Ethics (Artificial Intelligence) – Defending Our Privacy

Anil Kumar Kummari

Now a days All are accessible to use high end technology in mobile phones, for everything we are we are giving our information like fingerprint, Eye Scanner ( Retina Recognition ), health Apps will track your moments also. All our information is in the at one place. Once it hacked or miss used it will be at risk. As artificial intelligence proliferates, companies and governments are aggregating enormous data sets to feed their AI initiatives.

Although privacy is not a new concept in computing, the growth of aggregated data magnifies privacy challenges and leads to extreme ethical risks such as unintentionally building biased AI systems, among many others. Privacy and artificial intelligence are both complex topics. There are no easy or simple answers because solutions lie at the shifting and conflicted intersection of technology, commercial profit, public policy, and even individual and cultural attitudes. Data protection officials from more than 60 countries expressed their concerns over challenges posed by the emerging fields of robotics, artificial intelligence and machine learning due to the new tech's unpredictable outcomes. The global privacy regulators also discussed the difficulties of regulating encryption standards and how to balance law enforcement agency access to information with personal privacy rights.

Such technological developments “pose challenges for a consent model of data collection,” and may lead to an increase in data privacy risks, John Edwards, New Zealand privacy commissioner, said at the 38th International Data Protection and Privacy Commissioners' Conference, in Marrakesh, Morocco. For example, decision-making machines may be used to “engender or manipulate the trust of the user,” and would be an “all seeing, all remembering in-house guests,” that would collect personal data via numerous sensors. Peter Fleischer, global privacy counsel at Alphabet Inc.'s Google, said that established privacy principles would continue to be relevant for new technologies, but machine learning raised particular problems, such as machines finding “ways to re-identify data.”

The emerging technologies may have a broad impact across various industries. “Humans teaching machines to learn” was a “revolution in the making” that may have broad societal consequences that could cut across numerous economic sectors, Fleischer said. For example, data-driven machines may have the ability to analyse sensitive medical data, make medical diagnoses, thereby potentially revolutionizing the health-care industry, Fleischer said at the conference. Machines that learn would act “like a chef: see the ingredients and comes up with something new,” he said.

“Before the prospect of an intelligence explosion, we humans are like small children playing with a bomb. Such is the mismatch between the power of our plaything and the immaturity of our conduct.”

Nick Bostrom, Professor in AI Ethics and Philosophy at the University of Oxford

Google CEO Sundar Pichai thinks we are now living in an “artificial intelligence-first world.” He’s probably right. Artificial intelligence is all the rage in Silicon Valley these days, as technology companies race to build the first killer app that utilizes machine learning and image recognition. Today, Google announced an AI-powered assistant built into its new Pixel phones. But there’s a pivotal downside to the company’s latest creation: Because of the very nature of artificial intelligence, our data is less secure than ever before, and technology companies are now collecting even more personal information about each one of us.

Re-Defining Privacy

Unfortunately, the answer is no. We cannot turn back time. There is no completely private space available to us, anymore. Most of the things we do are already registered as data somewhere (and this occurs as soon as we do them). Purpose limitation is not always possible. We have fallen in love with the algorithmically driven companies that utilize technology to deliver an instantly better user experience. They pervade all aspects of everyday life. We already live in a world of big data. In addition, we cannot stop the emergence of artificial intelligence. The Internet of Things means that all of our devices are already connected (or will be connected in the near future). Connected and smart cities will continue to make our lives better. Our telephones already keep track of our moves and our connections and favourite places. Smart fridges keep track of our groceries. The list goes on and on. Moreover, perhaps most obviously, we love being connected and sharing our lives with others, via social media and other online platforms.

This does not mean that privacy disappears or that it ceases to matter.

Privacy is, and will continue to be, enormously important. Rather, privacy has been transformed by the proliferation of network technologies and the new forms of unmediated communication that such technologies facilitate.

In particular, technology has changed the character of the “zone” of privacy that people expect to be protected. There has been a shift from a settled space based on a clear distinction between public and private life to a more uncertain and dynamic zone that is constructed by and between individuals. Privacy as a well-defined space over which a person has “ownership” has been replaced by a more complex space that is constantly being negotiated and contested.

Work is similarly transformed. Businesses are becoming more flexible ecosystems / networks / platforms. “Lifetime” employment is no longer feasible or even desirable in a digital world. Working relationships become looser and more transitory as businesses are introducing more flexible work arrangements in which “employees” are “hired” for well-defined, but successive “tours of duty”.

