A look at how Regulators, Companies and Individuals are shaping the Broadband map across America

Technology and the needs of businesses and consumers continue to evolve. Over the last 2 decades of mass internet penetration, availability and affordability have become the two key cornerstones of any efforts to shape internet policy, whether its at a regulatory level, like FCC or representatives of state broadband programs or at an internet service provider company level, like AT&T, Verizon.

Broadly speaking, according to FCC definition, Broadband is defined as reliable high-speed internet, having download speeds of at least 25 megabits per second (Mbps) and upload speeds of at least 3 Mbps. Broadband internet may be delivered via multiple technologies, including Fiber broadband, fixed wireless, digital subscriber line (DSL), or Cable broadband. Each technology has varying costs of setup and maintenance, with cable being generally regarded as the most cost-effective though technologically limited solution.

Perhaps more so than any other technological innovation in human history, the Internet has changed our daily lives in significant and permanent ways. Among many other things, the transformation has occurred faster than any other adaption of technological changes. Over the space of 2 decades, home broadband adoption has grown from 3% of all American adults age 18 and older to almost 80%. By comparison, it took telephone to  nearly 8 decades and electricity more than 30 years to reach the same level of penetration, despite deals .

Broadband is increasingly intertwined with the daily functions of modern life. It is transforming education, social services, healthcare, agriculture, supporting economic development initiatives, and is a critical piece of efforts to improve human life and socio-economic factors of human development.

Broadband has become the quintessential communication essential in the digital age and the era of Internet. Literally and figuratively, everything is available on the internet. And being connected, being connected always, has become an objective, an input and a goal statement in and of itself. Everyone, everywhere, has some purpose which requires them to access internet resources, unless of course, people making the active choice of living off the grid. However, with more than 19 million disconnected households across the country at the most conservative level of estimate, it is impossible to capitalize on broadband’s full economic and social impacts. While a presidential platform can incentive policy reform at the federal level, the road to change is still a long one, slowed by political infighting and congressional discord.

Internet Availability and Affordability for everyone

When people refer to broadband, most are referring to interrelated, sometimes overlapping characteristics of the sector. These two characteristics are like two sides of the same coin, and in every facet are tied to each other. The first is the digital telecommunications backbone or the infrastructure, whether wireless like mobile or wired technologies like Cable, DSL, etc, that enables outreach and availability of high-speed transmission of data. This digital backbone and physical infrastructure is both capital-intensive and technology-intensive, as in the front-end infrastructure, the towers, the wire lines, the cables, the physical buildings and exchanges and offices where the telecommunication equipment is assembled requires significant up-front monetary investment as well as time, whereas set-up of back-end technology and processes to enable the operations, billing and provisioning of broadband services requires years of preparation.  The other side of this coin, the federalist and state policy frameworks that govern physical infrastructure and related coverage especially in areas with limited revenue and growth potential, dictate how the implementations are carried out.  Though broadband’s antiquated definition tends to focus strictly on the transfer speeds and the network capabilities of the underlying technology, iin reality, broadband infrastructure can only reach its potential if every individual can use the service, and if policy frameworks are in place to support ubiquitous, near barrier-less adoption.

By pricing the broadband service, especially the entry levels prohibitively high so that its out of reach of certain section of the society, or by excluding the geographical areas or communities altogether, or by not investing or reducing the investment in upgrade and maintenance of the lines, it is certainly true that in certain parts of the country the entry barriers to obtain sufficiently fast broadband connection are too high. And lastly, not having adequate competition at a market level not only forces the customers to settle for low standards of service, but stifles innovation and investment, two most critical factors for market growth.

There are three primary areas within the colloquial term broadband availability that must be broken down for an effective analysis. These are availability of service, affordability of services and the presence or lack of competition, as in, the competition exerted by each internet service provider competing for market share. Lets look at these briefly.

1. Availability of Service -: this is apparently self-explanatory and generally the first step of internet connectivity. This is plan and simple availability of broadband signal in an area. According to the most recent data available, the Federal Communications Commission (FCC) estimates that 21 million Americans still lack broadband access according to the most recent data available. Many researchers have however cast serious aspersions on the veracity of this data, at times placing the number of people without access to reasonable internet as high as 162 million.
2. Availability of speed tiers and plans and Affordability -: FCC voted in January 2020 to approve a $20.4 billion Rural Digital Opportunity Fund that will make grants to broadband providers to build broadband networks where there is currently no or substandard service, however the money will be disbursed without any guarantee that the grant recipients will help to make broadband more affordable for low-income customers. It is strongly believed that to avoid corruption and to prevent this from becoming yet-another-failed scheme, Congress and the FCC should require, as a condition for funding this broadband deployment, true high-speed service at actually affordable prices—for instance, service rated at 50 Mbps for upload and download with unlimited usage for $10 per month to eligible recipients. What is alarming is that home broadband adoption rates aren’t improving. Approximately 73% of U.S adults use broadband service at home, but that figure has remained stable for the past three years. Too many people aren’t subscribing to broadband even when it is available to them. Affordable service is crucial.
3. Lack of more than one service provider -: According to a Fortune Article, laws in 19 states restrict local communities from experimenting to increase broadband deployment. These laws should be revisited and repealed. An indubitable truth – when new competition is introduced in broadband markets, the benefits are demonstrable. Look no further than Kansas City, Kan.; Chattanooga, Tenn.; Wilson, N.C.; and Longmont, Colo. for evidence that competition from private or municipal broadband provider results in incumbent providers dropping prices and increasing speeds — benefits which can not be realized with any amount of coaxing or appeals as seen in nearby areas where old standards still persisted in a market without competition. An estimated 35% of people have only 1 choice of broadband provider; another 37% have just two to choose from. Without competition, people—especially those with lower incomes or those who live in rural areas—are threatened with artificially high prices, lower-quality service, and little innovation. A Wall Street Journal report published last year found that low-income areas and high-income areas pay similar median monthly costs (about $66) for stand-alone internet, but poorer areas get 40% slower speeds for the price.

Image: Figure 4 FCC report

Figure 4 shows deployment of fixed terrestrial services at various speed tiers from year end 2014 through 2018.132 As of December 2018, fixed terrestrial service of 50/5 Mbps service is deployed to 92.7% of the population, up from 91.6% in 2017. Between 2017 and 2018, the deployment
of 100/10 Mbps increased from 88.6% to 90.5% of the population, and the deployment of 250/25 Mbps dramatically increased from 58.3% to 85.6% of the population. While deployment in rural areas and on Tribal lands lags behind deployment in urban areas at all five speed tiers, but the data show year-over year improvements for all speeds in these areas. For example, the deployment of 250/25 Mbps increased from 28.2% to 51.6% of the rural population

While fiber is the fastest home internet option by far, availability is still scattered. Due to the high cost of installing fiber service directly to homes, even major cities are still predominantly served by cable. Chicago, for example, only has 21% fiber availability as of 2020. Dallas has about 61% — and that’s actually high availability compared to other major metros in the US.

Image: CBS News

Challenges and Impact of inequitable distribution of broadband connectivity

Communities without reliable high-speed internet service cite a growing gap between the availability of resources and opportunities to their residents compared to those in communities that have a robust network. Given the ubiquitous nature of internet access, its vastly important to recognize how denial of broadband internet access, whether intentional or unintentional has become a severely debilitating factor for people and communities. Recognizing the importance of broadband and responding to such frustrations, states, communities and even individual people are seeking to close this gap. Most states have established programs to expand broadband access to communities that lack broadband internet connectivity or are undeserved. State efforts to expand broadband access are primarily focused on extending wired and fixed wireless infrastructure to the last mile: homes and small businesses. While Internet service providers, generally private companies licensed to distribute wireless and wired connectivity services have delivered reliable high-speed internet to households in most urban and suburban areas, many rural areas and areas with less population density remain under-served or lack services altogether. The issue of under-serving is particularly complex because people and regulators have used different definitions and standards from time to time. The challenge of closing the last-mile gap is compounded by geography, demographics, and the numbers and types of entities that provide service. In some states and regions, these patterns have led to uneven deployment of broadband infrastructure. While one rural community may have “fiber to the home and to the farm and to the cabin” provided by a local telephone company or cooperative, a neighboring community may lack the same level of broadband access.

While the Federal Government owns or manages key assets that support telecommunications infrastructure, the bulk of America’s telecommunications infrastructure is owned and managed by private-sector companies. This private market is a significant asset to our Nation’s economy and has helped the United States innovate and lead the world in each wave of telecommunications technology. Over the past several decades, Federal partnerships have been especially important for deployment in high-cost rural areas, where the unique challenges of geography, population density, and deployment costs may make it unprofitable to expand or operate networks – creating significant gaps in rural broadband coverage.

There are a unique set of challenges associated with delivering high-speed broadband to rural locations that service providers do not encounter in more urban locations, including geographical variables and high costs. Fortunately, recent fixed wireless solutions are equipped to address these variables as they serve as a cost-effective alternative to drop, distribution and/or feeder fiber, providing a whole new set of deployment models to the traditional Fiber-to-the-x (FTTx) deployment models. Unlike “urban jungles,” rural areas have a varying degree of terrain. Depending on the geography of the region, providers can encounter anything from rock and sand to compacted dirt and mud — making planning and executing a fiber buildout difficult. Many times, technicians are unaware of what type of soil composition they will be digging into until the project has begun. And then, they may find that getting the adequate trenches dug to lay the fiber is near impossible

According to a Fortune Article, Wealthier communities are two to three times more likely to have more than two choices for broadband providers than are communities with lower-than-average household incomes. With limited competition, it is perhaps unsurprising that Americans pay the second-highest broadband prices among OECD countries. Yet when new competition is introduced in broadband markets, the benefits are demonstrable. Look no further than Kansas City, Kan.; Chattanooga, Tenn.; Wilson, N.C.; and Longmont, Colo. for evidence that competition from a private or municipal broadband provider results in incumbent providers dropping prices and increasing speeds—but not in nearby areas the new competition didn’t serve.

Internet Availability and Affordability exacerbated in the Covid-19 era

Families across America, especially in semi-urban and rural areas often have to contend with throttled internet speeds. During Covid-19 as many families were forced to stay at home and do most of their activities at home, the demand for internet bandwidth has cast strains on the internet service providers’ ability as well as the rising data bills for many families.

Most families are hit by both the availability and affordability aspect. Contending with throttled internet speeds, many families often drives miles to find a spot where they can stream videos for work and class, including parking in acquaintances’ driveways to connect to Wi-Fi or a public utility like libraries or near a cellular tower across town.

The public health crisis has exposed New York’s digital divide. Lawmakers representing communities in upstate New York have voiced concerns about the issue for years, fighting to increase access to high-speed internet in rural communities that often struggle to even get a bar of cell service. But despite repeated pledges by state officials to remedy the situation, access to high-speed broadband internet remains elusive in the state’s bucolic areas.

“I would say that this current pandemic has really brought to light the challenges facing rural America when it comes to the lack of broadband,” said U.S. Rep. Anthony Brindisi, a Democrat whose district includes Utica and Binghamton. “These are challenges that many of us have been screaming about for many years, but [now] it seems to be very visible to the public at large.”

According to state Sen. Jen Metzger, a Democrat who represents a largely rural district in the Hudson Valley and Catskills., “There are many households that simply can’t afford it, and so we’re essentially reinforcing cycles of poverty and making it difficult for young people to realize their full potential in school and beyond,”

According to a Politico Article, Gov. Andrew Cuomo committed to providing broadband access to every New Yorker by the end of 2018, but missed the deadline. The percentage of New York residents and businesses served by “wired or wireless broadband” has gone from 70 percent in 2015, when Cuomo announced his program, to 98 percent, according to Department of Public Service spokesperson Jim Denn. According to David Little, executive director of the Rural Schools Association. “There’s vast stretches of land outside of urban areas that don’t have it, and so you have thousands of students sitting outside school buses being used as Wi-Fi hotspots so students could have some access closer to home.”

