Author Archives: Candace Worley

Why AI Innovation Must Reflect Our Values in Its Infancy

In my last blog, I explained that while AI possesses the mechanics of humanness, we need to train the technology to make the leap from mimicking humanness with logic, rational and analytics to emulating humanness with common sense. If we evolve AI to make this leap the impact will be monumental, but it will require our global community to take a more disciplined approach to pervasive AI proliferation. Historically, our enthusiasm for and consumption of new technology has outpaced society’s ability to evolve legal, political, social, and ethical norms.

I spend most of my time thinking about AI in the context of how it will change the way we live. How it will change the way we interact, impact our social systems, and influence our morality.  These technologies will permeate society and the ubiquity of their usage in the future will have far reaching implications. We are already seeing evidence of how it changes how we live and interact with the world around us.

Think Google. It excites our curiosity and puts information at our fingertips. What is tripe – should I order it off the menu? Why do some frogs squirt blood from their eyes? What does exculpatory mean?

AI is weaving the digital world into the fabric of our lives and making information instantaneously available with our fingertips.

AI-enabled technology is also capable of anticipating our needs. Think Alexa. As a security professional I am a hold out on this technology but the allure of it is indisputable. It makes the digital world accessible with a voice command. It understands more than we may want it to – Did someone tell Alexa to order coffee pods and toilet tissue and if not – how did Alexa know to order toilet tissue? Maybe somethings I just don’t want to know.

I also find it a bit creepy when my phone assumes (and gets it right) that I am going straight home from the grocery store letting me know, unsolicited, that it will take 28 minutes with traffic. How does it know I am going home? I could be going to the gym. It’s annoying that it knows I have no intention of working out. A human would at least have the decency to give me the travel time to both, allowing me to maintain the illusion that the gym was an equal possibility.

On a more serious note, AI-enabled technology will also impact our social, political and legal systems. As we incorporate it into more products and systems, issues related to privacy, morality and ethics will need to be addressed.

These questions are being asked now, but in anticipation of AI becoming embedded in everything we interact with it is critical that we begin to evolve our societal structures to address both the opportunities and the threats that will come with it.

The opportunities associated with AI are exciting.  AI shows incredible promise in the medical world. It is already being used in some areas. There are already tools in use that leverage machine learning to help doctors identify disease related patterns in imaging. Research is under way using AI to help deal with cancer.

For example, in May 2018, The Guardian reported that skin cancer research using a convolutional neural network (CNN – based on AI) detected skin cancer 95% of the time compared to human dermatologists who detected it 86.6% of the time. Additionally, facial recognition in concert with AI may someday be commonplace in diagnosing rare genetic disorders, that today, may take months or years to diagnose.

But what happens when the diagnosis made by a machine is wrong? Who is liable legally? Do AI-based medical devices also need malpractice insurance?

The same types of questions arise with autonomous vehicles. Today it is always assumed a human is behind the wheel in control of the vehicle. Our laws are predicated on this assumption.

How must laws change to account for vehicles that do not have a human driver? Who is liable? How does our road system and infrastructure need to change?

The recent Uber accident case in Arizona determined that Uber was not liable for the death of a pedestrian killed by one of its autonomous vehicles. However, the safety driver who was watching TV rather than the road, may be charged with manslaughter. How does this change when the car’s occupants are no longer safety drivers but simply passengers in fully autonomous vehicles. How will laws need to evolve at that point for cars and other types of AI-based “active and unaided” technology?

There are also risks to be considered in adopting pervasive AI. Legal and political safeguards need to be considered, either in the form of global guidelines or laws. Machines do not have a moral compass. Given that the definition of morality may differ depending on where you live, it will be extremely difficult to train morality into AI models.

Today most AI models lack the ability to determine right from wrong, ill intent from good intent, morally acceptable outcomes from morally irreprehensible outcomes. AI does not understand if the person asking the questions, providing it data or giving it direction has malicious intent.

We may find ourselves on a moral precipice with AI. The safeguards or laws I mention above need to be considered before AI becomes more ubiquitous than it already is.  AI will enable human kind to move forward in ways previously unimagined. It will also provide a powerful conduit through which humankind’s greatest shortcomings may be amplified.

The implications of technology that can profile entire segments of a population with little effort is disconcerting in a world where genocide has been a tragic reality, where civil obedience is coerced using social media, and where trust is undermined by those that use mis-information to sew political and societal discontent.

There is no doubt that AI will make this a better world. It gives us hope on so many fronts where technological impasses have impeded progress. Science may advance more rapidly, medical research progress beyond current roadblocks and daunting societal challenges around transportation and energy conservation may be solved.  It is another tool in our technological arsenal and the odds are overwhelmingly in favor of it improving the global human condition.

