I Am Not Sam Bankman-Fried

by David Roth

It is some consolation that the photo of Sam Bankman-Fried that Google has listed as the top search result for my name for months is not nearly the worst available photo of Sam Bankman-Fried. There are some really rough ones, there—ones where he looks like the lost libertarian member of the Peanuts gang, ones in which he is making the sort of faces people only make when they are being photographed against their will outside a courthouse, a great many in which I found my brain somehow rounding whatever he was actually wearing up into a set of jumbo pajamas—but that’s not really the point. I am not Sam Bankman-Fried, is maybe the main issue I have with it. I might dislike the state of affairs somewhat less if I’d been confused with someone more photogenic than this particular prolific, infamous, and reliably underdressed financial criminal, but ideally for both Google’s purposes and mine this would be a picture of me. Which, again, I feel compelled to underline it is not.

It is possible to figure out how this happened in both the micro and the macro senses. The image ran atop a review I wrote of Michael Lewis’s book about Bankman-Fried for New York Magazine’s website; Google pulled it from Reddit, where it had populated automatically atop a thread of people talking about that story. Or, talking, anyway, about Bankman-Fried, and Lewis’s book, with that link and the photo that would wind up being made mine more or less incidental to it. There are other images alongside it in Google’s search results on my name—the portrait I’ve used as my headshot for the last five years more or less by default, a screengrab of me on a zoom call for a podcast, a somewhat more worrying one of me with a very wide/bad mustache that I had for less than five minutes taken while I was bothering my wife with the aforementioned mustache. None of these are really all that great, either—the portrait is good, but the ceiling for an image like this is where it is—but they are photos of me, and as such I can’t really argue with them appearing under my name.

The macro issue, though, is the more significant one. This is surely not an especially popular or frequent search on Google, as these things go—I’ve known about it for a while, and when Maria reached out to me with a “???” screengrab recently she became maybe the third or fourth person to do so. The bigger issue, of which this is a small and pretty funny part, is that Google is a lot wronger and stranger than it has ever been, and that it appears to be getting wronger seemingly more or less on principle. 

There is, to take an instance decently close to home here, my birthday, which Google lists as “1977 or 1978” which is both admirably open-ended and uncertain as AI-related bloopers go, but also confounding: I have never claimed to be a year older than I am, and can’t begin to imagine where I would have done that for Google to find in the first place. As with so many of the so many other things about the internet that have gotten worse, this seems to have begun as an Advanced Business Maneuver aimed at making the customer experience both leaner to produce and the maximum possible amount worse before it becomes entirely untenable. (In that case, the result would presumably involve my first image being the stoic image of MLB reliever Doug Jones that I have used as an avatar in the past, and a date of birth listed only as “old lol,” and ideally also the attribution of Jones’s impressive career saves total to me.) 

That Google’s search product has made oafish and vigorous use of AI as part of this campaign of making its core product worse has gotten a good deal of attention—I wrote about it at Defector, which may someday lead some future, even-worse version of the product to attribute this strategic gambit to me—primarily because of the slapstick, “you can eat a few small rocks per day, as a treat” results.

That’s understandable, although the central premise—that this product should be worse, due to business considerations far too high-level and sophisticated for the average user to comprehend—seems more like something that would be used to justify the deployment of a trendy, unhelpful, and rapidly worsening technology like AI than like that technology’s fault.

Someone—crucially, in Google’s case, not a programmer but a person with a MBA who gets paid the same sort of salary as a member of the All-NBA Second Team—made this decision, and other people with similar credentials and salaries agreed that it seemed like a good idea. They availed themselves of a technology that people in their cohort and seemingly only in that cohort find intriguing, and enlisted that janky technology to that witless and useless purpose. An essential thing got worse, to no obvious end, and it compounded and compounded as such things tend to do, and then at some point, sometime later you look in the mirror and don’t even recognize what’s looking back at you.