Splice Today and Twitter/XI attacked the great philosopher Daniel Dennett last year for an article he wrote in The Atlantic, in which he argued that the use of AI to create fake people should be severely punished. Dennett called fake people “the most dangerous relic in the history of humanity.”
Creating fake digital humans risks destroying our civilization. Democracy relies on the informed (not misinformed) consent of the governed. By allowing the most economically and politically powerful people, corporations and governments to control our attention, these systems would rule us. By distracting and confusing us and exploiting our most uncontrollable fears and insecurities, fake humans will lead us into temptation and from there into consenting to their own domination. They will convince us to adopt policies and beliefs that make us even more susceptible to manipulation. Or we will simply be distracted and become passive, ignorant pawns. This is a frightening prospect.
This is a shockingly, shockingly awful story. The counterfeiters are trying to talk us into it. But it’s also shockingly, shockingly vague. Dennett argued that the counterfeiters should get life in prison. Some kind of severe deterrent. It’s unclear what exactly he was proposing to punish. Was he referring to ChatGPT? Or what?
Dennett died in April at age 82, and by the time he took the next step, he was truly a philosopher of the first order, having written many classic books and essays marked by brilliant prose, daringly counterintuitive positions, and argumentative ingenuity, especially when it came to the question of the relationship between consciousness and the world. I never met him, but I have read his work and admired him from afar all my life, even when we disagreed.
But I thought the Atlantic essay was odd coming from a thinker of that calibre. Maybe he was ill. If so, I didn’t know it. He never said what he meant by “fake humans,” just that they were the biggest threat humanity has ever faced. He repeated the word “horrible” over and over to back up his claim. He seemed to be thinking about some kind of criminal crackdown on people working with AI, but he didn’t say anything practical or accurate about it, just hysterical. I assumed he was egged on by Atlantic editors like Adrien LaFrance, who called Facebook a doomsday machine.
I’ve been very vocal in criticizing The Atlantic on these issues before, but I also criticized Dennett in X, which I found quite amusing, but also, quite frankly, was trying to get Dennett to do a definitional study of “false man”, which is the sort of thing you should expect from (what I claim to be) the world’s greatest philosopher.
I started Xing with this: “Martin Amis told me he was relieved he died before he was charged with creating fake people.” I was wondering if the fake persona was my cat (she acts like that all the time), or my Roomba, or Siri, or a video game character, or a search algorithm, or… Dang, narrow it down before you criminalize it. Then I said:
Those of you who know me know that I have a tendency to throw out the most provocative yet flippant statements I can think of, and sometimes (though rarely) I bravely fight back. On the whole, these statements rarely elicit a response as people feel it is beneath their integrity to react. But to my surprise, Philosopher No. 1 fought back.
Perhaps my tweets did not suggest that I wanted to engage in rational debate about human nature, which I do. But Dennett was not receptive. I was also a professor of philosophy at the time, and he sent me an email on May 26 to his academic address:
You seem to be trying to continue to attract attention by misrepresenting my views on the dangers of hypocrites. I would like you to consider that my lack of public response may be because I do not want to provoke you into further reckless behavior. You appear foolish and needy (in my humble opinion). If you further attack, you will only have yourself to blame. Take a deep breath and reconsider. Surely there is a better way to gain attention and respect. If you follow this kind advice, I look forward to ignoring you and never mentioning you again.
Best of luck,
Dan Dennett
P.S.: Your paper was written together with my late good friend Richard Rorty, and I think he would agree with my suggestions.
I told people that Dennett was “threatening” me, which was probably a poor choice of words (though he definitely Googled me). He was threatening to call my dean or to trash-tweet me to his 300,000 followers while I replied to my 3,000. It had a “you’ll never work in this town again” vibe. I especially liked how bold and approachable it was, and was surprised by the tone. It sounded like it was straight out of 1890. I never thought I was being reckless, but in this case I was proud of having been reckless.
Meanwhile, he repeatedly said that I had misrepresented his views, but never said how specifically. As far as I can see, as far as he can tell, I had not. I also want to point out now that if there was a better way to get attention and respect, I was unaware of it. And we could always discuss Rorty later, I thought. Unfortunately, time was running out.
Anyway, I continued to screenshot the above email and upload it to X, which resulted in the following email on June 1, 2023:
Sharing my private emails on Twitter without my permission shows that you have a strange sense of honor and politeness. I would never do that. As I said in my email, I took the trouble to find out who you are in order to personally advise you. I thought you were making a mistake. You mistook this for a threat. I promised you that I would do nothing if you followed my advice. You misunderstood that too. You live in a strange world. I am much more interested in the issue of AI dangers than you are. I always try to make my ideas understood, but sometimes I fail.
Once again, I wish you the best of luck.
Dan Dennett
Meanwhile, in X, he said:
Dennett repeatedly speculated that I was a chatbot, an insult so 2022-ish that it was out of place for me to respond. But Dennett also claimed that chatbots are the most dangerous objects in human history. It’s kind of cool that he thought that about me!
In any case, this whole furor died down, just as any Twitter furor or moral panic eventually dies down, and was probably not that big of a furor overall. Also, it says a lot about Dennett in some ways that he was made to respond in the last year of his life: Adrien LaFrance, for example, still ignores not only my insults but also my arguments.
But I can’t help but think that Dennett’s reaction is also emblematic of the academia I just retired from. There is a clear hierarchy based on prestige, not knowledge. Institutions are ranked from 1 to 1000, and professors are ranked from 1 (Dennett) to 1 million (me). It makes no sense for a 1 to have a substantive reaction to a 1 million; we’d be better off squashing him like a bug. At times it seems like an inverted meritocracy.
Unfortunately, people like Dan Dennett may go unchallenged for endorsing a series of ridiculous but deadly positions. Because positions you have held in life, or up to a certain point, can be refined or abandoned. But positions you hold until you die, terminal positions, so to speak, are likely to be unchangeable. And so the greatest philosopher of mind of our time left us with the idea that we should start wiping out chatbots.
—Follow Crispin Sartwell on X: @CrispinSartwell