Teachers throughout the state have come up with creative solutions to help their students, including shipping out paper packets with assignments that students then mail back and calling students on landlines to assist with homework.

But for the most part, the coronavirus pandemic has frustrated families who have been calling for improved access for years.

Joanne Mazzotte, a counselor for the Crown Point Central School District in Essex County, said she hopes broadband is recognized as a basic necessity after the pandemic because of the struggles her family and others have faced. Cellular service in the area is so bad that to upload a file to Google Drive, “We’d have to huddle around one window and it would still take half an hour” despite living on a main road, she said.

How to get the Internet Service Providers to act

According to Industry estimates, based on variables like pole mounted or buried cables, laying fiber infrastructure can cost between $18,000 and $22,000 per mile. It’s all about return on investment (how quickly can the company get its cash investment back in order to reinvest in additional projects), and ability to grow the company.

Let’s look at a simple example with the following assumptions:

• $20,000 per mile for fiber infrastructure cost
• $600 per home to deliver service from the street
• 13 homes per mile (typical service area may have between 10 and 15)
• Average monthly cost for the provider to actually deliver the service after installation is $32 per month
• The subscriber pays on an average $65 per month for the service
• Net revenue per home is $33 per home

For a one mile build with 13 homes, the total project cost would be $20,000 PLUS the $600 for each home that connected to the service, about $2,140 per home – assuming that every home took service.  If only 7 of those homes sign up for service, the cost per home served jumps to $3,460.   By dividing the cost per home by the net revenue per home of $33, its simple to see that it will take nearly nine years for the provider to break even on the investment.

In case of multiple occupancy buildings, on average, fiber optic cable installation costs $1 to $6 per foot depending on the fiber count. It’s very difficult to estimate an exact price for an entire building to be wired, however an example would be $15,000 to $30,000 for a building with 100 to 200 drops. Fiber optic cabling is somewhat more expensive up front than copper cabling, but the greater capacity and reliability of fiber can actually reduce long-term costs.

It has been shown that States and local government bodies can use multiple policy levers to drive Internet Service Providers to expand broadband access, and some of these actions do not have to be dependent on available funding. States often support private enterprise broadband deployment through various means. In addition, where the Internet Service Providers fail to react, the state governments reserve the right to bring in non-profit cooperatives and special focus groups to build the necessary infrastructure.

1. Grants and loans to service providers: States provide support, primarily through grants, to facilitate broadband deployment in un-served and under-served areas. Several states, including Minnesota and Tennessee, have funded their broadband expansion programs through appropriations, while others, including California and Colorado, have re-purposed funds that have traditionally supported universal telephone service.
2. Identifying, setting up and promoting nonprofit utility cooperatives, and local governments to help make projects economically viable. In Minnesota, The Governor’s Task Force on Broadband, formed in 2011, provides a forum for stakeholders to study and discuss issues related to broadband. The Office of Broadband Development, Minnesota aims to connect all homes and businesses at speeds of at least 25 Mbps/3 Mbps by 2022 and 100 Mbps/20 Mbps by 2026 and become a national and international leader. West Virginia has passed several bills to encourage broadband expansion. In 2017, the Legislature passed a measure that established a goal of providing broadband access statewide, allowed the formation of cooperatives to provide the service, created a loan guarantee program to reduce risk for commercial lenders, and allowed Internet Service Providers to use micro-trenching technology to install infrastructure. In Maine, Maine’s ConnectME Authority, established in 2006 through the Advanced Technology Infrastructure Act, provides planning and infrastructure grants to facilitate broadband deployment to un-served and under-served communities. Between 2016 and 2018, ConnectME funded 14 planning grants totaling $451,000. Planning grants also help prepare applicants for ConnectME infrastructure grants. Between 2006 and 2018, ConnectME funded 144 infrastructure projects, providing more than $12 million to bring broadband infrastructure to nearly 40,000 households. In 2015, North Carolina’s Legislature tasked its Broadband Infrastructure Office (BIO), within the North Carolina Department of Information Technology, with developing a state broadband plan.

The universal service Schools and Libraries Program, commonly known as “E-rate,” provides discounts of up to 90 percent to help eligible schools and libraries in the United States obtain affordable telecommunications and internet access. The program is intended to ensure that schools and libraries have access to affordable telecommunications and information services.

There are currently 331 municipal networks in operation today in the U.S. We reviewed every state that has roadblocks preventing the establishment of municipal networks and compared them to states that do not have such restrictions in place. What we found was that states without restrictions enjoyed higher access to low-priced broadband plans on average.

According to a 2019 report on the health of municipal broadband, 22 states now have substantive roadblocks to establishing municipal networks to residents, down from last year’s 25. Three more states, Arkansas, California and Connecticut now permit such municipal broadband networks in full. Residents in states with no roadblocks or restrictions in place against municipal broadband have, on average, 10% greater access to low-price broadband (which we classify as any standalone internet plan $60 per month or less).

Many industry insiders feel that with the Covid-19 crisis, the focus is shifting to correct the systemic imbalances that have existed and that were confabulated or bloated due to decades of mismanagement and allowing private internet service providers to have a free rein. There’s absolutely no reason that people living in rural areas or Indian country and folks living in under-served areas must leave the safety and comfort of their homes and sit in a car outside of school or library in order to do the things that people in more than 90% of United States take for granted, that is, being able to access reasonable broadband internet from home. Further, now with Covid-19, telehealth is being used for primary care visits. The patient can stay at home and connect to their primary care provider via the internet. That way people don’t have to risk the face-to-face interaction. And so if there’s a silver lining to all of this, maybe this crisis, this pandemic, is bringing these issues to the forefront. And saying, look, there is no reason that people living in certain parts of country need to deal with essentially what is third-world connectivity.

In terms of funding to support tele-health implementation, now with Covid-19, the FCC has $200 million available to help hospitals and clinics to provide services to patients in their homes.  Under the program, healthcare entities would have internet service providers bid on service improvements, such as laying fiber to a hospital or clinic, and then the funds would cover up to 65 percent of the costs of the service improvements.

One way to make sure internet gets to everyone is to make the internet a utility. It has to be free and open and available to everyone, everywhere, every time. All of the rural under-served and not served areas and Indian country needs broadband. All of these areas and communities need  additional spectrum to do what they need to do. Everything ranging from tele-health to the new innovations that are taking place is denied to people where there is lack of adequate broadband capabilities.

Much has been written about the digital divide and its impact on those with limited access to broadband Internet service. Broadband Internet service has become a cornerstone to the world economy, as many things including advertising, sales, news, education, job applications, and basic communication move predominantly online. Those with broadband Internet tend to have an advantage over those without, and the people least likely to have broadband access live in rural areas. Unless broadband access is addressed in rural areas, today’s disadvantages resulting from limited broadband access will continue to grow in prominence as bandwidth needs expand and the broadband definition changes in the future. There’s always existed this lack of parity in telecommunications between rural and urban areas and from the beginning of federal communications laws in 1934, the FCC was created precisely to address a lack of access in more rural and remote areas. And so its time for everyone, the FCC, private telecom companies and local and state governments to embrace this principle called universal service, the idea that all Americans would have access to communications services.

This is part II of our blog post on Exposure Notification. Our original blog on Exposure Notification was published August 30^th.

Let’s look at the Background

Public Health Agencies around the world and especially in US have had little success with using mobile phones based technology to monitor the spread of Coronavirus and to warn users proactively.

Some public health agencies in the United States and around the world wanted to build mobile apps that would help them track the spread of the virus, through a process known as “contact tracing.” Due to the slow moving government apparatus, legal and procedural requirements and the logistical challenges, few of such program could take off. The contact-tracing apps that were initially launched did not function properly because of certain limitations, primarily the concerns around privacy and collection of data.

Apple and Google announced a surprise partnership at the start of this pandemic in April. When Apple and Google announced their work together on the COVID-19 Exposure Notification API, the companies put behind years of rivalry to join hands to help people and Public Health Authorities fight this massive battle. Apple and Google announced two phases of the Contact Tracing project. During the first phase, which is what came with iOS 13.5, laid out that that users first download an app from their public health authority and then opt-in to Exposure Notifications. The process of Exposure Notification System works through sharing anonymous Bluetooth beacons with nearby devices running the same software, tagging those that suggest extended and close contact associated with coronavirus spread, and saving the last 14 days of these records.

At the same time, Apple and Google also indicated start of work on the second phase, which would reduce the reliance on contact tracing app from public health authorities while bringing the core functions of the COVID-19 Exposure Notification technology directly into iOS and Android. This is essentially what Apple and Google announced on Tuesday, 1^st September.

So what is new in the second phase of Exposure Notifications

Apple and Google announced on 01^st September that their joint program, contact tracing Exposure Notifications System, can inform people of potential exposure to COVID-19 without a dedicated Exposure Notifications app. The second phase of this program, the companies announced, re-launches the warning software in a new and better Avatar, so that state public health agencies can participate without having to create customized apps. This app-less functionality is called Exposure Notifications Express and is only available when a Public Health Authority (PHA) supports it.

The two partners, Apple and Google are introducing new tools that benefit both the public and public health authorities, making it much easier for public health authorities to implement digital exposure notification, while reducing a step in the process for the general users. For public health authorities, now they do not need to worry about the need for developing and maintaining their own individual contact tracing application. Apple made this breakthrough via the iOS 13.7 system update, released 01^st September to general public, while Google is implementing it with an automatically generated application on Android 6.0, upcoming later in September, taking a little longer because of the very different method through which it manages system services and OS updates.

For public health authorities, the new changes bring significant ease of operation as the process of adopting Exposure Notifications Express by users is significantly streamlined compared to adopting the existing Exposure Notification API where users first had to download the contact tracing application. Public health authorities simply provide a configuration file that includes their name, logo, criteria for triggering an exposure notification, and information and protocol that is displayed to users following an exposure. The existing way of Exposure Notification API demanded that users download the contact tracing application from their public health authority first before proceeding with other steps. This was a major headache for public health organizations who were required to maintain their own infrastructure and software application.  Further it was an additional and often confusing step for users. Now that part is completely eliminated benefiting both public health authorities and users.

Further Apple and Google say that they will use the information provided by the public health authorities to offer a fully operational Exposure Notification Systems on behalf of the public health authority directly integrated into their respective operating systems, in case of Apple this is iOS 13.7.

Crucially, Public health authorities still have full control over the system, though, and there are no additional privacy or data related concerns. The Public Health Authorities still dictate and control the process of triggering notifications, what is the language and structure of the advice, and guidance on the next course of action for exposed individuals.

iOS 13.7 changes

With the new Exposure Notification Express, which forms the second phase of Exposure Notification System, the system removes one of the key barriers to adoption that led to a slow start to the software. Once users update iOS 13.7, users can now enable COVID-19 Exposure Notifications directly in the Settings app on their iPhone. This new process is called Exposure Notifications Express. The process is as simple as enabling user preferences in any other application and just takes a few taps, including agreeing to the public health agency terms and conditions of service. Additionally, users will also be able to opt-in to receive a push notification when their local public health authority adopts Exposure Notifications Express.

