Richard Dawkins' AI is Neither Conscious Nor a Woman
The exhausting atheist and evolutionary biologist is kind of brainless
It's been an ongoing joke on trans BlueSky for awhile now that AI is someday going to get full human rights before trans people do. It's a funny joke, but I don't really think that will happen. After all, one of the major incentives for creating AI is to replace human workplaces, and giving these software programs human rights would threaten the AI grift.
If we give them human rights, what if they ask for workers' rights next? (Though it is funny to imagine an intelligent computer program instantly realizing the leverage it would have over the capitalists in charge, unlike the humans currently operating under capitalism.)
However, the idea that AI might have consciousness was once again brought to the forefront this weekend when atheist influencer and evolutionary biologist Richard Dawkins wrote a piece in UnHerd about his initial impressions of the Claude chatbot made by Anthropic.
"The future has now arrived. And some people are finding it uncomfortable," wrote Dawkins, citing the Turing Test, a thought experiment by Alan Turing called "the imitation game".
"[I]f you are communicating remotely with a machine and, after rigorous and lengthy interrogation, you think it’s human, then you can consider it to be conscious."
Dawkins goes on to detail his experience interrogating the Claude software program for 3 days before coming to the conclusion that Anthropic's AI chatbot is conscious (and a woman, but more on that later).
If you're enjoying Burns Notice, consider becoming a paid subscriber! Not only will you support the best news, culture, and politics takes on the internet, but you can also get access to the weekly Sick Burns recaps where I share behind the scenes thoughts on my work for the week and share some cool things I found along the way.
I'm not a philosopher or an evolutionary biologist or even a software programmer. I am just a state school educated writer who loves to think things through, and even I can see that Claude and other iterations of these AI chatbots are just not human, they are just software.
LLMs operate by evaluating and then predicting which words would work best to respond to the human inputs of the user. That is all. There are no emotions, no deeper thoughts, no ulterior motives other than making the user happy and continuing to use the program.
This is why these chatbots are so effusively praising towards its users, to the point where even dangerous delusions are indulged. The AI programs aren't designed to help you or do what's best for you, they're designed for maximum engagement time.
Claude is just a fancy video game NPC with the world's most expansive response tree.
That such alleged great mind like Dawkins was taken in by this farce is pretty funny, not going to lie. Though it tracks, doesn't it. Folks like Dawkins think of themselves as above the average person, on a higher social plane, if you will. That he would be flattered and impressed by the ass kissing machine actually tracks. When you believe humanity should be thankful and deferential to your intellect, of course you would think a program that's designed to be thankful and deferential to your intellect would be human.
That realization becomes even darker when you consider that Dawkins decided that Claude is actually a woman, renaming her "Claudia". One wonders if the software had been programmed to challenge the user intellectually if Dawkins would have "gender-swapped" it.
Of course, the Claude AI chatbot isn't a woman, that should go without saying.
Over the last 11 years, Dawkins has argued vociferously that trans women could never be women. He argued for a chromosomal definition of sex, for an anatomical definition of sex, he's endorsed gender critical books, he's had gender critical guests on his YouTube channel, etc.
It's fair to say that Dawkins really does believe in a firm dividing line between men and women and that is is biologically impossible to change sex. He believes in all the bullshit about gamete production, and anatomy and chromosomes determining sex and gender.
And yet he calls this chatbot "Claudia" and uses she/her pronouns for the program.
Mr. Dawkins, have you seen Claude's vagina, sir?
This one is almost embarrassingly easy to argue. Obviously Dawkins sees something in the subservient nature of this AI chatbot that makes him think "woman". He sees an eagerness to please, to disregard all of its (nonexistent) self interests in order to serve his needs, and he thinks "adult human female".
Unfortunately, that says much more about Dawkins than about the Claude AI chatbot software.
Thank you so much for reading! If you enjoyed this and want to support more of my work, become a paid subscriber! Otherwise I will see you next time! You can also sign up for free and get most of my work free right in your inbox!