But realizing its advantages while mitigating its risks will require commitment and hard work from many conscientious minds from different quarters of our society. We as the technology community have an obligation to engage key stakeholders across the legal, political, social and scientific community to ensure that as a society we define the moral guardrails for AI before it becomes capable of defining them, for or in spite of, us.

Like all technology before it, AI’s social impacts must be anticipated and balanced against the values we hold dear.  Like parents raising a child, we need to establish and insist that the technology reflect our values now while its growth is still in its infancy.

The post Why AI Innovation Must Reflect Our Values in Its Infancy appeared first on McAfee Blogs.

I am an AI Neophyte

I am an Artificial Intelligence (AI) neophyte. I’m not a data scientist or a computer scientist or even a mathematician. But I am fascinated by AI’s possibilities, enamored with its promise and at times terrified of its potential consequences.

I have the good fortune to work in the company of amazing data scientists that seek to harness AI’s possibilities. I wonder at their ability to make artificial intelligence systems “almost” human. And I use that term very intentionally.

I mean “almost” human, for to date, AI systems lack the fundamentals of humanness. They possess the mechanics of humanness, qualities like logic, rationale, and analytics, but that is far from what makes us human. Their most human trait is one we prefer they not inherit –  a propensity to perpetuate bias.  To be human is to have consciousness. To be sentient. To have common sense. And to be able to use these qualities and the life experience that informs them to interpret successfully not just the black and white of our world but the millions of shades of grey.

While data scientists are grappling with many technical challenges associated with AI there are a couple I find particularly interesting. The first is bias and the second is lack of common sense.

AI’s propensity to bias is a monster of our own making. Since AI is largely a slave to the data it is given to learn from, its outputs will reflect all aspects of that data, bias included. We have already seen situations where applications leveraging AI have perpetuated human bias unintentionally but with disturbing consequences.

For example, many states have started to use risk assessment tools that leverage AI to predict probable rates of recidivism for criminal defendants. These tools produce a score that is then used by a judge for determining a defendant’s sentencing. The problem is not the tool itself but the data that is used to train it. There is evidence that there has historically been significant racial bias in our judicial systems, so when that data is used to train AI, the resulting output is equally biased.

A report by ProPublica in 2016 found that algorithmic assessment tools are likely to falsely flag African American defendants as future criminals at nearly twice the rate as white defendants*. For any of you who saw the Tom Cruise movie, Minority Report, it is disturbing to consider the similarities between the fictional technology used in the movie to predict future criminal behavior and this real life application of AI.

The second challenge is how to train artificial intelligence to be as good at interpreting nuance as humans are. It is straight forward to train AI how to do something like identifying an image as a Hippopotamus. You provide it with hundreds or thousands of images or descriptions of a hippo and eventually it gets it right most if not all the time.

The accuracy percentage is likely to go down for things that are perhaps more difficult to distinguish—such as a picture of a field of sheep versus a picture of popcorn on a green blanket—but  with enough training even this is a challenge that can be overcome.

The interesting thing is that the challenge is not limited to things that lack distinguishing characteristics. In fact, the things that are so obvious that they never get stated or documented, can be equally difficult for AI to process.

For example, we humans know that a hippopotamus cannot ride a bicycle. We inherently know that if someone says “Jimmy played with his boat in the swimming pool” that, except in very rare instances likely involving eccentric billionaires, the boat was a toy boat and not a full-size catamaran.

No one told us these things – it’s just common sense. The common sense aspects of interpreting these situations could be lost on AI. The technology also lacks the ability to infer emotion or intent from data. If we see someone buying flowers we can mentally infer why – a romantic dinner or somebody’s in the doghouse. We can not only guess why they are buying flowers, but when I say somebody’s in the dog house you know exactly what I mean. It’s not that they are literally in the dog house, but someone did something stupid and the flowers are an attempt at atonement.

That leap is too big for AI today. When you add to the mix cultural differences it exponentially increases the complexity. If a British person says put something in the boot it is likely going to be groceries. If it is an American it will likely be a foot. Teaching AI common sense is a difficult task and one that will take significant research and effort on the part of experts in the field.

But the leap from logic, rationale and analytics to common sense is a leap we need AI to make for it to truly become the tool we need it to be, in cybersecurity and in every other field of human endeavor.

In my next blog, I’ll discuss the importance of ensuring that this profoundly impactful technology reflects our human values in its infancy, before it starts influencing and shaping them itself.

*ProPublica, Machine Bias, May 23, 2016

The post I am an AI Neophyte appeared first on McAfee Blogs.