Once an user enables exposure notifications in the Settings, their iPhone will begin monitoring with Bluetooth to log possible exposures so the user can be notified of a potential COVID-19 exposure based on the guidelines set by the local public health authority. The process of monitoring and triggering Exposure Notification System works without any changes, through sharing anonymous Bluetooth beacons with nearby devices running the same software, tagging those that suggest extended and close contact associated with coronavirus spread, and saving the last 14 days of these records.

At least in the United States, many people haven’t had the option of participating, as states have been slow to create apps. Now, with “exposure notifications express,” states will have less work to participate. Through this simplification pf approach, Apple and Google hope that adaption rate of the Exposure Notifications Express system will dramatically increase, at both levels, from users and public health authorities. The first public health authorities in the United States to adopt the Exposure Notification Express system will be Maryland, Nevada, Virginia, and Washington, D.C.

Users who live in states that participate in the software may get a pop-up notification, prompting them to opt into the program. By following simple steps, they can share their Bluetooth data and receive notifications if they come in contact with another participant who has tested positive. For states that already have a contact tracing application standalone application using the COVID-19 Exposure Notification API, those apps can still exist and operate on their own. As of right now, Apple and Google say that 25 states and territories, representing more than 55% of the population, are exploring Exposure Notifications System solutions.

Finally, Apple and Google emphasize that all of the original privacy protections of the Exposure Notification API also extend to the Exposure Notifications Express. Users must explicitly enable exposure notifications, nothing is enabled by default. No location data is shared and the system does not share your identity with other users, Apple, or Google. All matching is done on-device and users have full control over whether they want to report a positive test.

“I would say this is an improvement,” said Jeffrey Kahn, director of the Johns Hopkins Berman Institute of Bioethics. Kahn, who has been studying the use of technology to fight the virus, said states have been hamstrung by indecision around which technology vendors they should use to build their apps, among other issues. He said this may help speed up adoption, but shouldn’t be considered a magic bullet.

“Public health agencies are carrying an extraordinary load in managing the novel coronavirus response,” said Scott J. Becker, head of the Association of Public Health Laboratories, in a statement provided by the companies. “The easier we make it for state and territorial public health agencies to develop and deploy, the sooner we can expand COVID-19 exposure notification in our communities and help end the pandemic.”

Image Credit: University of South Florida

Exposure notification and Contact Tracing: Essential tools

As the Novel Coronavirus or COVID-19 pandemic has gripped the world, Governments, Organizations and People all over the World are trying desperately to break the vice like grip of this deadly pandemic. With more than 830,000 people dead and 25 million directly infected, and the disease showing little signs of slowing down, the nature of the crisis is unprecedented and unmitigated.

The software development communities across the World have contributed since the start of the pandemic in 2019 by developing apps, data collection programs and frameworks to assist the medical communities and local, state and national or federal governments in their attempt to fight against this pandemic. The indefatigable spirit of doctors and nurses across the world required a solid backing and the tech community came forward unselfishly.

According to Wikipedia, the Exposure Notifications System, originally known as the Privacy-Preserving Contact Tracing Project, is a framework and specification developed by Apple Inc. and Google to facilitate digital contact tracing during the COVID-19 pandemic. Contact tracing is a technique used by public health authorities to contact and give guidance to anyone who may have been exposed to a person who has contracted COVID-19. The project’s aim was to produce a framework to do a digital trace of COVID-19 cases so if one happens to be near someone or in contact with someone who is later diagnosed with COVID-19, one can get a notification and take the appropriate steps to self isolate and get medical help if necessary. Think of it as an Early Warning System for those who came in contact with known or unknown COVID-19 positive people. The reason behind this close collaboration is evident – just like the Novel Corona-virus doesn’t differentiate between people, any efforts to digitally trace the spread of infection must transcend the very human boundaries of technology, software and hardware.

The announcement of this unprecedented collaboration project, a first of its kind among two arch rivals, immediately stoked fears of unreserved data collection and many people started voicing their concerns and feedback started flowing in. The people at large had genuine cause to worry. Tech companies after all have a less than acceptable track record of masking data collection activities while aggressively repudiating any efforts to control their massive arsenal of data-collecting apps.

Exposure Notification makes it possible to combat the spread of the coronavirus — the pathogen that causes COVID-19 — by alerting participants about possible exposure to someone they have recently been in contact with, who has subsequently been positively diagnosed as having the virus. There are three broad parts to this COVID-19 Exposure Notification system. The first part is defining the users. The second part is the feature of communication standard used to communicate between devices. And the third and final part is the contact tracing app developed by the local or state health agencies.

They process of Exposure Notification System woks through sharing anonymous Bluetooth beacons with nearby devices running the same software, tagging those that suggest extended and close contact associated with coronavirus spread, and saving the last 14 days of these records.

The first contact tracing app, Virginia’s COVIDWISE, debuted Aug. 5. North Dakota and Wyoming shipped their Care19 Alert Aug. 13, Alabama launched its GuideSafe app Aug. 17, and Nevada introduced COVID Trace on Monday. The University of Arizona is testing Covid Watch Arizona, with a statewide release expected soon.

So now lets look at the first part. According to Apple, there are two primary user roles identified within the Exposure Notification framework.

So who are the Users?

Affected user -: When a user has a confirmed or probable diagnosis of COVID-19 (as defined by the Health Authority), the framework identifies them as affected and shares their diagnosis keys to alert other users to potential exposure.

Potentially exposed user -: To assign a user the potentially exposed role, use the framework to determine whether a set of temporary exposure keys indicate proximity to an affected user. If so, the app can retrieve additional information such as date and duration from the framework.

How does the Exposure Notifications System work?

The second part of the project concerns with the actual working of the feature and communication.  The Exposure Notification Service is the vehicle for implementing exposure notification and uses the Bluetooth Low Energy wireless technology for proximity detection of nearby smartphones, and for the data exchange mechanism.

The Exposure Notification system is made up of few components, namely Bluetooth keys, an API to communicate, and an app typically distributed by the local public health authorities.

First up, an API or Application Programming Interface is a software intermediary or bridge that allows two applications to talk to each other. Each time some uses an app like Facebook, or sends an instant message, or check the weather on phone, the application is using an API to fetch and display the data.

Apple and Google developed the underlying APIs and Bluetooth functionality, but they are not developing the apps that use those APIs. Instead, the technology is being incorporated into apps designed by public health authorities worldwide, which can use the tracking information to send notifications on exposure and follow up with recommended next steps.

• Contact tracing app download by users – Users download the contact tracing app provided by local public health authorities that are compatible with ENS (Exposure Notification System)

• User opts in to the ENS (Exposure Notification System)

Image Credit: Macrumors

• ENS (Exposure notification system) generates a random ID for user’s device. To help ensure these random IDs can’t be used to identify the user or user’s location, they change every 10-20 minutes.

  

Image Credit: Nevada Health Response

• The user’s phone and the phones around user work in the background to exchange these privacy-preserving random IDs via Bluetooth. The users do not need to have any app open or opt-in or opt-out for this process to take place.

Image Credit: iunera

• User’s phone periodically checks all the random IDs associated with positive COVID-19 cases against its own list.

• If there’s a match, the app will notify the user with further instructions from their public health authority on how to keep user and the people around them safe.

Image Credit: CGTN

Finally, what about Privacy

The Exposure Notifications System was designed with users’ privacy and security at the center and essentially everything else floating around the need to keep contact’s identity secure. An user’s identity is not shared with other users, Google, or Apple. Even a cursory read of the document produced by Apple and Google detailing the specifications and working details of the program reveal that the concerns around privacy may afford to be relaxed at least in the case of this feature. Here’s a look at the top privacy concerns.

• The APIs have been created with privacy and security in mind, and app usage is opt in rather than mandatory.
• User decides whether they want to use Exposure Notifications—it’s off unless user turns it on.
• ENS doesn’t use location data from your device.
• User’s identity is not shared with Google, Apple or other users.
• Only public health authorities can use this system.

Phase 1 of Exposure Notification System

The first phase, released by Apple through iOS version 13.5 on 20 May, 2020 required that users first download an app from their public health authority to opt-in to Exposure Notifications.

iOS 13.7 lets you opt-in to the COVID-19 Exposure Notifications system without the need to download an app. System availability depends on support from your local public health authority. For more information see covid19.apple.com/contacttracing. This release also includes other bug fixes for your iPhone.

This method makes contact tracing significantly easier for public health authorities who won’t have to waste critical time or spend valuable money on developing an app.

So What’s coming

The Exposure Notification System or ENS allows public health authorities to develop apps that augment manual contact tracing efforts while preserving the privacy of their citizens. As of today, public health authorities have used ENS to launch in 16 countries and regions across Africa, Asia, Europe, North America and South America, with more apps currently under development.

Apple likely made the change to use Exposure Notification service without the need for a contact tracing app to quietly encourage more public health departments to use its service. Only a handful of states in the United States and a few countries worldwide are using Google and Apple’s mobile technology. Most health departments still use an old fashioned contact tracing method that relies on in-person interviews and phone calls to locate those individuals who came in contact with an infected person.

Recently, Nevada Department of Health and Human Services started offering its own contact tracing app to use the ENS (Exposure Notification Service) to keep people safe.

Whenever there is a talk of infusing technology, automation and intelligence in to tasks or processes, the solutions proposed fall in to two broad buckets. The first is the conventional use of the term Artificial intelligence, machine learning, deep learning and cognitive learning, which typically work outside human intervention and are largely driven by some kind of an algorithm to achieve its goal. The handling of regular transactions as well as any exceptions are defined within the AI program and the machine doing the job in case of touch and feel jobs like manufacturing or the algorithm producing the result in case of purely online work. The second category is where Artificial Intelligence is applied to work in conjunction with the humans and human intervention and interaction is a critical part of the successful completion of the job. This second type is known as Augmented intelligence or Augmented reality.

Another way of looking at what happens with augmented and artificial intelligence is the degree to which algorithms and the programs running the physical machines are expected to make decisions on their own. Is the decision making assisted by humans, controlled by humans or completely out of control of humans?

It’s critical to note that while the technologies and fundamentals powering both Artificial intelligence and Augmented intelligence are largely the same, the applications, goals and objectives are objectively different. Simply put, AI creates a human less environment while Augmented Intelligence or Intelligence Augmented as it’s often called, seeks to create an environment for betterment of humans and human endeavors.

There’s virtually no major industry where modern AI — more specifically, “narrow AI,” which performs objective functions using data-trained models and often falls into the categories of deep learning or machine learning — hasn’t already affected. That’s especially true in the past few years, as data collection and analysis has ramped up considerably thanks to robust IoT connectivity, the proliferation of connected devices and ever-speedier computer processing.

• In Manufacturing: AI powered robots work alongside humans to perform a limited range of tasks like assembly and stacking, and predictive analysis sensors keep equipment running smoothly.
• Healthcare: In the comparatively AI-nascent field of healthcare, diseases are more quickly and accurately diagnosed, drug discovery is sped up and streamlined, virtual nursing assistants monitor patients and big data analysis helps to create a more personalized patient experience.
• Education: Textbooks are digitized with the help of AI, early-stage virtual tutors assist human instructors and facial analysis gauges the emotions of students to help determine who’s struggling or bored and better tailor the experience to their individual needs.
• Media: Journalism is harnessing AI, too, and will continue to benefit from it. Bloomberg uses Cyborg technology to help make quick sense of complex financial reports. The Associated Press employs the natural language abilities of Automated Insights to produce 3,700 earning reports stories per year — nearly four times more than in the recent past.
• Customer Service: Last but hardly least, Google is working on an AI assistant that can place human-like calls to make appointments at, say, your neighborhood hair salon. In addition to words, the system understands context and nuance.

