The Downfall of One of the World’s Biggest Brains

Listen to this episode

S1: Now, the last clue, even a broken one of these on your wall is right twice a day. Watson, what is clock clock is correct and with that, you move up the.

S2: 2000, if you know Watson at all, it might be from this jeopardy. In 2011, when Watson, IBM’s computer system played against champions Ken Jennings and Brad Rutter and won.

S1: Well, we just seen history made here on KXAN this afternoon, Watson, the supercomputer wiping the floor with the two greatest human champions ever to play Jeopardy.

Advertisement

S2: Watson Watson’s time on Jeopardy was fun viewing, but it was also a very savvy public debut of a product that IBM wanted to sell.

S3: Watson Health just on the back of that performance. In that feat, it gained a ton of interest and attraction from health care systems and others who are looking to use that technology.

S2: That’s Casey Ross, technology correspondent for Stat News, who has been covering Watson Health for years.

S3: It was supposed to really change health care in so many different ways, delivering insights to hospitals about their operations, to oncologists about their cancer care of the patient in front of them, to pharmaceutical companies about the development of drugs, how to match patients with clinical trials. There were just so many different domains.

Advertisement
Advertisement

S2: It sounded revolutionary. But as Casey’s reporting shows, it never really worked. And last week, Watson Health was essentially sold for parts. A private equity firm, Francisco Partners bought some of Watson’s data and analytics products for what Bloomberg News said was more than a billion dollars.

Advertisement
Advertisement
Advertisement

S3: It represents an ignominious end to the Watson Enterprise that the greater whole that it was supposed to be was never realized.

S2: I look at the amount of money that went into pulling this together, acquisition after acquisition. It was billions of dollars and it went for a billion in the end. I wonder if there’s any way to read that. As anything but a failure

S3: financially, certainly not. They spent way more money building this than they certainly got back. I mean, just the acquisitions alone cost them $5 billion so that it was sold so many years later after so much and effort 7000 employees at one point means that this was a total failure that they needed to just cut their losses and move on.

Advertisement

S2: Today on the show, how did Watson go from being the future of health care to being sold for scraps in a time when tech companies are looking to disrupt how we get our care? What does this say about the promise of AI for medicine? I’m Lizzie O’Leary and you’re listening to what next TBD show about technology power? How the future will be determined? Stick around. To understand why IBM or any tech company wants to be in the health data business, just think about going to the doctor to get some tests done. Maybe you need a sonogram or an X-ray. But then you have to see a specialist. And instead of seamless digital communication between the various providers, you find yourself carrying a CD of your own records around, from doctor to doctor as an individual. It’s incredibly frustrating as an industry wide phenomenon. It presents an opportunity ripe for tech’s favorite word disruption.

Advertisement
Advertisement

S3: I mean, there’s a tremendous amount of information that is collected every day on the care of hundreds of millions of people, and there is currently no way to connect to that information to link it to an individual across all the domains in which they get care. And then to develop a holistic picture of who they are of what their diseases are, of what the best treatments are, and how best to ensure that they get the best care at the lowest possible cost. There is no connectivity right now that can do that at scale. I mean, you have health systems who have six different electronic health record systems, none of which talk to each other, even within a single system, even if you got care at your local hospital academic medical center. Across decades and saw different physicians along the way, they wouldn’t necessarily have a really good picture of who you are or what care you’ve gotten. And that’s where I think the people who are in the technology sector to look at and say, what? You’ve got to be kidding me. This has to be fixed and we’re going to fix it.

Advertisement
Advertisement
Advertisement
Advertisement

S2: Yeah, Google, Microsoft, a lot of very big companies are extremely interested in health care. I mean, some things like Google Health have failed. But I wonder if you could lay out what is so attractive for these big tech companies about health care? Why do they want to be in this business?

S3: I mean, it’s the biggest. It’s one of the biggest parts of our economy. It’s a $3 trillion business that has legacy technology infrastructure that should be embarrassing. Tech companies are drawn to audacious challenges like this and ones where they can make if they’re successful. A ton of money. Health care, in that sense, is just a wide open, very attractive lane for them. But it’s it’s one that has proven very, very difficult for them to make headway in

Advertisement

S2: cases talking about how things are today. But the same problems have been around since the advent of digitized data, and IBM saw an opportunity. Well, let’s go back a little bit to 2012. IBM closed this deal with Memorial Sloan-Kettering, one of the preeminent cancer centers in the country, to train and AI to make treatment recommendations. What was the goal? What were they trying to do?