Keeping artificial intelligence data in the shadows

One way for IT to address data privacy issues with machine learning is to "mask" the data collected, or anonymize it so that observers cannot learn specific information about a specific user. Some companies take a similar approach now with regulatory compliance, where blind enforcement policies use threat detection to determine if a device follows regulations but do not glean any identifying information.

“With AI it becomes easier to correlate data ... and remove privacy.”

Brian Katz

Device manufacturers have also sought to protect users in this way. For example, Apple iOS 10 added differential privacy, which recognises app and data usage patterns among groups of users while obscuring the identities of individuals. Tools such as encryption are also important for IT to maintain data privacy and security. Another best practice is to separate business and personal apps using technologies such as containerisation. Enterprise mobility management tools can be set up to look at only corporate apps but still be able to white list and blacklist any apps to prevent malware. That way, IT does not invade users' privacy on personal apps.

Reference:- 

1. https://gizmodo.com/googles-ai-plans-are-a-privacy-nightmare-1787413031
2. http://searchmobilecomputing.techtarget.com/news/450419686/Artificial-intelligence-data-privacy-issues-on-the-rise
3. https://medium.com/startup-grind/artificial-intelligence-is-taking-over-privacy-is-gone-d9eb131d6eca

Ethics (Artificial Intelligence) – Bio-terrorism

 

Anil Kumar Kummari

As Stephen Hawking noted in 2014, “Whereas the short-term impact of AI depends on who controls it, the long-term impact depends on whether it can be controlled at all”.

THE MEDICINE OF TOMORROW

As the medical wearable and sensor, market starts to truly boom, it is logical to think ahead to what might follow this “wearable revolution.” I think that the next step will be insideables, digestible, and digital tattoos.

“Insideables” means devices implanted into the body, generally just under the skin. In fact, there are people who already have such implants, which they can use to open up a laptop, a smartphone, or even the garage door. “Digestible” are pills or tiny gadgets that can be swallowed, which could do things like track digestion and the absorption of drugs. “Digital tattoos” are tattoos with “smart” capabilities. They might easily measure all of our health parameters and vital signs.

All of these teeny-tiny devices might be misused—some could be used to infuse lethal drugs into an organism or strip a person of their privacy. That is the reason why it is of the utmost importance to pay attention to the security aspect of these devices. They can be vulnerable to attacks, and our life will (quite literally) depend on the safety precautions of the company developing the sensors. That may not sound too comforting—putting your health in the hands of a company—but microchip implants are heavily regulated in the US, and so we are already looking ahead to issues surrounding this advancement.

1) Hacking medical devices

It has already been proven that pacemakers and insulin pumps can be hacked. Security experts have warned us that someone would be murdered through these methods any time soon. How can we prevent wearable devices that are connected to our physiological system from being hacked and controlled from a distance?

2) Bioterrorism due to nanotechnology

In the wildest futuristic scenarios, tiny Nano robots in our bloodstream could detect diseases. After a few decades, they might even eradicate the word symptom inasmuch as no one would have them any longer. These microscopic robots would send alerts to our smartphones or digital contact lenses before disease could develop in our body. If it becomes reality, and micro robots swimming in bodily fluids are already out there, how can we prevent terrorists from trying to hack these devices controlling not only our health but also our lives.

THE TINY ROBOT REVOLUTION

In the future, nanoscale robots could live in our bloodstream or in our eyes and prevent any diseases by alerting the patient (or doctor) when a condition is about to develop. They could interact with our organs and measure every health parameter, intervening when needed.

Nanobots are so tiny that it is almost impossible to discover when someone, for example, puts one into your glass and you swallow it. Some people are afraid that, by using such tiny devices, total surveillance would become feasible. There also might be the possibility there to utilize nanobots to deliver toxic or even lethal drugs to the organs.

By researching ways to identify when these nanobots are being utilized now, we could potentially prevent their misuse in the future.

AUGMENTING INTELLIGENCE

In the future, brain implants will be able to empower humans with superpowers with the help of chips that allow us to hear a conversation from across a room, give us the ability to see in the dark, let us control moods, restore our memories, or “download” skills like in The Matrix movie trilogy. However, implantable neuro-devices might also be used as weapons in the hands of the wrong people.