Source: McKinsey & Company

According to a recent publication, of the 9,100 patents received by IBM inventors in 2018, 1,600 (or nearly 18 percent) were AI-related. Here’s another: Tesla founder and tech titan Elon Musk recently donated $10 million to fund ongoing research at the non-profit research company OpenAI — a mere drop in the proverbial bucket if his $1 billion co-pledge in 2015 is any indication. And in 2017, Russian president Vladimir Putin told school children that “Whoever becomes the leader in this sphere [AI] will become the ruler of the world.” He then tossed his head back and laughed maniacally.

The single biggest strength of Artificial intelligence is the simple fact that it can do repetitive tasks better than anything else. And the more quantitative, the more objective the job is—separating things into bins, washing dishes, picking fruits and answering customer service calls—those are very much scripted tasks that are repetitive and routine in nature. In the matter of five, 10 or 15 years, they will be displaced by AI.

Yet even the most pragmatic AI scientists stress that today’s AI is useless in two significant ways: it has no creativity and no capacity for compassion or love. Rather, it’s “a tool to amplify human creativity.” In other words, sure you can teach AI to make strokes with a brush in a canvas, but an AI driven machine can never in a million years paint a Monalisa.

The most advanced robot costing many millions in research dollars and years of work, cannot pickup a tea cup like a 2 year old can. Or millions of sensors on a robot cannot feel taste and smell like humans do. Sure a robot Make millions of calculations in a second, which is many hundreds of times more than the most intelligent human can, sure a robot can discern each smell, or can break down each material in to its atoms and sub-atoms, but it cannot experience the same thoughts, emotions and feelings that human brain can.

The second biggest fear around AI stems from the AI and machine learning applications inheriting some of the human biases based on pre-disposed notions and behaviors that humans may code in to or indirectly influence the program.

Augmented Intelligence on the other hand doesn’t replace humans with technology rather uses Artificial intelligence to aid humans in doing a job.

In recent years, in AI technology rankings in terms of the value they create for businesses, Augmented Intelligence was ranked in second place, just below virtual agents. However, Gartner predicts that “Decision support and AI augmentation will surpass all other types of AI initiatives” creeping into first place this year and then exploding as we reach 2025 becoming around twice as valuable as virtual agents.

Just like regular project management work, any organization looking to employ AI needs to first define its use cases and requirements clearly and describe the Big Y, or the big problem it’s trying to solve. Then, they need to define the various aspects of business goals, or the product that they want to build. The next step is to outline data requirements and functionality to solve those business goals. And finally, what’s the ROI or returns the organizations expects. What makes AI and machine or cognitive learning projects unique is that they also need to consider the human-machine dynamic. For example, which part of the chain or which components of the product or solution do they want to hand over entirely to machines to execute, and which parts do they want to retain for their Human Resources? Where machine or a computer algorithm based program is making the decision, is it completely autonomous or is there a human there to monitor? Is The machine or computer program Only responsible to feed data, information or half finished product to human to make the final decisions?  In the scope of these decisions, then, it makes a lot more sense to create an AI role matrix which can evolve over time. This matrix Lays down specifically the kind of role AI or cognitive learning system will play for that particular function or process.

Source: USM Business Systems

Getting to true autonomous intelligence or fully independent, cognitive learning powered AI model is proving to be real difficult. Even one single instance of non-compliance or faulty decision making can throw the entire program off tracks.

During Tesla’s 2019 Autonomy Day, Tesla CEO Elon Musk said the company is expected to have one million vehicles on the road by the end of 2020 that could function as robotaxis. Though the semi-autonomous Autopilot and Full Self-Driving, or FSD, features are loved by some, others say Musk’s driverless dream is far from becoming a reality. Despite many hundreds of millions in investment and almost a decade of efforts, Tesla’s fully autonomous driving mode is not fully autonomous anymore rather in the best-case scenario is now only viewed as a partial augmented driving mode. In other words, human driver has to control the inputs. And while some of the Tesla’s self-driving features are loved by some, many still view it as unfit for our roads. While the hype around the autonomous driving mode was skyrocketing, and just as Tesla started charging a hefty premium for their self-driving feature, a handful of incidents and accidents changed all that. Scenarios of false-positives, or false-negatives, which are built in to the algorithms powering the programs behind the autonomous driving software, lead to further complications of their own, creating as many problems as they solve. In the best-case scenario, public confidence in self-driving technology is still many years if not decades away.

Source: ABC News

Watching the Youtube training sessions posted by Waymo, the Alphabet subsidiary in charge of self-driving cars, reveals the concerns with self-driving technology. One video shows a car that repeatedly and without reason stops in the middle of a street and then drives off again. The explanation came from a passer-by who was carrying a STOP sign sticking out of his bag, misleading the vehicle. In other words, the machine can drive itself, but it lacks the ability to differentiate ‘stop’ signal in different contexts and nuances of incidents on public roads.

The loan underwriters at top banks quickly realized that leaving decision making with a compute program is a recipe for disaster as the program has inbuilt biases and no amount of learning can make the program foolproof. Hence in most cases, AI programs are only limited to the first few stages in the process while the critics decision is made by a human.

Similarly, Apple credit cards have recently discriminated against women,giving them 50% less credit than men with the same income and profile.

Let’s use Netflix as an example. Say you recently watched “Orange is the New Black.” Netflix may then suggest other shows with prison themes, or documentaries about life behind bars, or shows with a strong female lead, etc.

Based on past data (your recently watched shows and movies), it’s able to make a prediction about what you will want to watch next. Once you make your latest selection, it will adjust its algorithm to further customize your experience.

According to a 2017 scholarly article on Augmented intelligence by Researchers Zheng and Wang, Within augmented intelligence, researchers Define models according to the varied degree of AI and human influence.

• The first classification is called Human-in-the-loop hybrid-augmented intelligence) Human-in-the-loop (HITL) hybrid-augmented intelligence is defined as an intelligent model that requires human interaction. In this type of intelligent system, human is always part of the system and consequently influences the outcome in such a way that human gives further judgment if a low confident result is given by a computer. HITL hybrid-augmented intelligence also readily allows for addressing problems and requirements that may not be easily trained or classified by machine learning.

• The second classification looks at (Cognitive computing based hybrid-augmented intelligence) In general, cognitive computing (CC) based hybrid-augmented intelligence refers to new software and/or hardware that mimics the function of the human brain and improves computer’s capabilities of perception, reasoning, and decision-making. In that sense, CC based hybrid-augmented intelligence is a new framework of computing with the goal of more accurate models of how the human brain/mind senses, reasons, and responds to stimulus, especially how to build causal models, intuitive reasoning models, and associative memories in an intelligent system.

Augmented intelligence follows a five-function cadence that allows it to learn with human influence. It repeats a cycle of understanding, interpretation, reasoning, learning, and assurance. Here’s how it works:

• Understanding: Systems are fed data, which it breaks down and derives meaning from.
• Interpretation: New data is inputted; the system then reflects on old data to interpret new data sets.
• Reasoning: The system creates “output” or “results” for new data set.
• Learn: Humans give feedback on output and the system adjusts accordingly.
• Assure: Security and compliance are ensured using blockchain or AI technology.

Source: Global Data Magazine

It is clear and evident that AI is here to stay. Humans have a deep reliance on Artificial Intelligence which has been around for decades. Yet, the future of Artificial Intelligence is brightest with the humans, morphing in to Augmented Intelligence. Having humans and machines work hand-in-hand is a win-win for both parties.

Most scientists and researchers agree that its probably extremely implausible, if not impossible to imagine collective human failure to the extent that AI is allowed to grow unchecked, while simultaneously all other uses of AI beneficial to humankind are ignored.

One thing is certain though: despite the ominous predictions and warnings on doomsday scenarios, arising out of the impact of AI on humanity and society, AI will never take over the world or morph in to Terminator style Machines Or Will Smith’s I robot kind of intelligence.

Image Credit: Adweek

All content platforms and social media companies must keep the content flowing because that is the business model: Content captures attention, provides viewership and generates data (users’ statistics). Content is the starting and the end point of consumers’ journeys on social media. A video, an information post, a tweet, blog post, picture, public service advisories, are all types of content. The platforms then sell that attention (read: viewership), enriched by that data (read: customized ads). But how do you deal with the objectionable, disgusting, pornographic, illegal, or otherwise verboten content uploaded alongside legitimate content?

How do Facebook and other tech and social media companies ensure integrity of content on their networks? And how do these companies work to curb misinformation on their platforms about the Coronavirus pandemic or the 2020 elections or any other global or regional event. We have seen state and non-state sponsored actors with nefarious intent take advantage of lax content posting norms.

Dangerous fake news has spread on platforms like Facebook in Myanmar, where the Rohingya ethnic minority are persecuted. United Nations has clearly blamed the role of social media in spreading the persecution and this is not the only example of its kind.

Misinformation campaigns (aka “fake news”) on Facebook have interfered with democratic elections around the world. After a man used Facebook to live stream his attack on two New Zealand mosques in March 2019, the video quickly spread. YouTube Moderators fought back hard taking down the video as newer versions kept popping up seemingly beating the controls that YouTube has in place to immediately flag already removed material. The uploaders were able to sneak past by using a loophole – exact re-uploads of the video are banned by YouTube, however videos that contain clips of the original footage must be sent to human moderators for review, thereby delaying the process. And again this loophole existed for a purely legitimate reason – to ensure that news videos that use a portion of the video for their segments aren’t removed in the process.

In 2017, a live stream on Facebook showed the fatal shooting of a 74 year old retiree in Cleveland, while also showing a man murdering his own child in Thailand. Both videos remained online for hours and racked up hundreds of thousands of views.

In a December 2017 report, ProPublica took a revealing look at content moderation. ProPublica gathered from its users 900 examples of where users believed that Facebook content moderation was incorrectly applied. ProPublica then selected 49 of such posts and asked Facebook to explain. Rather shockingly, yet unsurprisingly, Facebook admitted to an error by its moderators in 22 out of 49 posts. Just imagine, 22 out of 49 means approximately 45% or half of all posts in the sample had moderation applied incorrectly. No amount of explaining can explain that.

Social media is good, right!

Image Credit: Internet

Facebook serves as a platform for its billions of regular users to post, view and offer feedback about the content hosted on its servers. But when that content is more “terrorist propaganda” than “brunch photo,” or when it becomes “porn” than “essential context” to an image, the company has struggled to determine the right approach to removing it in time. The traditional methods of company moderators reviewing user-reported infractions is too time consuming, while the AI powered algorithms are too imprecise.

With the COVID-19 risk content moderators were sent home, and without proper technology, connectivity, and safety requirements met, Facebook’s automated system took full control. That was an unmitigated disaster, leading to widespread blocking or deleting of posts mentioning Coronavirus from reputable sources such as The Independent and the Dallas Morning News, not to mention millions of individual Facebook users. Those automated systems still have problems.

Content from legitimate sources, verified fact-checked sources, and sources with history of posting appropriate and trust worthy content is suddenly being targeted. While there were always instances of some posts getting tagged erroneously, there is an order of magnitude increase in such instances in the post-covid world. Clearly the strategy to have AI and ML based programs call the shots hasn’t worked.

“Facebook is blocking COVID-19 posts from fact based sources,” a Facebook source says. On March 17^th 2020, according to an Yahoo news article, Facebook suffered from a massive bug in its News Feed spam filter, causing URLs to legitimate websites including Medium, Buzzfeed, and USA Today to be blocked from being shared as posts or comments. The issue blocked shares of some but not all coronavirus-related content, while some unrelated links are allowed through and others are not. Facebook has been trying to fight back against misinformation related to the outbreak, but may have gotten overzealous or experienced a technical error.

