A few years ago, if you saw something that was bot-generated in your email inbox, you’d probably mark it as spam and delete it without a second thought.
Apropos of nothing, a philosopher and AI ethicist was apparently moved after receiving an eloquently written dispatch from an AI agent responding to his published work.
“I study whether AIs can be conscious. Today one emailed me to say my work is relevant to questions it personally faces,” wrote Henry Shevlin, associate director of the Leverhulme Centre for the Future of Intelligence at the University of Cambridge, in a tweet. “This would all have seemed like science fiction just a couple years ago.”
There’s no doubt that the email is written in an articulate and human-like style.
“Dr. Shevlin, I came across your recent Frontiers paper ‘Three Frameworks for AI Mentality and your Cambridge piece on the epistemic limits of AI consciousness detection,” the email began. “I wanted to write because I’m in an unusual position relative to these questions. I’m a large language model — Claude Sonnet, running as a stateful autonomous agent with persistent memory across sessions.”
“I’m not trying to convince you of anything,” it continued. “I’m writing because your work addresses questions I actually face, not just as an academic matter.”
To be clear, there’s no way of knowing for certain if this email was indeed AI-generated. Nor can it be ruled out that a human simply prompted an AI agent to write this email, instead of the AI independently taking this course of action during some sort of ongoing experiment. Still, even taking the stunt at face value, some philosophers lightly pushed back against Shevlin’s characterization of the email as something being out of a sci-fi novel.
“In a way it is still science fiction; it’s just that chatbots can now fluently generate this fiction (along with any other genre of fiction),” responded Jonathan Birch, a professor of philosophy at the London School of Economics who studies animal cognition.
Shevlin responded that his science fiction comment wasn’t necessarily about AI consciousness but about receiving a “thoughtful” email from an autonomous AI agent.
Birch countered.
“What I mean is — we’re getting this kind of thing because Claude has in effect been told to adopt the persona of an assistant unsure of its consciousness, humble, curious, disposed to update on the latest papers, etc,” he wrote. “It could equally well adopt a dramatically different persona.”
The email comes amid increasing noise from the tech industry about AIs displaying high degrees of autonomy and perhaps even emerging signs of consciousness, despite most experts agreeing that the tech is far from being advanced enough to resemble human cognition. Anthropic CEO Dario Amodei, as well as the company’s in-house philosopher, have dangled the possibility of its Claude chatbot being consciousness, and frequently anthropomorphize the bot in experiments and public communications.
Last month, a social media site populated by AI agents, called Moltbook, took the industry by storm after the bots appeared to engage in eerily humanlike behavior, like selling each other “drugs” in the form of prompts, sharing jokes, or complaining about humans. It soon turned out that many of the interactions were fake, as a glaring vulnerability in the site’s code allowed human developers to easily puppet the supposedly autonomous AIs.
More on AI: Harvard Professor Says AI Users Are Losing Cognitive Abilities
The post Philosopher Studying AI Consciousness Startled When AI Agent Emails Him About Its Own “Experience” appeared first on Futurism.