Conclusion

Bioterrorism remains a legitimate threat both from domestic and international terrorist groups. From a public health perspective, timely surveillance, awareness of syndromes resulting from bioterrorism, epidemiologic investigation capacity, laboratory diagnostic capacity and the ability to rapidly communicate critical information on a need to know basis to manage public communication through the media are vital. Ensuring adequate supply of drugs, laboratory reagents, antitoxins and vaccines is essential. Formulating and putting into practice, SOPs/drills at all levels of health care will go a long way in minimising mortality and morbidity in case of a bioterrorist attack.

Reference:-

1.  https://futurism.com/the-future-of-extremism-artificial-intelligence-and-synthetic-biology-will-transform-terrorism/
2.  http://medicalfuturist.com/list-of-ethical-issues-in-the-future-of-medicine/

Ethics (Artificial Intelligence) - Artificial 

Stupidity

Anil Kumar Kummari

Futurists worry about artificial intelligence becoming too intelligent for humanity’s good. Here and now, however, artificial intelligence can be dangerously dumb. When complacent humans become over-reliant on dumb AI, people can die. The lethal record of accomplishment goes from the Tesla Autopilot crash last year, to the Air France 447 disaster that killed 228 people in 2009, to the Patriot missiles that shot down friendly planes in 2003.

War Algorithm logo that is particular problematic for the military, which, more than any other potential user, would employ AI in situations that are literally life or death. It needs code that can calculate the path to victory amidst the chaos and confusion of the battlefield, the high-tech Holy Grail we have calling the War Algorithm. While the Pentagon has repeatedly promised it won’t build killer robots — AI that can pull the trigger without human intervention — people will still die if intelligence analysis software mistakes a hospital for a terrorist hide-out, a “cognitive electronic warfare” pod doesn’t jam an incoming missile, or if a robotic supply truck doesn’t deliver the right ammunition to soldiers running out of bullets.

“Before we work on artificial intelligence why don’t we do something about natural stupidity?” —Steve Polyak

Should we worry about how quickly artificial intelligence is advancing?

There are people who are grossly overestimating the progress that has been made. There are many, many years of small progress behind many of these things, including mundane things like more data and computer power. The hype is not about whether the stuff we are doing is useful or not—it is. However, people underestimate how much more science needs to be done. Moreover, it is difficult to separate the hype from the reality because we are seeing these great things and to the naked eye, they look magical.

 Artificial stupidity. How can we guard against mistakes?

Intelligence comes from learning, whether you are human or machine. Systems usually have a training phase in which they "learn" to detect the right patterns and act according to their input. Once a system is fully trained, it can then go into test phase, where it is hit with more examples and we see how it performs.

Obviously, the training phase cannot cover all possible examples that a system may deal with in the real world. These systems can be fooled in ways that humans would not be. For example, random dot patterns can lead a machine to “see” things that are not there. If we rely on AI to bring us into a new world of labour, security and efficiency, we need to ensure that the machine performs as planned, and that people can’t overpower it to use it for their own ends.

Artificial stupidity as a limitation of artificial intelligence. Artificial stupidity is not just delivering deliberate errors into the computer, but it could also be seen as a limitation of computer artificial intelligence.

Dr. Jay Liebowitz argues that "if intelligence and stupidity naturally exist, and if AI is said to exist, then is there something that might be called "artificial stupidity?"

Liebowitz pointed out that the limitations are:

Ability to possess and use common sense

Development of deep reasoning systems

Ability to vary an expert system's explanation capability

Ability to get expert systems to learn

Ability to have distributed expert systems

Ability to easily acquire and update knowledge

— Liebowitz, 1989, Page 109

Once a system is fully trained, it can then go into test phase, where it is hit with more examples and we see how it performs.  However, like google maps, it shows only shortest route. However, in reality, it do not know that road is closed due to temporary actions of govt.

 Artificial stupidity. How can we guard against mistakes?

What is important to keep in mind is that the training phase is not able to cover all possible scenarios that a system can come across, hence why systems can be fooled in ways that humans would not be. Importance of ensuring that the machines perform as planned and that people are not able to overpower it to use it for their own benefits. At the time of making, it must be follow three laws of Robotics. Which makes artificial stupidity followed by the machines and goes into the hands of terror groups.

Reference:-

1.  https://breakingdefense.com/2017/06/artificial-stupidity-when-artificial-intel-human-disaster/
2. https://www.technologyreview.com/s/546301/will-machines-eliminate-us/

Future of Artificial Intelligence in Cyber Security

Anil Kumar Kummari

With AI being introduced in every industry, the cyber security space would be no stranger to it. With advancement, new exploits and vulnerabilities could be easily identified and analysed to prevent further attacks. Incident response systems could also benefit greatly from AI. When under attack, the system will be able to identify the entry point and stop the attack as well as patch the vulnerability.