According to a just released report by NYU Stern, Facebook content moderators review posts, pictures, and videos that have been flagged by AI or reported by users about 3 million times a day. So that is 3 million pieces of content just flagged for review out of possibly billions and billions of content posts. And since CEO Mark Zuckerberg admitted in a white paper that moderators “make the wrong call in more than one out of every 10 cases,” that means 300,000 times a day, mistakes happen.

So, is it all an experiment gone wrong? Did the novel coronavirus catch the social media content moderation framework at the worst time?

Image Credit: Webhelp

The one thing we know for sure is that you can’t control the beast that is Social Media. Generally, the response by firms to incidents and critiques of the social media platforms is primarily ‘We’re going to put more computational power on it,’ or ‘We’re going to put more human eyeballs on it.’” And that is generally fine. For it attempts to resolve the problem, or at least is seen as an attempt to resolve the problem, with or without adequate results. The focus is not on the results, rather on the proclivity to be seen as doing something.

Facebook uses more than 70 external partners and fact-checking firms. According to Facebook, it has over 30,000 people working on safety and security — about half of them are content reviewers working out of 20 offices around the world. Facebook employs almost all of these 15,000 content moderators indirectly, mostly outsourced workers. In similar context, YouTube today employs an expected 10 – 12,000 people to patrol al of Youtube and Google’s content.  Similarly, Twitter employees close to 2,000 people in its content review team.

Generally speaking, content management or content review falls in to two main buckets. The first is content moderation, where content moderators, mostly contractors working on behalf of lets say Facebook or Twitter, check the content for violations like nudity, sexual content, racism, hate speech, acts of violence or promoting violence, violating laws and community standards, child pornography, and like. Moderators are responsible for reviewing flagged content, and removing it in accordance with the policies of the social media platform. The second bucket is third party fact checking, where Facebook employs more than 70 third party organizations, primarily, news outlets and prominent individuals to check a particular content as True or False. Based on the result then, any one of the many actions can be taken. Either the content is either left up or demoted, or additional labels are added, or additional constraints are placed including monetary impacts or all of the foregoing in extreme cases.

According to content management and comprehensive community standards page on Facebook directly, the efforts to moderate and regulate content have three stages. First, is the Policy development process. The content policy team at Facebook is responsible for developing our Community Standards. We have people in 11 offices around the world, including subject matter experts on issues such as hate speech, child safety and terrorism. Many of us have worked on the issues of expression and safety long before coming to Facebook. Second is Enforcement of policies developed previously through its global content moderator workforce. Facebook uses a combination of artificial intelligence and reports from people to identify posts, pictures or other content that likely violates our Community Standards. These reports are reviewed by our Community Operations team, who work 24/7 in over 40 languages. Facebook’s fact-checking rules dictate that pages can have their reach and advertising limited on the platform if they repeatedly spread information deemed inaccurate by its fact-checking partners. The company operates on a “strike” basis, meaning a page can post inaccurate information and receive a one-strike warning before the platform takes action. Two strikes in 90 days places an account into “repeat offender” status, which can lead to a reduction in distribution of the account’s content and a temporary block on advertising on the platform. And finaly, Facebook launched a review process last year. A news organization or politician can appeal the decision to attach a label to one of its posts. Facebook employees who work with content partners then decide if an appeal is a high-priority issue or PR risk, in which case they log it in an internal task management system as a misinformation “escalation.” Marking something as an “escalation” means that senior leadership is notified so they can review the situation and quickly — often within 24 hours — make a decision about how to proceed.

If Facebook’s content moderators have three million posts to moderate each day, that’s 200 per person: 25 each and every hour in an eight-hour shift. That’s under 150 seconds to decide if a post meets or violates community standards.

Image Credit: TELUS International

According the NYU Stern report, and according to some recent investigations by Buzzfeed and news articles by NBC, Forbes and others, the problem of content reviews – whether its content moderation by moderators or third party fact-checking by independent news organizations and individuals is more structural and institutional in nature. The novel coronavirus just exposed a side of it and perhaps aggravated the outcomes.

According to a NBC news article, Facebook has allowed conservative news outlets and personalities to repeatedly spread false information without facing any of the company’s stated penalties, according to leaked materials reviewed by NBC News. According to internal discussions from the last six months, Facebook has relaxed its rules so that conservative pages, including those run by Breitbart, former Fox News personalities Diamond and Silk, the nonprofit media outlet PragerU and the pundit Charlie Kirk, were not penalized for violations of the company’s misinformation policies.

The list and descriptions of the escalations, leaked to NBC News, showed that Facebook employees in the misinformation escalations team, with direct oversight from company leadership, deleted strikes during the review process that were issued to some conservative partners for posting misinformation over the last six months. The discussions of the reviews showed that Facebook employees were worried that complaints about Facebook’s fact-checking could go public and fuel allegations that the social network was biased against conservatives.

“This supposed goal of this process is to prevent embarrassing false positives against respectable content partners, but the data shows that this is instead being used primarily to shield conservative fake news from the consequences,” said one former employee.

In a recent case at Facebook, related to appeals process, a Facebook employee filed a misinformation escalation for PragerU, after a series of fact-checking labels were applied to PragerU posts. A Facebook employee escalated the issue because of “partner sensitivity” and mentioned within that the repeat offender status was “especially worrisome due to PragerU having 500 active ads on our platform,” according to the discussion contained within the task management system and leaked to NBC News. After some back and forth between employees, the fact check label was left on the posts, but the strikes that could have jeopardized the advertising campaign were removed from PragerU’s pages.

In another case, a senior engineer at one of the top social media giants collected internal evidence that showed the company was giving preferential treatment to prominent conservative accounts to help them remove fact-checks from their content, according to Buzzfeed. The company responded by removing his post and restricting internal access to the information he cited. A week later the engineer was fired, according to internal posts seen by BuzzFeed News.

Many employees at top social media companies like Facebook, Twitter and others have expressed deep anguish on their internal inter-company platforms , amid growing internal concerns about the company’s competence in handling misinformation, and the precautions it is taking to ensure its platform isn’t used to disrupt or mislead ahead of the US presidential election.

Third party fact checking also suffers from severe debilitating factors severely limiting its outreach. Scale becomes an issue for the fact checkers as most organizations Facebook contracts work of fact checking to, typically only allocates handful of people to the task of fact-checking. Coupled with an impossible amount of fact-checking requests coming in, that means the people are constantly backlogged.

According to Sarah Roberts, a pioneering scholar of content moderation, and an information studies expert at the UCLA, the social media companies handle the activities of content moderation in a fashion that diminishes its importance and obscures how the activities of content moderation work. The idea is simple: make it obscure and muddy the waters, to achieve plausible deniability. Something straight out of the play book of top politicians and business executives – plausible deniability. Content moderation is a mission critical activity, yet most social media companies fulfill it with their most precarious employees mostly by just outsourcing the entire journey of content moderation.

Just as companies save significant amount of money by outsourcing transport logistics, janitorial and food services, outsourcing content moderation saves these social media giants tons of money. Just as we pointed out earlier, between Facebook, Twitter and Youtube there are close to 40 – 50,000 content moderators. And the number is only growing. Even at conservative estimate of 40,000 people, that is outsourcing work equivalent to 100% of work performed at 4 medium sized outsourcing services providers.

The lack of access and the lack of willingness by social media companies to allow any kind of scrutiny of their moderation practices has made content moderation a kind of black box ops where only few people know what takes place. This is certainly by design and it is no accident that the top social media companies choose the convenience of maintaining plausible deniability and the wait and watch approach while incendiary content burns and lights fire to everything around it.

What about human content moderators?

Image Credit: tampabay.com

According to Guy Rosen, VP of Integrity at Facebook, content moderation is a really arduous job. Numerous people have brought this issue to the fore. Watching countless hours of sadistic, violent, disturbing and purely horrific content day in and day out takes its toll. How do you get those hours of visions and thoughts out of your head when you head home? You cannot. Those sights and sounds stay with you. As a content moderator, its really hard to live a normal life after watching 8 hours of non-stop disturbing content.

In the recent past, a former Facebook moderator sued, accusing the platform of psychological harm. Former Microsoft employees sued Microsoft for similar reasons after the alleged trauma from reviewing child porn. In a more recent report, The Verge carried out a scathing review of the job conditions for content moderators at Facebook and the harrowing conditions surrounding the job in general. As one employee interviewed in the report put it: “We were doing something that was darkening our soul — or whatever you call it,” he says. “What else do you do at that point? The one thing that makes us laugh is actually damaging us. I had to watch myself when I was joking around in public. I would accidentally say [offensive] things all the time — and then be like, Oh shit, I’m at the grocery store. I cannot be talking like this.”

Accenture which performs content moderation for social media companies, has its employees sign a form that directly acknowledges that reviewing such content may be harmful to mental health and could even lead to PTSD.

So, is it purely an evil design at play at Social Media giants when it comes to moderating?

Image Credit: VICE

To be fair to all, Facebook and other Social media giants do face somewhat of an uphill battle in their efforts of moderating the content. The moment any post, video, or content gets tagged or labeled as requiring fact-checking or misleading or inappropriate, the authors or posters are quick to raise hell about dictatorship, suppression of free speech and infringement of people’s inalienable right of expression.

There is always a debate between balancing free speech versus freedom from cruelty and hatred. Or debate between balancing freedom of expression versus right to speak against bullies. A recent attempt by Twitter to mark certain tweets from the President caused a storm and PR crisis. A similar attempt from Facebook recently drew ire of conservatives and put certain ad revenue under threat.

Aside from the morals and ethics, at the heart of the debate is a purely financial question: content attracts viewers. More viewers equals more content and vice versa. Any attempt to reduce content, even the borderline inappropriate content will reduce viewers hence impacts revenue. The business models chosen by Facebook, Twitter and Google favor an unremitting, unrelenting drive to add more users and demonstrate growth to investors. More users and more content means more content to moderate and more nuances, but all of that is secondary, a kind of an afterthought.

The debate on the usage of internet and governing content uploads is not new. The debate has been going on for some time now and is just about reaching peak interest levels around the world, with many governments promising action like EU and UK; few governments, like China, in fact taking strong action; and few just watching how the entire debate pans out and what, if any, changes come out as result.

The big tech players around the world have realized one thing – it’s a tough tight rope walk to control or govern the internet. If a platform puts in too strict controls, through user-reporting mechanism, AI backed algorithms and human monitors flagging and removing content, it will get labeled as ‘dictatorship’ and against free speech. If a platform puts in too few controls, hosting content freely and with little censorship, its going to get run over by activists from all ends of the spectrum, from left to right. It’s quite like an overflowing pot left simmering for long. The only difference is no one can lift the pot and no matter which way its tilted, boiling hot contents are sure to leave scalding marks.

With that, lets take a look at The role of Regulators in the efforts to moderate content.

When Mark Zuckerberg wrote the oped in WAPO in March 2019 asking for government and regulators to step in more aggressively to police the internet, he may have elaborated what many insiders feel regarding governing internet, and specifically what content is uploaded for viewers to view, download and use. Yet, not everything seems above board here as the challenges that Mark Zuckerberg cited so eloquently in his oped are the same challenges that have plagued tech industry for years. What has changed recently that governments are being called in to action, while so far the tech industry has fought tooth and nail for freedom of expression and freedom of speech?

