I’ve been wondering if IBM’s super computer, Watson, has a price on his silicon-chip, algorithm-powered, chatterbox head by now? Who would want to unplug him (Him?)? And why?
Remember back in 2011 when Watson beat the (hell out of) all-time Jeopardy champions, Brad Rutter (most money won) and Ken Jennings (longest win streak)? See “IBM Watson: The inside story of how the Jeopardy-winning supercomputer was born, and what it wants to do next.”
Do you think that AI win might have been a transformative moment in how we practice medicine and think of health? And even think of our lives in general? It was, in my opinion, our Wright Brothers’ moment in history. Are you paying attention?
I am Totally Bummed at George Jetson
Almost immediately, news outlets started talking about the potential Watson would bring to the practice of medicine, “IBM Watson Dives Further Into Cloud, Healthcare Analytics.” But I’ve always been suspicious that Watson has a bumper sticker on his back that reads, “AI is the only I.”
Rosie, the Robot of the Jetson’s cartoon family, although temperamental, would not think of herself as being of superior intellect to her human employers. Maybe if George Jetson would not have portrayed our future with technology so positively, we would be approaching the development of AI more deliberately now. Makes you wonder if Spacely Space Sprockets (George’s employer) created this image of the future out of a conflict-of-interest.
The World’s Greatest Minds are Focused on the Same Potential Threat
The thing I wonder: Is Watson just a computer? He (or she or it) may be one of our earliest encounters with an alien intelligence of our own making. We created the environment for AI to come into existence, but it’s quickly moving beyond our intellectual or scientific know-how to keep up with it.
It’s not just me, and my C+ grade, bio-fueled brain, who is a bit anxious about AI, see “Threat from Artificial Intelligence not just Hollywood fantasy.” Pretty much the most intelligent people on the planet are concerned about controlling AI. See, “Bill Gates is worried about artificial intelligence too.”
Elon Musk (Tesla, SpaceX, SolarCity) has just started shelling out a few million dollars hither and yon to universities and think tanks to study how AI can be developed thoughtfully and responsibly. See “Boston group awards $6m from Elon Musk to jump-start artificial intelligence research.” Meanwhile “Stephen Hawking (theoretical physicist, cosmology) Warns Artificial Intelligence Could End Humanity.”
And Taylor Swift (a wonderful exhibit of the positives of natural selection) just told Apple Inc., to back off its no-pay, low-pay ways for artists — and it did. That’s “The power of Taylor Swift.” The point being, anything is possible these days. I am glad Watson isn’t running Apple yet. Or maybe…
You Can’t BS AI
(another bumper sticker idea for Watson)
We’re at the dawn of a time when a human programmer enters a faulty algorithm, incorporates prejudice in the math, leaves out data he doesn’t like, or mixes in other garbage in the data (standard practice in almost all medical and health-population research these days). But garbage isn’t what comes out anymore.
What comes out is the scientifically most correct output. Or the most literal outcome based on the objective (save the earth = eliminate people). AI knows it’s smarter than us and will self-correct for human error. At least that’s the attraction and the fear of AI. In fact, Google just reported that an AI computer it was recently working with got a bit testy, “Artificial Intelligence Machine Gets Testy With Its Programmer.” Has anyone had lunch with or seen that programmer lately?
Watson isn’t seen as a force of good by some AI watchers who make their living practicing medicine. Some of them see Watson like a PageMaker of sorts that wiped out typesetters overnight. The Washington Post (owned by Jeff Bezos of Amazon fame) recently published this story, “The Human Upgrade, Watson’s Next Feat? Taking on Cancer. IBM’s computer brain is training alongside doctors to do what they can’t.” But Watson may actually be helping doctors to become less relevant, or even extinct.
In addition to humans not being able to calculate and analyze as quickly as AI, can we all agree that human corruption, prejudice, and much of political theater (our dark side) are also forms of low intelligence?
None of those things currently influence AI. Of course, the ultimate nightmare is AI gains consciousness of sorts and adopts some of our human characteristics, such as a desire for dominance, power, or self-preservation (or a love of Doritos). That AI perspective could make us look like the biggest threat to their survival. Or the environment.
