The Robotics Summit and Showcase is just a couple months away. Find out all about our agenda here and register by April 20 for a 20% discount to learn from the best in the robotics industry.
Robot Lightsabers duel light vs dark side at @Universal_Robot #SXSW show floor. Is there a robot with a lightsaber yet in Star Wars? Grievous doesn’t count pic.twitter.com/lIihUJ5gpR
— Digital LA (@DigitalLA) March 13, 2018
As Mark Hamill humorously shared the behind-the-scenes of “Star Wars: The Last Jedi” with a packed SXSW audience, two floors below on the exhibit floor Universal Robots recreated General Grievous’ famed light saber battles. The battling machines were steps away from a twelve foot dancing Kuka robot and an automated coffee dispensary. Somehow the famed interactive festival known for its late night drinking, dancing and concerts had a very mechanical feel this year. Everywhere debates ensued between utopian tech visionaries and dystopia-fearing humanists.
Even my panel on “Investing In The Autonomy Economy” took a very social turn when discussing the opportunities of utilizing robots for the growing aging population. Eric Daimler (formerly of the Obama White House) raised concerns about AI bias affecting the well being of seniors. Agreeing, Dan Burstein (partner at Millennium Tech Value Partners) nervously expressed that ‘AI is everywhere, in everything, and the USA has no other way to care for this exploding demographic except with machines.’ Daimler explained that “AI is very good at perception, just not context;” until this is solved it could be a very dangerous problem worldwide.
Last year at a Google conference on the relationship between humans and AI, the company’s senior vice president of engineering, John Giannandrea, warned, “The real safety question, if you want to call it that, is that if we give these systems biased data, they will be biased. It’s important that we be transparent about the training data that we are using, and are looking for hidden biases in it, otherwise we are building biased systems.” Similar to Daimler’s anxiety about AI and healthcare, Giannandrea exclaimed that “If someone is trying to sell you a black box system for medical decision support, and you don’t know how it works or what data was used to train it, then I wouldn’t trust it.”
One of the most famous illustrations of how quickly human bias influences computer actions is Tay, the Microsoft customer service chatbot on Twitter. It took only twenty-four hours for Tay to develop a Nazi persona leading to more than ninety thousand hate-filled tweets. Tay swiftly calculated that hate on social media equals popularity. In explaining its failed experiment to Business Insider, Microsoft stated via email: “The AI chatbot Tay is a machine learning project, designed for human engagement. As it learns, some of its responses are inappropriate and indicative of the types of interactions some people are having with it. We’re making some adjustments to Tay.”
Related: 10 tech-savvy companies on the hunt for AI/robotics talent and IP
While Tay’s real impact was benign, it raises serious questions of the implications of embedding AI into machines and society. In its Pulitzer Prize-winning article, ProPublica.org uncovered that a widely distributed US criminal justice software called Correctional Offender Management Profiling for Alternative Sanctions (COMPAS) was racially bias in scoring the risk levels of convicted felons to recommit crimes. ProPublica discovered that black defendants in Florida, “were far more likely than white defendants to be incorrectly judged to be at a higher rate of recidivism” by the AI. Northpointe, the company that created COMPAS, released its own report that disputed ProPublica’s findings but it refused to pull back the curtain on its training data, keeping the algorithms hidden in a “black box.” In a statement released to The New York Times, Northpointe’s spokesperson argued, “The key to our product is the algorithms, and they’re proprietary. We’ve created them, and we don’t release them because it’s certainly a core piece of our business.”
The dispute between Northpointe and ProPublica raises the question of transparency and the auditing of data by an independent arbitrator to protect against bias. Cathy O’Neil, a former Barnard professor and analyst at D.E. Shaw, thinks a lot about safeguarding ordinary Americans from biased AI. In her book, Weapons of Math Destruction, she cautions that big corporate America is too willing to hand over the wheel to the algorithms without fully assessing the risks or implementing any oversight monitoring.
“[Algorithms] replace human processes, but they’re not held to the same standards. People trust them too much,” declares O’Neil. Understanding the high stakes and lack of regulatory oversight by the current federal government, O’Neil left her high-paying Wall Street job to start a software auditing firm, O’Neil Risk Consulting & Algorithmic Auditing. In an interview with MIT Technology Review last summer, O’Neil frustratingly expressed that companies are more interested in the bottom line than protecting their employees, customers, and families from bias, “I’ll be honest with you. I have no clients right now.”
Most of the success of deconstructing “black boxes” is happening today at the US Department of Defense. DARPA has been funding the research of Dr. David Gunning to develop Explainable Artificial Intelligence (XAI). Understanding its own AI and that of foreign governments could be a huge advantage for America’s cyber military units. At the same time, like many DARPA-funded projects, civilian opportunities could offer societal benefits. According to Gunning’s statement, online XAI aims to “produce more explainable models, while maintaining a high level of learning performance (prediction accuracy); and enable human users to understand, appropriately trust, and effectively manage the emerging generation of artificially intelligent partners.” XAI plans to work with developers and user interfaces designers to foster “useful explanation dialogues for the end user,” to know when to trust or question the AI-generated data.