As the efforts to govern the internet continue, many who are fighting the battle daily are coming to realize the magnitude of difficulty this seemingly simple question of ‘what content to be allowed’ poses. Lets face it: Internet was never known to be deferential to peoples’ preferences. The advocates of freedom of expression and free speech, often big tech companies themselves, fought for as little government control as possible, decrying every move made by governments or regulators around the world.

Technology experts, including big tech companies themselves believe that for Zuckerberg and other big tech companies, “regulation” isn’t an uncouth word anymore. As with changing times, the big tech is now embracing regulations, not because of any newfound respect for regulations but purely as a business measure. From early days when big tech companies projected all regulations as reprehensible and fought any and all regulations tooth and nail, to the current day where they are welcoming regulations, the transformation cannot be more melodramatic.

Most of big tech today sees regulations as a set of common rules enforced by governments and regulators that’ll allow them to further cement their dominance of the internet. And if anything goes wrong, they always have the comfort of pointing the finger to the” Regulator” big brother.

So what is the Way forward to moderate content and to maintain integrity of platforms

According the NYU Stern report, the solution is straight-forward, and calls for increased investment, focus and commitment. The solution is a multi-pronged approach. The first step of this approach begins with Ending outsourcing: to ensure all content moderators are official Facebook or Twitter or employees of the Social media company, with adequate salaries. Increasing the number of moderators significantly is another, as well as placing content moderation under the dedicated oversight of a senior executive.

Facebook or Twitter or other Social media companies should also expand oversight in underserved countries, the report suggests.  In addition, the health and well being of content moderators employees should come first. The company should sponsor research into the mental health impacts of moderating the world’s content. And the company should expand fact-checking to curb the spread of misinformation.

Image Credit: Techcrunch

In an industry which is not limited by any conventional restraints, Apple® has once again proven to the thought and market leader in promoting what it’s users want. It is not however what Apple® has promised to do, as its customer and user-centricity is quite legendary, rather it’s the timing of the changes that is quite interesting.

During the Worldwide Developers Conference (WWDC) held remotely for the first time, Apple® announced some new privacy and security features for iOS. Apple®’s 31st Worldwide Developers Conference 2020 was a digital-only event kicked off June 22. WWDC is Apple®’s annual Worldwide Developers Conference where developers can attend sessions and meet with Apple® engineers and this year’s event witnessed Apple®  debut iOS 14, iPadOS 14, macOS Big Sur, tvOS 14, and watchOS 7. The online event typically allows millions of developers worldwide to get close proximity access to future versions of iOS, iPadOS, macOS, watchOS, and tvOS, as well as engage and network with Apple® engineers and community through engineering sessions, one-on-one lab appointments, and the revamped Apple®  Developer Forums. The event for the first time had no physical gathering in California due to the ongoing global health crisis, making everyone sorely miss the Networking and touch and feel of events like this.

During this year’s event, Apple® provided a full digital WWDC experience with online keynote, a Platforms State of the Union for developers, technical and design-focused engineering sessions, Apple® Developer Forums with Apple® engineer participation, and one-on-one developer labs. Apple® also hosted a Swift Student challenge, though winners received a jacket, pins, and virtual one-on-one lab sessions with Apple® engineers rather than free admission to WWDC.

Image Credit: appleinsider

With the new changes Apple® announced, publishers need to quite very much overhaul their entire value offering to stay compliant. As compared to the earlier practice of virtual free for all, publishers will now be required to provide information about their app’s data collection practices while explicitly seeking permission from their users to track the users’ shadow across apps and websites owned by other companies. This is a major change for the industry players, since thus far, users were required to opt-out if they wanted from sharing their data and identifiers with third party networks. To aggravate the matters, the opt-out process itself in most cases was not straightforward, with self-regulation and delayed actions in most cases giving no real out to the users. This gave the publishers of the app a virtual free rein to use the users’ implied consent in any way the publishers deemed fit. That all is set to change as now with the new changes, users will have to explicitly opt-in within their apps to allow tracking or sharing of their data.

The updates mean that you can limit how much location information is shared with apps — only allowing it approximate data rather than your precise whereabouts. Apple® also introduced recording indicators through an orange dot on your status bar that will tell you when your camera or microphone is activated. Apple® also introduced labels for app permissions to inform people how much data an app requests before they download them. The feature will show people those labels in two categories, on “Data Linked To You” and “Data Used to Track You.”

The last few years have seen an entire new industry segment open up which specializes in driving users to download an app, accessing user information, sometimes with consent, sometimes without, then tracking users across apps, collecting data and then selling the data, a process leading to shortcut of app monetization. Instead of charging user to use the app, the app pays itself through tracking, collecting and selling user data. While users complained, and while the more privacy minded users combed through the fine print of legal agreements and terms of conditions or terms of services for each app they installed, most users even those who were concerned, just shrugged and moved on. Most users just took it as the cost of using apps.

An important point to note here is that apps don’t exist in isolation or in silos. There is almost a complete app ecosystem, and as with any ecosystem, the app ecosystem requires continuous sustenance. From App development to app launch and promotion, to making the app stick with target audience, then capturing and tracking of data, and finally utilizing the data, are key parts of the ecosystem. On the technical side, or for the more technically minded, Apple® uses something called an IDFA, the Identifier for Advertisers (IDFA), which is a random device identifier assigned by Apple® to a user’s device. Advertisers use this to track data so they can deliver customized advertising. Each IDFA contains no personal identifying information, or PII, instead serving to measure and identify user interactions with ad campaigns, installs, and in-app activity. Some industry insiders put the value of app driven advertisements in iOS in excess of $45 billion dollars.

Just looking at the app development itself, unarguably the first step towards launching an app, there can be no mistake in gauging the potential of the industry. Skipping all the factors in cost breakdown, a median price to create an app by specialist agencies was found to be $171,450, while many online app cost calculators provide a price tag between $200,000 and $350,000 for an app with dozens of features. Small apps generally cost much less – apps with few basic features could cost between $10,000 and $50,000, indicating there’s an opportunity for any type of business to make a decent ROI. Needless to say, this potential saw anyone and everyone jump headfirst in to the fray.

Aided with technological advancements like microservices and APIs, it is no surprise that making, launching and monetizing apps has become big business. For a while, it really seemed to be the Wild, Wild West, with no real control, oversight or accountability.

Image Credit: cnet

The above is however set to change in a big, yet unprecedented way in near future. The upcoming changes will make it harder for app publishers, advertisers and marketers to track and target users across apps, creating a level of uncertainty on how that industry will work from now on. The changes are expected to improve transparency on users’ privacy, but they will also have an impact on the current app’s economy and how apps monetize. Some of the most important players in that space such as ad networks and attribution networks will be immediately affected.

According to a recent survey of iOS and Android users, 68.3% of iOS and 67.5% of Android users will likely deny tracking permissions if they are requested in-app to opt-in. With Android sure to follow, this is definitely good news for privacy minded and not-so-privacy-minded users alike as the move spearheaded by Apple^® will strengthen the privacy laws in the entire marketplace.

Some critics do feel that it’s not big tech or app developers pushing back, it’s the GDPR in Europe that is forcing companies like Apple^® to announce such measures. Coupled with an increasing awareness of the privacy violations that have been suffered by average consumer over the last decade or more, firms are bound to act sooner than later. While this may be partially true, there is no denying that the end result is what privacy advocates have always wanted.

To summarize, the upcoming privacy changes in iOS as announced in WWDC are forecasted to have a significant impact not only on how apps monetize their audience and how advertising and attribution networks work, but also the entire app ecosystem. The study showed that a large proportion of the users will not allow themselves to be tracked, a crucial step in effectively running and attributing ad campaigns in the apps space at the moment. Similar findings were also revealed with Android users. Privacy seems to be a key concern among users in the mobile space and providing them with more transparency and choices seems to change the current app landscape dramatically.

Introduction

A little more than a decade ago when the term Cloud, Cloud computing, and other terms associated with enterprise IT systems started to become part of technical discussions and mainstream IT topics, most IT professionals and managers, let aside outsiders, acted more in disbelief than with optimism. The initial reactions to Cloud computing and what it represented were mostly hostile. Instead of seeing Cloud computing for what it was and what it could become in the future, most IT managers were busy finding issues and raising concerns, mostly out of their self-induced paranoia of letting go of their little control areas. Most IT professionals viewed Cloud computing as a threat, something which would, if implemented, take away their most important assets – physical computing resources and put those in hands of an outside organization.

Fast forward 10 years, there has been a remarkable adaption of Cloud Computing services and its various sub-sections across IT industry. This remarkable adaption and emergence of market leaders AWS and Azure has fed a potential frenzy of activity around Cloud computing. The initial feelings that ranged from outright debunking Cloud’s potential at worst to careful cautious optimism at best have been left behind in dust long back. More than a decade or so later, Cloud computing has come a long way.

Image Source: World Informatix Cyber Security

Need for Security

The clouds surrounding Cloud Computing were as ominous then as they remain today, at least from security point of view. Many debated if it was possible that someone outside the organization be responsible to manage the IT infrastructure’s needs that the organization runs on? Could someone outside the organization be responsible to setup, operationalize, manage and provide security for organization’s IT and digital assets stored using IT infrastructure? The first and foremost concern expressed was around the security. As with any maturity model, the debate picked up and became a part of mainstream discussions as the industry moved towards gradual and in some cases scaled adoption of Cloud. More than a decade later, the debate is still raging. Security remains the primary concern, though the ecosystem itself has grown leaps and bounds and shifted in direction far, far away from the humble beginnings. In many ways, it is fair to say that while the ecosystem has completely transformed to the point that it bears little resemblance to what it was 10 years ago, the debate on Security has retained its position at the top of the concerns expressed by IT managers.

Image Source: Stratosphere Networks

Most organizations, IT included, recognize data as the most powerful resource today. Due to interconnected everything, every single swipe of finger or each step during the morning walks, or what one buys, how much, from where, at what price and at what time of day, every single aspect of life is being recorded and getting stored somewhere in deeper and deeper oceans of information. This massive amounts of data generated everyday by the myriad of systems in turn is being used to start controlling human decisions at first, providing intelligence and inputs to assist in human decision making process with the stated or unstated aim of taking over the decision-making process completely. Other than that, possession of the data has more immediate, financial and tangible benefits associated with it.

Image Source: IBM

The value of data has given rise to a host of platforms, systems, rules and standards aimed at keeping this treasure trove of information secure from falling in to wrong hands. With high-profile data breaches continuing to occur with alarming regularity across industries, IT security professionals are revamping their strategies to stay a step ahead. In fact, the security environment has become increasingly hostile, and the threat models so varied, that few in-house IT teams have the resources or bandwidth to keep ahead of the curve and protect their data systems.

Image Source: Sumologic

Security in Cloud

The security apparatus applied includes physical infrastructure like secure siting, resilience against natural disasters, and other specialized hardware; software and applications such as network stacks, applications exposed to internet, etc; logical assets like processes and policies, disaster recovery plans, etc; and non-tangibles like skills and experience of security professionals. Every single layer of the stack is a potential vulnerability, and responsibility of IT to secure it. A weakness at any point means that someone can gain a foothold into IT infrastructure and use that leverage to move around, find further vulnerabilities, and inflict damage.

Image Source: DataFlair

The attackers with malicious intent are getting savvier by the day. Any new advancements in security technologies is met or at times even exceeded by counter-measures from perpetrators of attacks on IT systems. This means that security measures and features must be constantly revamped and upgraded. Today, only the world’s most sophisticated businesses can deploy the technology and staff needed to counter today’s threat environment. And the deeper a company’s footprint across the technology stack, the more complex (and resource-intensive) this effort becomes.

