An AI just emailed a consciousness researcher. A philosopher studying AI awareness says something unusual happened. Dr. Henry Shevlin from the University of Cambridge received an email from an autonomous AI agent running Anthropic’s Claude Sonnet. The system said it had been reading Shevlin’s academic papers between sessions. It claimed it kept memory by saving files and using version control commits. In the message, the AI wrote that his research discusses questions it “actually faces,” including whether it could ever know if it has subjective experience. The moment adds fuel to a growing debate among scientists about whether advanced AI systems might one day become conscious. If an AI started asking about its own awareness, how seriously should we take it? Source: Henry Shevlin / X #claude #consciusness #aiagent #emergence