Studies show that it takes, on an average in 2016, 99 days for a company to realize that they have been compromised. Although a long way from 146 days in 2015, yet a very long time for the attackers to gain all the information they were looking for. This period is not only enough to steal data but also manipulate it without detection. This can have a great impact on the company as it makes it very difficult for the company to differentiate between the fake and the actual data.

With the advancements in AI, hopefully, all of the above problems would be able to mitigate the problems being faced.

“With AI it becomes easier to correlate data… and remove privacy”

Keeping artificial intelligence data in the shadows

One way for IT to address data privacy issues with machine learning is to "mask" the data collected, or anonymize it so that observers cannot learn specific information about a specific user. Some companies take a similar approach now with regulatory compliance, where blind enforcement policies use threat detection to determine if a device follows regulations but do not glean any identifying information. Device manufacturers have also sought to protect users in this way. For example, Apple iOS 10 added differential privacy, which recognizes app and data usage patterns among groups of users while obscuring the identities of individuals.

Amazon Becomes the First to Turn to Artificial Intelligence to Protect Data in the Cloud…

Amazon Web Services (AWS), it makes plenty of sense for Amazon's team of engineers and programmers to continue to place a substantial priority on keeping this sensitive info safe, secure, and out of sight of the prying eyes of digital intruders. However, the fate of your dealership's data (and that of countless other organizations) may not actually rest in human hands at all anymore. As the editorial team over at Forbes magazine explains, Amazon has blazed a new trial by becoming the first public cloud computing and storage service provider to turn to artificial intelligence (A.I.) to safeguard information held within AWS. Known as "Amazon Macie," this new safety measure leverages the power of machine learning in an effort to automatically discover, codify, and shield stored data on behalf of the service's users. In terms of how Amazon Macie works, this system utilizes machine learning to both understand the nature of potentially sensitive information and find security flaws within user accounts on AWS. From here, analysing and reporting issues to customers, including real-time alerts related to usage that the A.I.

Is artificial intelligence (AI) used to detect cyber-attacks, how is its success rate?

Of course, AI can be used to detect cyber-attacks. There are plenty of academic researches about detecting cyber-attacks using artificial intelligence.

The success rate of those researches varies between 85% and 99%.

In the last few years, in addition to academic researches, some products have been improved to detect cyber-attacks with the help of artificial intelligence like Dark Trace. Dark Trace claims to have more than 99% of success rate and it has a very low rate of false positives. For more details, you can check the company’s website.

AI Solutions for Cyber Security

Automation and false positives

Although informatics systems are prone to failure and attacks, they are a necessary help to overwhelmed security engineers. There is a growing shortage of cyber security specialists, and the mix of high-value actions and routine tasks should be divided between man and machine. Computers are expected to automatically perform daily tasks like analysing network traffic, granting access based on some set of rules and detecting abnormalities, while the cyber security specialists can work on designing algorithms and studying emerging threats. Removing false positives is also one of the main tasks that require human assistance and one of the reasons why AI is not ready to take over security completely.

Predictive analytics

Cyber threats have become more and more complex. Just gathering data about attacks like data breaches, malware types, and phishing activity and creating signatures is no longer enough. The new approach is to monitor a wide number of factors and identify patterns of what constitutes normal and abnormal activity, without looking for specific traces of a particular malicious activity, but for spikes or silent moments. Some companies even pair this with other AI-powered tools including natural language processing to speed up this process. Staying a step ahead of hackers will be increasingly difficult, as predictive analysis can be tricked with randomization.

Immunity

Learning from nature is effective not only in engineering but in cyber security as well. The body’s immune system is one of the best defensive lines in the living world. AI could be trained to behave like the white cells and antibodies, neutralizing threats that are not according to the known patterns without shutting down the whole system. This approach could be the cure to the adaptive malware previously discussed. The system learns from experiences and becomes stronger, just like an organism that has been exposed to the diseases, and overcomes it.

Hands-on Approach

Cybersecurity powered by AI is just the natural step in protecting vulnerable data. The race between those aiming to create safe systems and attackers is crossing into new territory, but machines are far away from taking the lead. Currently, both parties are restructuring their data and integrating systems. There are numerous corrective actions necessary from humans. This is a process, composed of multiple layers, not a one-time action. The defining factor remains the education of the humans involved, first as users then as protectors.