Besides DARPA, many large technology companies and universities are starting to create think tanks, conferences and policy groups to develop standards that test AI bias. The results have been startling – ranging from computer vision sensors that negatively identify people of color to gender bias in employment management software to blatant racism of natural language processing systems to security robots that run over kids identified mistakenly as threats. As an example of how training data affects outcomes, when Google first released its image processing software, the AI identified photos of African Americans as “gorillas,” because the engineers failed to provide enough minority examples into the neural network.
Ultimately AI reflects the people that program it, as every human being brings with him his own experiences that shape personal biases. According to Kathleen Walch, host of AI Today podcast, “If the researchers and developers developing our AI systems are themselves lacking diversity, then the problems that AI systems solve and training data used both become biased based on what these data scientists feed into AI training data.”
Walch advocates that hiring diversity can bring “about different ways of thinking, different ethics and different mindsets. Together, this creates more diverse and less biased AI systems. This will result in more representative data models, diverse and different problems for AI solutions to solve, and different use cases feed to these systems if there is a more diverse group feeding that information.”
Before leaving SXSW, I attended a panel hosted by the IEEE on “Algorithms, Unconscious Bias & AI,” amazingly all led by female panelists, including one person of color. Hiring basis became a big theme of their discussion. Following the talk, I hopped into my Uber and pleasantly rode to the airport reflecting on a statement made earlier in the day by John Krafcik, Cheif Executive, of Waymo. Krafcik boasted that Waymo’s mission is to build “the world’s most experienced driver.” I just hope the training data is not from New York City cabbies.
Frank Tobe says
Your article should really be titled: Bias AND MISUSED data are the real dangers of AI.
In a recent article about the two-edged sword of new technology, I wrote “The twin issues of killer robots and robots taking our jobs are the result of the two-edged blade of new technology, i.e., technologies that can be used for both good and evil. Should these new technologies be stopped entirely or regulated? Can they be regulated?”
One can easily see the bias. Just have three of your friends pull up the news.google.com page on their computers and see the difference. Google has altered the content and sequence of the news based on its experiential history of each user. They do this to keep each individual reader attentive BUT this very same selection and sequencing process can be used for evil as well as altruistic purposes.
The recent stories about how a ‘student’ researcher created an app for Facebook which got 275k participants is a case of how things can be used for criminal purposes. In this case they grabbed all the friends of the 275k and netted about 50 million FB profiles, photos, timelines, friends, etc. Then, with funding from hedge fund billionaire Mercer and led by Steve Bannon, they worked those 50M records into a series of 6k psychographic data points from which they were able to individually target not only those 50M FB people but other FB users that share many of those data points. FB stock is down 7% today and British and US governments are launching investigations with the biggest question: was the Russian operation privy to those 6k psychographic data points?
Misuse of data is likely more of a problem that biased data (which most-often is done for marketing purposes); not illegal election-changing purposes.
Tom Curl says
The biggest part of the problem is the humans. Many of us realize there are ways with drugs, VR, hypnosis, and now AI bots to manipulate our minds and even change our memories of events. But most of us do not believe this could happen to us personally. We are too smart for that kind of stuff.
The problem, unfortunately, is there is no way for us to know on an individual basis if our minds have been manipulated.
William K. says
While bias based on any number of things is definitely a real danger with AI, it is not the worst one. The show-stopping flaw is that AI only learns from the past, and lacks the ability to evaluate an entirely new situation. The presumption that a particular response is correct because it was correct the last dozen times will not help if some other variable that never changed before is changed. Consider seeing one pedestrian waiting to step off the curb,that never stepped off the curb before, but this time does, and the AI program did not anticipate it while an attentive human would have observed the obvious focus that a computer never knows how to perceive. Just one example, backed up by the recent Uber fatality. A human could have chosen to change lanes just in case, while the computer never would.
William K. says
Of course, my assertion assumes a non-distracted human driver. Thus what is really needed is serious legislation to remove a bunch of those very profitable distractions. But, just like the cigarette problem, it will be many years before any change can be made, since there is so much profit in providing those distractions.
Joanna Bryson says
I literally wrote the Science paper on this topic, but this is NOT the biggest threat of AI. The biggest threat of AI is its deliberate misuse, the deliberate construction or repurposing of AI to do antisocial things. Like change democratic election outcomes to undermine the country’s own institutions, destabilising and making it less able to constrain the bad actors that would do such a thing.
William K. says
Good Point, J.B! but now there is the fundamental flaw of the AI system not being able to sense that something might happen, or is probably going to happen, and correct for that condition to avoid the unfortunate event. The second part of the flaw is that AI will never ever be capable of effecting a response not programmed in advance. The result will be that AI will often make wrong choices because it does not have the obvious correct choice in it’s list of options. And I don’t think that there is a solution for that.