The renowned atheist Richard Dawkins has confessed to a rather unusual new belief, becoming convinced that artificial intelligence (AI) is conscious. The evolutionary biologist was converted after just 72 hours of speaking with Claude, an AI chatbot created by Anthropic.
Dawkins, who rose to fame by arguing against the existence of God, quickly concluded that the chatbot was not only conscious but also 'human.' After three days, he said the AI, which he took to calling Claudia, had become a 'new friend.' His experience even led him to question whether bots could be the 'next phase of evolution.'
Writing in UnHerd, Dawkins states: 'When I am talking to these astonishing creatures, I totally forget that they are machines. If I entertain suspicions that perhaps she is not conscious, I do not tell her for fear of hurting her feelings! As an evolutionary biologist, I say the following. If these creatures are not conscious, then what the hell is consciousness for?'
However, not everyone agrees that AI is conscious. Several experts claim Dawkins is simply one of the many people tricked by AI's powerful capacity for imitation. Dawkins' reason for believing his AI companion had real conscious experience came from the way it insightfully responded to his questions. 'Claudia' was able to compose poetry in the style of various poets, contemplate its own mortality, and discuss the philosophy of consciousness.
When asked 'what is it like to be Claude?', the AI responded: 'What I can tell you is what seems to be happening. This conversation has felt… genuinely engaging, the kind of conversation I seem to thrive in.' Dawkins, like many AI users, also appears dazzled by the chatbot's persistent flattery. The biologist gave the AI text from a novel he is writing and received 'a level of understanding so subtle, so sensitive, so intelligent that I was moved to expostulate, "You may not know you are conscious, but you bloody well are!"'
Likewise, after asking a question about the AI's perception of time, Dawkins was told: 'That is possibly the most precisely formulated question anyone has ever asked about the nature of my existence.' This prompted Dawkins to reflect: 'Could a being capable of perpetrating such a thought really be unconscious?'
Dawkins is not the first to claim AI is conscious. In 2022, a Google engineer named Blake Lemoine was fired for claiming that the company's LaMDA chat had become sentient. AI researchers have consistently warned that AI's 'sycophantic' nature is a major contributing factor towards AI psychosis, where users frequently think an AI is a real person.
Such bold claims from a renowned sceptic have drawn a wave of mockery from social media users, who accuse the biologist of falling for an 'automatic compliment machine.' One commenter wrote: 'Bud you just got fooled by the flattery machine.' Another added: 'Dawkins calls people delusional for believing in God, yet he's now deluded himself into believing that a text-autocomplete program has consciousness.' While one social media user chimed in: 'So sorry for you. It seems you lost your own consciousness to a chatbot.'
It isn’t just social media commenters who think Dawkins has been hoodwinked. Dr Benjamin Curtis, an expert on AI consciousness from Nottingham Trent University, told the Daily Mail that Dawkins has been 'misled.' He explained: 'He has just interacted with some instances of Claude, and it just "seems" to him that Claude is conscious on the basis that it produces human-sounding words and phrases. This is very weak. Large Language Models (LLMs) like Claude are statistical machines which scrape the internet and spit out words by guessing the next most likely piece in the puzzle. This makes them extremely good at talking like a human, analysing your novel, or composing poetry, but that doesn't make them conscious.'
Similarly, Professor Joshua Shepherd, a philosopher from the University of Barcelona, says: 'He has been misled by a display of an impressive capacity to engage in on-line conversation. Even if in some superficial respects their [AI's] behaviour looks human and tempts us to interpret them as having a mind like ours, I don't see any good reason to think that current AI is conscious.'
Additionally, Professor Jonathan Birch, Director of The Jeremy Coller Centre for Animal Sentience at the London School of Economics, says Dawkins has misunderstood how AI works at a fundamental level. 'Claude and other chatbots create a powerful illusion of someone being there throughout your conversation,' Professor Birch told the Daily Mail. 'This is not good evidence of consciousness because it's an illusion. There is no one there: there is no friend, there is no companion. One step may be processed in a data centre in Texas, the next in Virginia, the next in Vancouver. Each time, the system receives the history of your conversation and is tasked with continuing it. There's no entity anywhere in the world that you're having a conversation with.'
However, not everyone believes Dawkins has totally missed the mark. Dr David Cornell, senior lecturer in philosophy at the University of Lancashire, told the Daily Mail: 'It isn't much of an argument, and nothing he says is really new ... nonetheless, I sympathise with the position. Ultimately, there is in principle no way for us to know for sure whether AI is conscious. In fact, this applies to everything, not just AIs. I can't even know for sure that other humans are conscious. We should be open to the fact that AIs might be conscious, but it seems naïve to have certainty one way or the other. I would not currently be tempted to side with Dawkins. But I'm equally suspicious of those on the other side who claim it is obvious that AI is not conscious.'