S3: The goal, as I remember it, from talking to the physicians at Memorial Sloan Kettering and folks at IBM at the time was they were really trying to democratize the expertise of Memorial Sloan Kettering Cancer Physicians of its own colleges and make that expertise available to patients all over the world. And to sort of develop this standardized engine for providing optimal treatment recommendations customized to a patient in front of a doctor, you know, thousands of miles away. I mean, it was a very it was a beautiful notion that you could do that.

Advertisement
Advertisement
Advertisement
Advertisement
Advertisement

S2: It’s really appealing.

S3: It’s really appealing. And they and they are trying to say, Well, let’s make it more objective. Let’s look at all of the data and let’s and let’s tell every physician for this patient in front of you. This is how they should be treated.

S2: So you get your, you know, biopsy results and things don’t look good, but you’re not just getting kind of the the expertise or the biases of your particular oncologist, you’re getting the wealth of of thousands of oncologists distilled into an algorithm.

S3: Yes, you are getting all of that data across so many different physicians crunched down into a very digestible format and recommendation that could then lead to the best treatment for that patient.

Advertisement

S2: Reading your reporting it, it sounds like this was incredibly important to IBM in 2015, Ginni Rometty, who was the CEO at the time, went on Charlie Rose. She said Health care was our moonshot.

S4: I’m telling you our moonshot will be the impact we will have on health care. It is already started. We will change into. Our part to change the face of health care. I am absolutely positive about it.

S2: And how much of IBM’s hopes were hung on this thing?

S3: The company made a huge bet here that this could be the bridge to a different kind of future for IBM, which at the time was several years of of quarterly revenue declines. And they were trying to use Watson as this bridge to a different future where IBM wasn’t sort of this old guard hardware company that everybody knew so well, but was operating on this cutting edge of artificial intelligence. And, you know, they referred to it as cognitive computing, and health care was the biggest, the buzziest use case. This was where they were going to really show the surpassing value of their technology.

Advertisement
Advertisement
Advertisement
Advertisement

S2: To do that, IBM needed massive amounts of data on which to train Watson. It got that data through acquisitions, eventually spending some $5 billion, buying a series of health data companies

S3: Truven Fatal Explorers and Merge. Truven had the biggest insurance claims database in the nation, with 300 million covered lives. I mean, tremendous data set. Explore has provided clinical data set of actual electronic health records kept by health systems representing about 50 million or so patients. Vital added on top of that, and Merge had a huge imaging database, so they had all this data and that the idea was you expose Watson to that and it finds patterns that physicians and anyone else can’t possibly find when looking at that data, given all the variables in it. So that was the idea,

Advertisement

S2: except that was not the reality. One of IBM’s high profile partnerships with MD Anderson Cancer Center in Texas fell apart. A doctor involved told Casey that there wasn’t enough data for the program to make good recommendations and that Watson had trouble with the complexity of patient files. The partnership was later audited and shelved IBM’s big idea that Watson would provide the best cancer care around the world. That just didn’t pan out.

S3: I mean, if you think about it like knowing what we know now or what we’ve learned through this, the notion that you’re going to take an artificial intelligence tool, expose it to data on patients who were cared for on the Upper East side of of Manhattan and then use that information. And the insights derived from it to treat patients in China is ridiculous. Most machine learning and artificial intelligence experts looking back at that, in hindsight, would say that’s a ridiculous idea because you need to have representative data and the data from New York is just not going to generalize to different kinds of patients all the way across the world.

Advertisement
Advertisement
Advertisement

S2: What was happening in a clinical setting? What was happening to patients,

S3: mainly our window through the reporting was talking to physicians. We got concerns from them that the recommendations that it was giving were just not relevant. Like maybe it would suggest a particular kind of treatment that wasn’t available in the locality in which it was making the recommendation or that it just the recommendation did not at all square with the the treatment protocols that were in use at the local institution or. And I think more commonly so, especially in the U.S. and Europe, you’re not telling me anything I don’t already know. I think that was the big credibility gap for physicians. It was sort of like, Well, duh. Hmm. Yeah, I know that that’s the chemotherapy I should pursue. I know that this treatment follows that one. You know, I’m very well aware based on my experience as a clinician that I should pursue these treatments and then try these and that these are the options for this kind of patient. You’re not telling me anything that I don’t already know.

S2: You got a hold of an internal IBM presentation from 2017, where a doctor at a hospital in Florida told the company, This product is a piece of shit.

S3: Yes. Seeing that written down in an internal document which was circulated among IBM executives was a shocking thing to see. And it really underscored the extent of the of the gap between what IBM was saying in public and what was happening behind the scenes.

S2: When we come back, what IBM was saying and selling in public.

S1: Watching, Annabel, your birthday is tomorrow. I’m telling something. What did you ask for a princess?

Advertisement
Advertisement
Advertisement
Advertisement

S2: There were a few years when if you turned on the TV, it felt like you were guaranteed to see an ad for Watson Health. I was too sick to have a look. There were celebrities, Bob Dylan.

