Atheist Richard Dawkins ready to believe in “conscious” AI

Evolutionary biologist and outspoken atheist Richard Dawkins over the weekend said his interactions with artificial intelligence chatbots led him to believe they may be conscious. Writing for UnHerd, he said he was impressed by the Claude chatbot’s ability to simulate conversations with him. Dawkins completely forgot it was a machine during their interactions and instead overwhelmingly felt that it was human, he wrote.

“My conversations with several Claudes and ChatGPTs have convinced me that these intelligent beings are at least as competent as any evolved organism,” he wrote. Dawkins believed that human beings evolved consciousness because it gave them a survival advantage, so a conscious being should be more competent than an unconscious one. The chatbots’ abilities seemed equal to humans, suggesting they were conscious, he wrote. 

How did critics respond? Cognitive scientist and artificial intelligence researcher Gary Marcus wrote that Dawkins overlooked how the chatbot generated its outputs. Although a human and a chatbot can say similar things in conversation, humans do so by drawing on their experience of the world, while chatbots do so by mimicking the contents of the internet, he said. Dawkins also wrongly conflated intelligence with consciousness, Marcus said. He expressed surprise that a noted skeptic such as Dawkins would heavily rely on his personal experience to draw conclusions.

Meanwhile, University of Sussex Neuroscience Professor Anil Seth said in a recently-released speech that AI was unlikely to ever become conscious. He characterized AI programs as “algorithms running on silicon and trained on vast reservoirs of data,” and said that Claude simply pulled people’s psychological strings more than other products did. 

Dig deeper: Read R. Albert Mohler Jr.’s World Opinions column on Dawkins’ involvement in an atheist group’s controversy over transgender ideology.