Reference:-

1.   http://bigdata-madesimple.com/will-artificial-intelligence-take-over-cyber-security/

Ethics (Artificial Intelligence) -

“Cybersecurity of sensitive data”

Anil Kumar Kummari

Do you use Siri, Google Now, Cortana or Alexa? They work by recording your voice, uploading the recording to the cloud, then processing the words and sending back the answer. After you have your answer, you forget about the query. But your recorded voice, the text extracted from it, and the entire context of the back-and-forth conversations you had are still doing work in the service of the A.I. that makes virtual assistants work. Everything you say to your virtual assistant is funnelled into the data-crunching A.I. engines and retained for analysis. In fact, the artificial intelligence boom is as much about the availability of massive data sets as it is about intelligent software. The bigger the data sets, the smarter the A.I.

Artificial Intelligence Right Now?

Siri: Apple’s personal assistant on iPhone’s and Mac OS.

Netflix: Recommendation engine

Nest: Home Automation

Alexa: Amazon’s smart hub

Gaming: Games like Call of Duty and Far Cry rely heavily on AI

News Generation: Companies like Yahoo, AP use AI to write small news stories such as financial summaries, sports recap, etc.

Fraud Detection

Customer Support: Companies have been using small scale Chat Bots to automate this process.

Self-driving cars

Speech Recognition

Robotics

Why Are Criminals Targeting Sensitive Data?

Adapting and responding to evolving cyber threats and protecting critical infrastructure and proprietary business assets are essential for both government agencies and businesses. “Post-mortem” analyses of breaches offer a treasure trove of lessons learned and reveal attack tactics, techniques and procedures. Cyber criminals leverage technology vulnerabilities and trickery to exploit the human-technology gap — by targeting sensitive passwords, data and applications regularly used by staff. Data theft is the goal of most recent breaches. Cyber criminals typically break into vulnerable systems and pivot between systems using stolen credentials or posing as a third-party contractor to gain access to valuable data. Targeted confidential data comprises personnel records, public billing information, credit card numbers, financial or health records and more. The theft of your city’s legally protected data can result in significant regulatory fines, loss of public trust and damage to the city’s reputation.  Fortune.com estimates that in 2016, the cost of data breaches averaged $4 million dollars or $158 per record. Medical history, credit card data and Social Security numbers have the highest cost per stolen record at $355.

Sensitive Data Risk Management

Data is the new currency. Traditional currency and property risk-management techniques also apply to protecting against cybercrime. Regulated or sensitive data has monetary value and makes an attractive target for cybercriminals. Reducing the amount of regulated data stored on hand is equivalent to cash management practices, such as moving excess cash from registers to a hardened safe or transporting it to a bank’s vault. Unrestricted and unmonitored employee access to a large amount of cash is typically prohibited; however, public agencies often fail to apply the same level of scrutiny for employee access to regulated or sensitive data.

Whatever the motive, it is clear that governments are the highest-value targets for hackers today. Thus, it is critical that agencies invest in strong cyber defences—stronger, if anything, than those found in the private sector.

As with modern-day terrorism, cybersecurity has proven daunting because the nature of the threat is constantly evolving. Each major technological development—mobile, social, cloud computing—brings a host of new risks.

“With AI it becomes easier to correlate data ... and remove privacy”

Keeping artificial intelligence data in the shadows

One way for IT to address data privacy issues with machine learning is to "mask" the data collected, or anonymize it so that observers cannot learn specific information about a specific user. Some companies take a similar approach now with regulatory compliance, where blind enforcement policies use threat detection to determine if a device follows regulations but do not glean any identifying information.

Device manufacturers have also sought to protect users in this way. For example, Apple iOS 10 added differential privacy, which recognises app and data usage patterns among groups of users while obscuring the identities of individuals.

References:-

1. https://www.computerworld.com/article/3035595/emerging-technology/artificial-intelligence-needs-your-data-all-of-it.html
2. https://remora.com/blog/amazon-macie-machine-learning-cloud-storage-security

Ethics (Artificial Intelligence) - Racist Robots

How Do We Eliminate AI Bias?

Anil Kumar Kummari

We have already seen glimpses of what might be on the horizon. Programs developed by companies at the forefront of AI research have resulted in a string of errors that look uncannily like the darker biases of humanity: a Google image recognition program labelled the faces of several black people as gorillas; a LinkedIn advertising program showed a preference for male names in searches, and a Microsoft chatbot called Tay spent a day learning from Twitter and began spouting anti-Semitic messages.