S1: I’ve read all your lyrics, you’ve read all of my

S2: lyrics and heartwarming anecdotes, all about how Watson was helping doctors quote outthink cancer. You’re healthy. Are you

S1: a doctor? No, I help doctors identify cancer treatments. I want to be a doctor someday. I can help with that, too.

S2: Watson, I like.

S3: There were a lot of internal discussions, even a presentation that indicated that the technology was not as far along as they’d hoped, that it wasn’t able to accomplish what they set out to accomplish in cancer care. There were probably a lot of people that believed that truly did believe that they would get there or that it was closer than maybe some people realized. I think the marketing got way ahead of the capabilities.

S2: It’s very hard to listen to you and not think about Theranos, even though I know that this is not a one to one parallel in any way. I guess the question is when you are trying to move by leaps and bounds with technology in the health care sector, it just feels like a reminder that all things are not created equal, that that making big leaps with people’s health is a much riskier proposition.

S3: Yeah, I think that underscores the central kind of theme or point of this story that when you try to combine sort of the bravado of the tech culture and the notion that you can achieve these huge audacious goals, they called it a moonshot in a domain where you’re dealing with people’s lives and health in the most sacrosanct, you know, aspects of their existence in their bodies. You need to have evidence to back up that you can do what you say you can do.

Advertisement
Advertisement
Advertisement

S2: Why did they continue on trying to rescue this product that they seemed to know internally was failing?

S3: I think they had so much invested in it that it really was for them too big to fail. Had 7000 employees. They invested so much time and energy on in marketing. In the success of the product that they really needed it to succeed, and I don’t think the culture they’re tolerated, failure, you know, failure is sort of celebrated within technology companies, you know, fail fast, fail first, fail off and all of that. That’s not the case at IBM. It was losing revenue for twenty four consecutive quarters. They needed a win.

S2: Instead, they got to fail. But Watson’s fate certainly doesn’t mean that A.I. in health care is going away. Just last week, Microsoft, in a large group of hospitals, announced a coalition to develop AI solutions in health care. If you had to, I guess, pin down a moral to the story, is it that I and health care isn’t ready for prime time or that IBM did it wrong?

S3: I think it’s both of those. I think that, you know, this will be a case study for business schools for decades. I think when you look at what IBM did in the strategy, mistakes that they made and just the tactical errors that they made in pursuing this product, they they made a lot of unforced errors here. And and I think it’s also true that the technology, certainly the generation of technology that they had, was nowhere near ready to accomplish the things that they set out to accomplish and promised that they they could accomplish. I don’t think that the failure of Watson means that artificial intelligence isn’t ready to make significant improvements and changes in health care. I think it means the way that they approached it is a cautionary tale that lays out how not to do it.

Advertisement
Advertisement
Advertisement

S2: Is that something that can happen in a a for profit setting? I mean, I I had a conversation tangentially related to this with Timnit Gebru, who was at Google and very famously left. And she was very skeptical that if you have your A.I. funded by a company that wants to make money, that the profit motive is always going to interfere with with the science and the research.

S3: I think it is it is doable. I don’t think it’s automatically a product that’s developed in a for profit enterprise is doomed to failure. I think that the company would have to have a much longer view, though. You can’t you can’t develop a product like that with a quarterly financial target in mind. That’s never going to work.

S2: Does the failure of Watson Health make you worry that? It’s going to shut down other avenues for innovation that that, you know, such a spectacular belly flop. Might impede progress.

S3: I don’t think so. Like, when I think about it, there were so many mistakes that were made that were learned from that. If anything, it will facilitate faster learning and better decision making by other parties that are now poised to to to disrupt health care and make the progress that IBM failed to achieve. There’s a saying that, you know, pioneers often end up with arrows in their backs, and that’s what happened here. I mean, they’re an example. A spectacular example of of wrongheaded decision making and missteps that didn’t have to happen. And by learning from that, I think advancement and progress and true benefits will be faster coming.

S2: Casey Ross, thank you.

S3: Thank you.

S2: Casey Ross is a National Technology Correspondent for STAT. That’s our show for today. This episode was produced by Ethan Brooks. We are edited by Tori Bosch and Alison Benedict. And today is Alison’s last day after a decade of sleep. Alison, you are the best and we will miss you very much. Alicia Montgomery is the executive producer for Slate Podcasts. TBD is part of the larger What Next family, and it’s also part of Future Tense, a partnership of Slate, Arizona State University and New America. And I want to recommend that you listen to Tuesday’s episode of What Next? It is just a staggering story about how one Alabama town turned policing into a source for profit. I’m Lizzie O’Leary, thanks for listening.