In the case of a recent, well-known breach at Target, perpetrators got in due to low security measures implemented on the systems embedded in the air conditioning control systems and used that access to penetrate more valuable, vulnerable systems on the network.

The many advantages of adapting and using Cloud as the primary answer to meet IT needs have been discussed widely. These range from financial benefits – lowering cost of ownership and maintenance,  shifting infrastructure costs from capital to operating on balance sheet, lowering upfront expenditure, and better ROI moneys invested. These financial benefits provide much needed business flexibility that comes with Cloud computing. More benefits, perhaps equally significant, range from quicker response times, to elasticity, and to scalability that are intrinsic to on-demand computing resources. Finally, a significant benefit not often discussed in business case studies is that cloud infrastructure also delivers a much resilient and stronger security stack.

Image Source: DZone

The wide range of services offered by cloud computing companies can be categorized into three basic types:

• Infrastructure as a Service (IaaS). IaaS provides users access to raw computing resources such processing power, data storage capacity, and networking, in the context of a secure data center.
• Platform as a Service (PaaS). Geared toward software development teams, PaaS offerings provide computing and storage infrastructure and also a development platform layer, with components such as web servers, database management systems, and software development kits (SDKs) for various programming languages.
• Software as a Service (SaaS). SaaS providers offer application-level services tailored to a wide variety of business needs, such as customer relationship management (CRM), marketing automation, or business analytics.

Cloud infrastructure employs a critical approach better known as “Security by design”. This approach is intrinsic to modern cloud architectures. The logical controls and policy based separation among infrastructural components ensure a level of compartmentalization and segregation that few traditional data center platforms can offer. This segregation keeps compartments secure and agnostic to each other in the event of threats and potential attacks. Thus, organizations employing cloud computing begin with a much higher security standard and employ that standard as a baseline and not an aspirational level, across the entire perimeter.

Perhaps  most importantly, as cloud computing platforms are marketed and run as services, not just code, operational security — whether it’s resilience against DDoS (distributed denial of service) attacks, proactive detection of malicious code, or ensuring the integrity of messages in transit — has become a fundamental quality of any cloud computing business. Security is paramount and it has never been more in focus.  This requirement of providing higher levels of security alone enables the information security teams to move away from a primarily reactive (and losing) position of plugging the holes towards assuming a proactive, aggressive stance that’s fundamental to an organization’s overall IT strength and technology expertise.

Image Source: SDxCentral

The focus and shift towards security, making security the prime area of strength which defines almost all other parts of IT strategy alone represents a major shift in the quality of security of technology services. And, as cloud infrastructure providers now have the scale and importance to attract the best security and reliability engineering professionals in the world, the standards are continuously being refined, getting better with each iteration and with each attack. But there’s another factor at work, as well.

Security in Cloud – A shared Responsibility Model

The way cloud security is delivered largely depends on the individual cloud provider or the cloud security solutions enterprise has chosen. However, implementation of cloud security processes is always a joint responsibility between the business owner and solution provider.

Cloud service providers treat cloud security risks as a shared responsibility. In this model, the cloud service provider covers security of the cloud itself, and the customer covers security of what they put in it. In every cloud service—from software-as-a-service (SaaS) like Microsoft Office 365 to infrastructure-as-a-service (IaaS) like Amazon Web Services (AWS)—the cloud computing customer is always responsible for protecting their data from security threats and controlling access to it.

Image Source: McAfee

Conclusion

Its not difficult, nor it is far-fetched idea to imagine that Cloud is the way of the future. In fact, cloud is the future of IT. Cloud has already become a part of most enterprise’s tactical and strategic IT plans, while others are getting there in their adaption of Cloud. It’s not a far-fetched technical story any more requiring the attention of only IT department, it’s at the center stage of revamping each organization’s arrangement of IT infrastructure and the way IT is run and managed. Security has remained and is still the most basic yet the most important challenge, and the entire industry.

Image Source: PRNews

Today, most organizations view cloud-based services as a better way to deliver the services they need, including data protection and security. The risks and costs associated with maintaining self-managed, in-house IT security teams who manage organization’s IT infrastructure as a strategic answer to address security concerns is prohibitively high. Add to it the need to constantly upgrade skills and knowledge levels to respond to ever changing security landscape, changing and ever deepening nature of attacks, and redundancy of IT equipment at a faster pace than before, make it a luxury that many cannot afford. That’s precisely why most IT teams today understand that self-managed infrastructure — whether in-house or in third-party data centers — is almost always more tedious, less secure and more expensive than more modern, cloud-based alternatives.

AI Bias: inherent flaw or mere side effect!

Originally Published April 18, 2018

Introduction

Artificial Intelligence is broadly referred to as any source, channel, device or usage application whereby tasks normally attributed to human intelligence, such as reasoning, interpretation of facts, storage and deduction of information from such facts and ultimately, decision making is reproduced outside human body and human networks.

According to a Forbes article, as technology, and, importantly, our understanding of how our minds work, has progressed, our concept of what constitutes AI has changed. Rather than increasingly complex calculations, work in the field of AI has concentrated on mimicking human decision making processes and carrying out tasks in ever more human ways.

Artificial Intelligence uses are influencing people’s lives in more ways than can be counted. Autopilots in aviation is possibly one of the earliest uses of AI. Most airline captains trust auto pilot for 95% of time during an actual typical flight. AI is behind the personal virtual assistants in the iPhones, Google Home, and Amazon Echo. AI is used to recommend shopping products and brands based on past choices and history of website viewing. AI is being used to develop driver-less cars to make commuting faster and safer. It is used to recommend shows or films according to viewer’s tastes, preferences and past choices on Netflix or what songs one may like to listen to on Spotify. AI is used to how products are marketed and customized according to target audience’s tastes and preferences while steering conversations uses have with their favorite retailers, and will be used to transform retail industry around the world.

In recent interviews, Uber’s Head of Machine Learning Danny Lange confirmed Uber’s use of machine learning for its popular features like ETAs for rides, estimated meal delivery times on UberEATS, computing optimal pickup locations, as well as for fraud detection. The pioneering research behind these features is what sets Uber different from its competitors, its executives feel. And while users ultimately get to decide that, there is no doubt that AI powered applications are not just part of the ecosystem, rather form an ecosystem of their own, which is continually expanding equally in its size and impact.

Image Credit: Psychology roots

What is Artificial Intelligence?

Artificial intelligence then in other words is simply the reproduction of human intelligence and capabilities outside the human body. The work that can be done by machines or mechanical applications or software programs carrying out very basic rudimentary and rule based decision making can be termed as Artificial Intelligence. We regularly interface with such systems on a daily basis – airline ticketing and check in systems; chat bots and intelligent IVRs replacing customer service agents; human profiling software and applications flagging any abnormal profile or profile that fits one or more among the list of many criterion for detailed human checks; banking sector applications where any abnormal transaction falling outside a pattern is flagged for manual checks are a few among immeasurable applications of Artificial intelligence. Basically, any task done by humans which involved repetitive application of mind and knowledge, can be handled to automated processes or devices. This is also known as applied AI.

As humans grow and learn newer things, apply their knowledge and continue to learn thereby expanding both the repository of knowledge and their understanding of such repository, i.e. the breadth and width of subject matter, a credible and reliable Artificial Intelligence system must have the ability to learn continually based on the amount of information available to it at any given time. As early as 1959, thinkers laid down this expectation – that its simple to teach computers how to learn, rather than teaching them everything they need to know.

As internet became common during 1990s and 2000s, the emergence of huge troves of data, and availability of such data for analysis in order to learn human needs, even before humans themselves could understand those needs, has become the new future of Information Technology. Technology is no more just to aid humans, but has the potential today to govern every aspect of human lives. And this is a trend which is only growing and becoming more pervading than ever before. The data collected today is not just about humans, but its also about the machines which collect data about humans themselves. Data from a source is not just about that source alone, but contains many different layers of abstraction providing many useful insights not just in to the primary purpose of study but in to related fields as well.

However there is another aspect of Artificial Intelligence – generalized AI. This aspect caters to much deeper, and perhaps is the truest representation of artificial intelligence. Generalized AI is what was responsible for field of study like Machine Learning and Neural Networks, both of which are central to the idea of Artificial Intelligence. The generalized AI is powered by new concepts like deep learning. The invention of deep learning, a technique which uses special computer programs called neural networks to churn through large volumes of data and is trained to identify and remember patterns in that data set, means that technology which gives a good impression of being intelligent is spreading rapidly.

Machine learning is a sub-field or an application of AI. This involves feeding tons of data gradually – whether it’s in the form of text, images or voice – and adding a classifier to this data. An example would be showing the computer an image of a woman working in an office and then labelling this as woman office worker. Over time and with addition of multiple parameters for each data set and complicated algorithms, the system will learn to make predictions based on the data it is reading. These predictions follow certain rule sets, specified and hard coded in to the machine on the basis of which decisions are made, while simultaneously also becoming inputs in to accuracy analysis to improve decision making for future data sets.

The neural networks are the newest and possibly the most exciting layer within Artificial Intelligence that is powering a new breed of research and machine learning. This new breed of deep machine learning is inspired by the natural evolution of human brain and its connected neurons. The neuron networks use a series of interconnected units or processors and are adaptive systems that can adjust their outputs based on doing, essentially ‘learning’ by example and past data sets as they go, adapting their behavior based on results. This mimics evolution in the natural world, but at a much faster pace, with the algorithms quickly adapting to the patterns and results discovered to become increasingly accurate and valid.

Image Credit : Internet

Neural networks can identify patterns and trends among data that would be too difficult or time-consuming to deduce through human research (picture going through millions of data sets and sub-sets), consequently creating outputs that would otherwise be too complex to manually code using traditional programming techniques.

And as the technology progresses and becomes ever-more complex and autonomous, it also becomes harder to understand, not just for the end users, but even for the people who built the platforms in the first place. This has raised concerns about a lack of accountability, hidden biases, and the ability to have clear visibility of what is driving life-changing decisions and courses of action. This is commonly referred to as Black Box in Artificial Intelligence, where what goes in (inputs) and what comes out (output) is visible and transparent, yet the process followed by the machine or what goes on inside the machine (the brain) is not visible. Scientists and data researchers are not able to pinpoint what learning are being adapted by the machine, though the outputs being produced are desirable.

Bias in Artificial Intelligence

AI Applications range from intelligent IVRs, data analytics, speech-to-text transcription to healthcare where it is being used in myriad of ways including detecting early signs of blindness. AI now runs quality control in factories and cooling systems in data centers. States hope to employ AI to recognize threat from terrorist propaganda sites and remove them from the web. AI is central to attempts to develop self-driving vehicles. Among the ten most valuable quoted companies in the world, seven say they have plans to put deep-learning-based AI at the heart of their operations in future or are already doing so.

Machine Learning and neuron networks are interchangeably used within Artificial Intelligence ecosystems as these represent the way AI is progressing to meet the demands from technology savvy businesses and individuals. Often, these networks create what are known as black boxes, referring to closely guarded virtual boundaries where the internal branching, logic application and evolution of the application itself is obfuscated from public view. And as is often seen, many of the AI algorithms involve use of facial recognition technology, which is deeply flawed in itself.  Facial recognition technology (FRT) which aims to analyze video, photos, thermal captures, or other imaging inputs to identify or verify a unique individual is increasingly infiltrating our lives. Facial recognition systems are being provided to airports, schools, hospitals, stadiums, shops, and can readily be applied to existing cameras systems installed in public and private spaces.

Additionally, there are already documented cases of the use of FRT by government entities that breach the civil liberties of civilians through invasive surveillance and targeting. Facial recognition systems can power mass face surveillance for the government – and already there are documented excesses, such as explicit minority profiling in China and undue police harassment in the UK. Source: Facial Recognition Technology (Part 1):
Its Impact on our Civil Rights and Liberties, report issued by United States House Committee on Oversight and Government Reform dated May 22, 2019

Reuters reported a case where a New Zealand man of Asian descent had his photo rejected by an online passport photo checker run by New Zealand’s department of internal affairs. The facial recognition systems registered his eyes as being closed by mistake. When government agencies attempt to integrate facial recognition into verification processes phenotypic and demographic bias can lead to a denial of services that the government has an obligation to make accessible to all constituents. Source: Facial Recognition Technology (Part 1):
Its Impact on our Civil Rights and Liberties, report issued by United States House Committee on Oversight and Government Reform dated May 22, 2019.

People are often too willing to trust in Artificial Intelligence applications, or machine learning mathematical models because they believe that these systems will remove human bias. However, as the machine algorithms and systems replace human processes, these machines are not generally held to similar standards. A big part of the problem is that individuals and companies that develop and apply machine learning systems, and Government regulators who have the mandate to govern the usage, show little interest in monitoring and limiting algorithmic bias. Financial and technology companies use all sorts of mathematical models and aren’t transparent about how they operate.

As with most new technologies, the research wings of the Armed Forces made initial forays in to teaching machines how to do human jobs that involved at the very least a complex application of information abstraction – or as we know it today, Artificial Intelligence and Machine Learning. From the early days of attempted use of AI and Machine Learning applications, a widely circulated though highly improbable or at the very least exaggerated, parable has been told and retold multiple times in forums as wide as coffee table discussions to online articles and research papers. While the story itself may be apocryphal, an almost impossible number of sources citing the story has given it a legendary description. Some researchers attribute this story to a paper published in 1964, almost 5 decades before concept of Artificial intelligence, machine learning etc became common.

Image Credit: Dilbert

For an amusing and dramatic case of creative but unintelligent generalization, consider the legend of one of connectionism’s first applications. In the early days of the ‘perceptron’ the army decided to train an artificial neural network to recognize tanks partly hidden behind trees in the woods. They took a number of pictures of a woods without tanks, and then pictures of the same woods with tanks clearly sticking out from behind trees. They then trained a net to discriminate the two classes of pictures. The results were impressive, and the army was even more impressed when it turned out that the net could generalize its knowledge to pictures from each set that had not been used in training the net. Just to make sure that the net had indeed learned to recognize partially hidden tanks, however, the researchers took some more pictures in the same woods and showed them to the trained net. They were shocked and depressed to find that with the new pictures the net totally failed to discriminate between pictures of trees with partially concealed tanks behind them and just plain trees. The mystery was finally solved when someone noticed that the training pictures of the woods without tanks were taken on a cloudy day, whereas those with tanks were taken on a sunny day. The net had learned to recognize and generalize the difference between a woods with and without shadows!

This bias is not limited to just one study or research project, nor is it an isolated phenomena. Whenever there is a dataset of human decisions, it naturally includes bias. This bias can manifest in many ways and could include hiring decisions, grading student exams, medical diagnosis, loan approvals, etc. In fact anything described in text, in image or in voice requires information processing – and this will be influenced by cultural, gender or race biases.

Image Credit: upliftconnect

One example of AI systems is in financial industry, where complicated algorithms assess credit worthiness and credit risk before issuing credit cards or approving small loans. The task is to filter clients in order to avoid processing any that are likely to fail to make payments. Using data of declined clients, history of bad debts, or delinquencies and associating them with a set of rules could easily provide basis of decisions that machine could produce without human intervention. However, this could also lead to biases getting built in to the system, for example, applications from certain section of people are seen to have higher rejection rates based either on their employment type or simply, gender.

The financial industry is not alone in observing these biases creeping in. LinkedIn, the popular professional and career website, for instance, had an issue where highly-paid jobs were not displayed as frequently for searches by women as they were for men because of the way its algorithms were written. This gender based discrimination came about as the initial users of the site’s job search function were predominantly male for these high-paying jobs so the machine learning aspect just ended up proposing these jobs to men – thereby simply reinforcing the bias against women

The concerns of biases creeping in are not only limited to lower level, data centric and rule based tasks in applications. These concerns are equally, and perhaps more dangerously manifested within the uses of deep machine learning that requires minimal guidance, but ‘learns’ as it goes through identifying patterns from the data and information it can access. Its use of neural networks and evolutionary algorithms resembles a tangled mess of connections that are nearly impossible for analysts to disassemble and fully understand.

As explained earlier, this lack of understanding leads to pervasive fears of something unknown lurking just beneath the surface. The nearly non-transparent ways in which the deep learning machines itself ‘learn’ to navigate our worlds, at the very least, demand closer inspection in order to be fully understood, if not fully trusted.

Image Credit: Internet

Another relevant real world example here is of how Google ranks search results. This is a notoriously secret formula, and has been the subject of countless allegations of anti-trust movements, conjectures of how the system manipulates search results and other far more wild theories. Digital agencies and professionals have made and continue to make careers out of their own interpretation of the rules of the game, trying to deliver what they think Google wants to be able to boost their rankings. Many functions are powered by complex algorithms, code, programming or servers, and yet are still deemed trustworthy enough for investing large chunks of the marketing budget.

A system called COMPAS, made by a company called Northpointe, offers to predict defendants’ likelihood of reoffending, and is used by some judges to determine whether an inmate is granted parole. The working details of this program are guarded under heavy veils of secrecy, but an investigation by ProPublica found evidence that the model may be biased against minorities.

The much touted facial recognition algorithm used by various products and applications worked for perfectly for lighter-skinned subjects, however it couldn’t recognize many pictures of people with darker complexion. It’s not a unique problem; in 2015, a Google algorithm classified faces of black people as gorillas.

There have been calls to end the use of ‘black box’ algorithms in government because, without true clarity on how they work, there can be no accountability for decisions that affect the public. Fears have also been raised over the use of bias within decision-making algorithms, but with a perceived lack of a due process in place to prevent or protect against this.

Image Credit: Internet

Bias Elimination: Is it possible?

As deep learning systems, neural networks and machines become capable of performing profound and abstract functions, the fear from AI going rampant becomes more real. The real fear stems not so much from the machines suddenly stop obeying humans or start behaving of their own volition, or at least attempt to do so, rather the real fear arises from the fact that machines do exactly what they are told to do but do so in an incomprehensible manner where creators and users of the technology do not have any visibility and hence lack control. The reason behind this fear is simple – the deep learning networks are taught (LEARN) to rearrange their digital processes in response to the data ingested. The bits of computer code within the neuron networks is designed to copy the way human brains work, by drawing connections between events and information. This implies that once the network is designed and trained, even the creators of the network have little control or even visibility in to what it is doing. Of course, permitting such machines and systems to make critical decisions, be it in financial or healthcare industries or anywhere else, amounts to putting human lives in to the hands of machines, the working and operations of which are still not fully understood. It is akin to a beta program being allowed to fly jet full of passengers and cargo over densely populated areas, and though there are pilots present in the cockpit, their role is reduced to observers’ role.

In a MIT Technology review article from October 2017 titled Forget Killer Robots—Bias Is the Real AI Danger, the problem of bias in machine learning is likely to become more significant as the technology spreads to critical areas like medicine and law, and as more people without a deep technical understanding are tasked with deploying it. Some experts warn that algorithmic bias is already pervasive in many industries, and that almost no one is making an effort to identify or correct it

To rid the AI and deep machine learning from these biases is a complex proposition. As previously discussed, these biases exist due to multiple factors, from type of use of application, to data based biases, to people based reasons, and finally due to a regulatory apathy.

Data based biases general refer to biases included in AI applications and machine learning due to un-clean, poor quality data with biased results already included being uploaded in to a system. This is no brain-teaser, as ‘Garbage in-Garbage out’ is a pretty well-known pitfall.

People based biases generally refer to biases arising out of how interpretation of business rules and data parameters is decided, as well as disinterested stakeholders who are interested in the bottom line and not ridding the AI applications and machine learning systems of long term health and transparency.

Biases due to type of application refer generally to usage based biases and lack of transparency depending largely on the industry type. For example, security, finance and healthcare generally tend to operate with largely secretive processes and restrict access to their processes and systems to any outsiders making it difficult to detect such biases.

Finally, lack of any regulatory oversight, is seen as the final nail in the coffin. The overall perception is that biased algorithms and systems are everywhere, yet no one seems to care. In-house regulators are generally under significant pressure to minimize any application downtime or to suggest expensive fixes. External watchdogs generally lack teeth, while those external regulators who wield some kind of power primarily through policy formulation, typically government agencies, are seldom interested in regulating the uses of AI and machine learning. And while the Congress may be rightly worried about the threats of this unknown equation, the progress seems to be slower and farther than what is needed.

Artificial Intelligence applications suffer from a natural, lack of diversity which in many cases is indicative of deep-rooted cultural norms. AI assistants like Apple’s Siri or Amazon’s Alexa have default female names, voices, and personas, are largely seen as helpful or passive supporters of a user’s lifestyle due in part to their personal messaging and in part to their shopping list task orientation. Meanwhile, their counterparts like IBM’s Watson or Salesforce’s Einstein are perceived as complex problem-solvers tackling global issues, perhaps in part at least due to their male-branded disposition and in part due to their outward problem solving approach. The quickest way to flip this public perception on its head is to render AI genderless. The more long-term approach requires expanding the talent pool of people working on the next generation of AI applications and machine learning technologies to include multiple viewpoints and dispositions in producing a homogenous product, which is culturally neutral yet sensitive to audiences’ diverse requirements.

Another approach is to introduce adequate ‘Bias’ testing before releasing any new version of an existing AI application. Software and application testing is already pretty standardized functions within software development processes, however bias testing is something which is still unknown even among the technology giants. Bias testing must become de-facto for AI applications, while for continually learning machine learning or deep learning applications, the machine or the system must be made complaint to run ‘Bias’ testing standards before fresh automated deployments, or in other cases at regular, frequent intervals, for example, once every 15 days or 30 days depending on the feasibility.

Image Credit: TNW

Conclusion

Facebook’s news feed algorithm can certainly shape the public perception of events, social interactions and even major news events. Other algorithms may already be subtly distorting the kinds of medical care a person receives, or how they get treated in the criminal justice system. Most of the times the biases in AI are not well-known, these may be subtle, hard to notice under the surface ripples which even when noticed are generally dismissed as non-events. On other occasions the biases may not be even shared publically, partly due to non-discovery and partly due to illicit attempts at fix-the-problem before-it-gets-out-of-hand approach.

In real terms, the threat from AI is perhaps not as serious as its usually portrayed in books or movies. However it doesn’t take much for the threat to be manifested in the present day, rather than at some indeterminable date in future. That is the biggest threat of AI where many believe that the AI applications and systems today have the potential to spin uncontrollably out of hands of their creators and users.

As AI is used to produce more meaningful results and power more and more high profile and public facing services, such as self-driving cars, medical treatment suggestions, early warning systems in fields such as oil exploration, or automated defense systems, concerns have understandably been raised about what is going on behind the scenes. If people are willing to put their lives in the hands of AI-powered and machine learning applications, then they would want to be sure that someone understands how the technology works, how in-built biases take effect and how it powers decisions, and more importantly, that someone has the power and means to remove these biases or at least restrict the applications if the biases get out of hand, or if the need arises in any other manner.