Jason Lim is a Washington, D.C.-based expert on innovation, leadership and organizational culture.
Bradykinin hypothesis
By Jason Lim
I'll be the first to admit that I had never heard of the term, “the bradykinin hypothesis” before. Then again, I had never heard of “cytokine storms” either before the pandemic made it into a household term.
But I recently found a fascinating article recently written by Thomas Smith on Medium titled, “A Supercomputer Analyzed COVID-19 ― and an Interesting New Theory Has Emerged,” that told the story of how a supercomputer named Summit at Oak Ridge National Lab in Tennessee crunched a massive amount of data about COVID cases and came up with an insight about how the virus causes seemingly unrelated and random symptoms from brain damage to blue toes.
He writes: “When Summit was done, researchers analyzed the results. It was, in the words of Dr. Daniel Jacobson, lead researcher and chief scientist for computational systems biology at Oak Ridge, a 'eureka moment.' The computer had revealed a new theory about how COVID-19 impacts the body: the bradykinin hypothesis. The hypothesis provides a model that explains many aspects of COVID-19, including some of its most bizarre symptoms. It also suggests 10-plus potential treatments, many of which are already FDA approved. Jacobson's group published their results in a paper in the journal eLife in early July.”
OK, so I understand that a supercomputer is not artificial intelligence. But the point I am trying to make was that this is the first time in my lifetime that technology didn't just aid in doing what medical professionals do better and faster, but, in fact, came up with insight about a pathological condition that might have escaped human doctors altogether for very long time. In this case, it can be argued that technology wasn't just “complementary” or “augmentative” to human judgment or skills; it supplanted human judgment.
I remember touring the IBM Watson headquarters in New York City several years back where they showcased the different ways that Watson can serve in the medical field. Chief among them was how Watson could be fed millions of scans of MRI images of cancers and taught to read the images much better than a human could. Or, how Watson could connect the dots of seemingly unrelated symptoms to diagnose the Kawasaki disease that had escaped doctors' notice. But the IBM reps who showed us the capabilities were very careful to label Watson as a complementary piece to what the doctors do, careful not to usurp the primacy of human skill and cognition.
But how long can that primacy hold?
There have been dire warnings for the past few years about how AI is going to take over jobs across industries. The AI revolution is going to be more than the mere automation that threatened blue collar jobs. This time, it is also targeting the white-collar work of lawyers, financial analysts, accountants, etc. This is understandable. Consuming large amounts of information, finding patterns, and then correlating the current situation to historical data and patterns to interpret that information ― which is what many white-collar jobs consist of ― can definitely be done better, cheaper, and faster by AI.
However, the medical field seemed safe from such a takeover. Who in their right mind would trust a machine over the judgment of a medical doctor who has seen thousands of cases, treated hundreds of patients, and seen everything from minor cuts to horrendous injuries of all types? Who would know better why you are feeling fatigued all the time, why you can't breathe right, or why you suddenly have pains wracking your body? Who but a trained doctor?
Well, Summit apparently does. And it almost seems an inevitability that AI will be pretty much better at everything than a human being. Apparently, AI can now teach itself and, even, reproduce itself. What then? What happens when it becomes impossible to not to admit that we are second to AI?
I don't pose this question as an apocalyptic koan. Rather, I pose it as a human system integration problem. How do we design these sociotechnical systems so that we can maximize the utility that they provide while not killing ourselves with ennui, lack of ingenuity and disengagement? How can we stay motivated and productive when machines can do everything for us better than we can do for ourselves?
I consider the eventual AI primacy as a human system integration problem because it really goes to the heart of designing in human agency ― or, more cynically, the perception of the primacy of human judgment ― in a world where human judgment is suspect compared to that of AI. How do we ensure meaning in our lives and feel as if we are still the ones ultimately in charge when we have become obsolescent? How do we maintain a sense of purpose and autonomy when we are no longer productive?
Or, going back to the bradykinin hypothesis, maybe that event is where this “humanity” show has jumped the shark, and we are all just waiting for cancellation; hopefully, we will get picked up as a late-night rerun.
Jason Lim (jasonlim@msn.com) is a Washington, D.C.-based expert on innovation, leadership and organizational culture.