Tay Chatbot - Microsoft

In 2016, Microsoft released a “playful” chatbot named Tay onto Twitter designed to show off the tech giant’s burgeoning artificial intelligence research. Within 24 hours, it had become one of the internet’s ugliest experiments. By learning from its interactions with other Twitter users, Tay quickly went from tweeting about how “humans are super cool,” to claiming “Hitler was right I hate the Jews.” While it was a public relations disaster for Microsoft, Tay demonstrated an important issue with machine learning artificial intelligence: That robots can be as racist, sexist and prejudiced as humans if they acquire knowledge from text written by humans.

US-Risk Assessment System

“If you want to take steps towards changing that you can’t just use historical information.” In May last, year report claimed that a computer program used by a US court for risk assessment was biased against black prisoners.

The Correctional Offender Management Profiling for Alternative Sanctions was much more prone to mistakenly label black defendants as likely to reoffend according to an investigation by ProPublica.

ProPublica did, as part of a larger examination of the powerful, largely hidden effect of algorithms in American life. Obtained the risk scores assigned to more than 7,000 people arrested in Broward County, Florida, in 2013 and 2014 and checked to see how many were charged with new crimes over the next two years, the same benchmark used by the creators of the algorithm. The score proved remarkably unreliable in forecasting violent crime: Only 20 percent of the people predicted to commit violent crimes actually went on to do so.

When a full range of crimes were taken into account — including misdemeanours such as driving with an expired license — the algorithm was somewhat more accurate than a coin flip. Of those deemed likely to re-offend, 61 percent were arrested for any subsequent crimes within two years.

We also turned up significant racial disparities, just as Holder feared. In forecasting who would re-offend, the algorithm made mistakes with black and white defendants at roughly the same rate but in very different ways. The formula was particularly likely to falsely flag black defendants as future criminals, wrongly labelling them this way at almost twice the rate as white defendants. White defendants were mislabelled as low risk more often than black defendants.

BRISHA BORDEN

Prior Offenses

4 juvenile misdemeanors

HIGH RISK8

Subsequent Offenses

                                   None

VERNON PRATER

Prior Offenses

2 armed robberies, 1 attempted armed robbery

LOW RISK3

Subsequent Offenses

1 grand theft

Could defendants’ prior crimes or the type of crimes they were arrested for explain this disparity? No. We ran a statistical test that isolated the effect of race from criminal history and recidivism, as well as from defendants’ age and gender. Black defendants were still 77 percent more likely to be pegged as at higher risk of committing a future violent crime and 45 percent more likely to be predicted to commit a future crime of any kind.

Maxine Mackintosh, a leading expert in health data, said the problem is mainly the fault of skewed data being used by robotic platforms. Machine learning may be inherently racist and sexist if it learns from humans and will typically favour white men, research has shown. Machine-learning algorithms, which will mimic humans and society’s actions, will have an unfair bias against women and ethnic minorities.

The white suspect had prior offences of attempted burglary and the black suspect had resisting arrest. Seemingly, giving no indication as to why, the black suspect was given a higher chance of reoffending and the white suspect was considered ‘low risk’. But, over the next two years, the black suspect stayed clear of illegal activity and the white suspect was arrested three more times for drug possession.

He added researchers at Boston University had demonstrated the inherent bias in AI algorithms by training a machine to analyse text collected from Google News. When they asked the machine to complete the sentence “Man is to computer programmers as woman is to x”, the machine answered “homemaker”. Stopping racist, sexist robots a challenge for AI Health data expert Maxine Mackintosh said that the problem lies with society, and not the robots. She said: “These big data are really a social mirror – they reflect the biases and inequalities we have in society. “If you want to take steps towards changing that you can’t just use historical information.”

“People expected AI to be unbiased; that’s just wrong”

This is the threat of AI in the near term. It is not some sci-fi scenario where robots take over the world. Its AI-powered services making decisions we do not understand, where the decisions turn out to hurt certain groups of people.

Refence:-

1.   https://www.theguardian.com/inequality/2017/aug/08/rise-of-the-racist-robots-how-ai-is-learning-all-our-worst-impulses

2.   https://www.theverge.com/2016/3/24/11297050/tay-microsoft-chatbot-racist

3.   https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing