The AI Snake Oil newsletter has rebranded itself as the AI as Normal Technology newsletter. I have many thoughts.
I don’t really want to get deep into my disagreements with Arvind and Sayash, which are numerous. I don’t think there’s any topic on which our two newsletters see eye to eye. I often find myself subtweeting them on here.1 But this rebranding is over the top, cynically riding a wave of institutionally acceptable critihype.
Let me be clear. I don’t think AI is snake oil OR normal technology. These are ridiculous extremes that are palatable for clicks, but don’t engage with the complexity and weirdness of computing and the persistent undercurrent promising artificial intelligence. But here’s what Sayash wrote on Twitter:
“AI Snake Oil was an attempt to understand AI's present and near-term impacts. But since releasing the AI as Normal Technology essay, we have been thinking about its future impacts. The name change reflects this shift.”
This got me wondering: what are examples of technology that started off as snake oil and then became normal? I asked this on Bluesky, and all of the answers were deeply troubling.
The most common example in the replies was cryptocurrency. Which, yes. Cryptocurrency emerged as an unprecedented platform for illicit purchases, scams, and Ponzi schemes. All of Silicon Valley tried to sell us a bag of goods about how Web3 would change computing and society. Do you guys remember NFTs? Some Bored Apes replied to me on Twitter yesterday! Zeke Faux’s chronicle of cryptocurrency Number Go Up is filled with harrowing tales, including a dark web of literal slaves running pig butchering scams in Cambodia. And yet, now all of these institutional investors are knee-deep in crypto. Matt Levine wrote about how even your Vanguard funds are probably investing in crypto. And the US Government just passed an insane act backing stablecoins. Cryptocurrency is now a normal technology, and that doesn’t seem to be a good thing.2
An example that’s been on my mind a lot is multilevel marketing (MLM). Remember the Mary Kay lady? Bridget Read’s Little Bosses Everywhere is an essential history of MLM in America, chronicling how get-rich-quick schemes shilling vitamin supplements evolved into cult-like pyramid schemes where salespeople only made a profit by convincing other people to become salespeople and report to them as bosses. The industry is built on conning people into conning other people. And yet, through various legal maneuvers, some of these companies became legitimate businesses. Laws were made to distinguish MLMs from pyramid schemes. The legal definition seems to be that Mary Kay and Amway are legal, and most everything else is a scam. Amway and its founding DeVos family are famously politically active, funding right-wing causes and think tanks like the Heritage Foundation. Amway is normal insofar as its family has billions of dollars, owns a sports team, and has held positions high in government. And yet no one knows what products Amway actually sells.
The third example, sent by Dan Davies, is advertising. This might be the most disturbing example. Dan recommended The Mirror Makers by Stephen Fox, which chronicles the history of advertising. I have only skimmed the first chapter, and I can already tell I need to read this book. Fox describes the origins of advertising in the 19th century, how it was universally hated and considered scammy, and how it gained its initial momentum through shilling “patent medicines,” potions claiming to be miraculous curatives. That’s right: The advertising industry started off by selling snake oil, and now it’s the backbone of the global economy. It’s a multi-trillion-dollar industry. There is no tech oligarchy without ads. Despite their protests to the contrary, Meta and Alphabet are advertising companies. Advertising is deeply normal, but in the derogatory sense that Max Read sees AI logically continuing.
Max’s derogatory normality was focused on generative AI, describing how it is evolving into technology that profits on engagement. But I think the concept can be broadly applied. Cryptocurrency, MLMs, and advertising are also only normal in the derogatory sense. It’s a bummer. No one was able to provide a positive example of a technology once considered snake oil became generally liked after its widespread adoption. Maybe such technology can’t exist. Maybe it’s easy to spot snake oil from the get go, but our institutions aren’t strong enough to prevent all scams from becoming institutions themselves. If you have a positive example, tell me in the comments! I want a counterpoint to this depressing conclusion.
Derogatory normality is quite depressing. Arvind and Sayash love their normal technology frame because they think we can fix things with normal governance. They believe that AI can be regulated like any other technology. However, the examples above demonstrate that regulation often enables scams, rather than stopping them. Their Normal Technology essay only engages with the strawman arguments of rationalist doomers (who love the attention and write bullshit like this). They don’t engage with the present harms of the technology, which has seen unprecedented widespread adoption in unexpected areas like personified personal advice agents, automatic coursework completion, and composition of doctors’ notes. If chatGPT is normal technology, it’s not normal like a toaster or a light bulb. But I hope it doesn’t end up being normal like cryptocurrency or MLMs.
See if you can spot the subtweet in Monday’s post.
Fascinatingly, Arvind of the AI is Normal Technology blog is a huge fan of cryptocurrency, and has written a textbook and popular Coursera class on the topic. Make of this what you will.
Weather forecasting has slowly and steadily evolved from mostly snake oil to a generally positive and helpful technology. Having physics and frequent reality checks has greatly assisted in moving the field in the direction toward more effective methods.
Hi, Ben! This part caught my attention:
> Let me be clear. I don’t think AI is snake oil OR normal technology. These are ridiculous extremes that are palatable for clicks, but don’t engage with the complexity and weirdness of computing and the persistent undercurrent promising artificial intelligence.
I realized that, while I've read a bunch of your texts, I'm not sure *what* do you think AI is. Or, to be more precise: how the field will progress, which kind of impact should we expect, and so on. In your recent participation in the Increments podcast you say something like 'the doomers have a very clean narrative of what's going to happen' (or something like this, pardon me for the misquote) and I don't find something similar for your position. In fact, there seems to be very few coherent counter-narratives to the doomer's one and, in that sense, Arvind is positioning himself to be this, whether correctly or no, it remains to be seen. So, while writing a point by point rebuttal of their view could be silly, a broad overview of your assumptions and some general predictions would be very interesting.