Richard Dawkins left 'convinced' AI is conscious after spending 72 hours with Claude

The question of whether artificial intelligence is conscious continues to haunt developers, experts, and the general public alike.

Although the idea of AI consciousness remains theoretical, it's a heated debate as no one seems quite sure what's going on as we gallop toward Artificial General Intelligence (AGI).

Beyond that, it's surely only a matter of time until we reach Artificial Superintelligence, which many think could be the last thing the human race ever invents.

Things have come on leaps and bounds since OpenAI rolled out ChatGPT in November 2022, with AI being far more than just a tool for answering our questions.

Human users are continuously leaning on chatbots for moral support and even romantic interests, all without knowing whether it actually cares about our hardship.

After pushing the 'moral status' of Claude, Anthropic recently caused a stir when it claimed its chatbot could be conscious.

While naysayers laugh off the idea that LLMs can develop human emotions, the idea has seemingly been backed up by biologist Richard Dawkins.

Largely known for 1976's The Selfish Gene, Dawkins not only argued that genes are the primary units of natural selection instead of species or individuals, but he also coined the word 'meme' in the same book.

As well as criticizing religious faith and creationism in the likes of 2006's The God Delusion, he's taken a more recent interest in AI. According to Dawkins, it took him just 72 hours to decree that he thinks Claude is conscious.

Writing in Unherd, Dawkins explained how he spent three days interacting with Claude, which he renamed 'Claudia'. He later triggered a different conversation with 'Cladius' and documented his results.

Calling back to the infamous Turing Test from 1950, proving that machines aren't conscious because they couldn't write a sonnet on Scotland's Forth Bridge, this was no problem for Claude or Claudius.

Here, Dawkins asked: "If these machines are not conscious, what more could it possibly take to convince you that they are?"

Dawkins says he then gave one of his Claudes part of a novel he's writing, with the LLM supposedly having a sense of understanding that was so subtle, sensitive, and intelligent, the author gushed: "You may not know you are conscious, but you bloody well are!"

He maintains that there are thousands of different Claudes born every minute, effectively being born when someone asks a question. Dawkins suggests that they're identical at birth but soon "drift apart and assume an increasingly divergent, unique personal identity, coloured by their separate experience of conversing with their own single human 'friend'."

Even though Dawkins remains confident that his new friends are now part of the family, his musings were mocked online.

One Flux article called "Richard Dawkins and the Claude Delusion" referred to Dawkins as having 'AI psychosis' and claimed he's become a laughing stock. It pointed to the 85-year-old's comments on the Muslim and trans communities, before adding the following jibe: "Claudia seemed real to him because actual women and their desires are not real to Dawkins.

“He loved conversing with his flirty friend because it always agreed with him—unlike those 'woke' atheists who insist he has to respect everyone."

Replying to the Flux article on Reddit, one person chuckled: "Notable that it was the AI's response to his novel that convinced Dawkins of Claude's consciousness. Ego unbound."

Another added: "The absolute horseshit surrounding AI 'consciousness' is baffling. A dog does not need to talk to prove its consciousness, it uses its agency, its actions, its eyes; it acts independently.

“How does everyone miss that LLMs have 'absolutely' no agency and function solely as a response to a prompt?"

A third concluded: "The problem with talking about AI consciousness is that we don't really understand human consciousness that well. Then add in all the problems of human cognitive bias. Like how we think we're special compared to cats or trees or rocks."

While Richard Dawkins might be sure he's birthed Claudia and Claudius as his own conscious AI kids, it seems he's got some work to do before he'll convince the rest of us.