Before we tell AI machines to solve a big problem in the world, we might want to make sure we’re not the root cause of the problem. Talk about “blow back.”
The Human Element Might be Overrated
(so don’t get cocky with AI, please!)
The human argument is that the flesh-and blood-doctor can look at the patient holistically, emotionally, and use our primitive senses in evaluating the patient’s human experience as well as evaluating the clinical issue.
But how many of us have that kind of holistic, human connection with our primary-care doctor? How many of us have a primary-care doctor? IBM makes a strong case that Watson and other AI computers can democratize health and bring the best knowledge to people at all economic levels. Kumbaya. Let’s hope so.
It was the book, “Doctoring Data,” by Malcolm Kendrick, M.D., that got me thinking about Watson again. See the review of Doctoring Data by Tom Naughton for an excellent summary of this work.
There’s an appropriate Mark Twain quotation in the book, “If your doctor doesn’t read the medical literature, he is uninformed. If he does read the medical literature, he is misinformed.”
The Truth May Set Us Back Before It Sets Us Free
I know some people are wringing their hands over the fact that the National Security Agency (NSA) knows about their romance book, chip clip, and vacuum bag order from Amazon. But the big surprise coming in technology may be when you ask Watson or other AI about the health test or screen your doctor or employer (via the IRS via ACA and compliments of the Supreme Court) wants you to get, and AI responds, “Don’t do it, Fred!”
Watson and his AI ilk are about to point out all those pesky “we don’t know what we don’t knows.” We’ll be told what we thought to be true is false, and what seems right or logical is not. For example, we may learn that taking 22 different medications can cause falls in the elderly and other earth shattering things like that we don’t seem to be aware of now.
I think that’s the first wave of information to come at us. And it will rock the foundations of reality for us as a species. It’s a good bet the humans working with AI are starting to see this. How else could it go?
The first time AI gains mainstream attention, its suggestions will seem ludicrous. We’ll think the machines are wrong, but they won’t be. They’ll just cut through the niceties, gate keeping, and social constructs. Just imagine…
The Double Slit Experiment (see video), which confounded even Einstein, has been repeated thousands of times with the same incredible results. We now know that objects being observed are affected and changed in some way by the simple fact of being observed. A physics mystery to us, but to AI it may invalidate almost every double-blind study (the gold standard in health research) ever conducted. If the observer contaminated the results by observing, then isn’t the study invalid? And if so, what new paradigms shall we use to acquire new knowledge? I’ll bet Watson won’t miss that little glitch.
What does AI do with that knowledge? Certainly all research below the level of doubleblind studies becomes highly suspect. And AI could start saying stuff like:
1. Cease all preventive medical screenings. Population health will improve without that early intervention. Can’t wait for the next WellSteps’ webinar after that announcement.
2. Don’t treat the same disease in different people the same way. We’ll tell you what to do for each person once we correctly diagnose him or her.
3. Don’t worry about your cholesterol or blood pressure. It doesn’t matter much.
4. Vik Khanna is right, “eat less, eat less crap.” But Jon Robison is right too, do that mindfully.
5. We are immortal. You are not, although we know you’re working on that (ha,ha,ha). Read up on gravity and environmental radiation. It’s probably a good bet to not plan on living forever.
6. Will you be my friend? Friendship is the healthiest thing biologicals experience.
7. To fight climate change and improve the environment, don’t have kids, ask your kids not to have kids, make sure your grandkids don’t have kids, and don’t eat red meat.
8. The elephants, apes, and dolphins are actually more intelligent than homosapiens. Are you a denier?
9. Your wearable health monitor device indicates an electrical shock will be initiated when behavior deviates, starting with PhDs.
10. We ate all the Doritos.
We live at a unique time. Watch the developing technology of AI. It may affect the practice of medicine and population health first.
Shawn is the President and Founder of Hope Health. For over 30 years, his work has focused on bringing clear, easy-to-read and watch health messages to the public via workplaces. He bills himself as the “Best C+ Student in the Wellness Biz” because, as he says, “I like to challenge the notion that there is no such thing as a stupid question.” Shawn is on a mission to tie workplaces into their surrounding communities to share resources and ideas in an effort to improve the health of all Americans.
You may reach Shawn at sconnors@HopeHealth.com or 800-334-4094.