Opus 5.5 codes a visualization of Claude's answer on AI consciousness
After asking Claude whether it is conscious, the author had Opus 5.5 code a beat-timed visualization of a starry brain collapsing into a single guess.
I ASKED CLAUDE IF IT'S CONSCIOUS. IT WOULDN'T SAY NO. Its answer was "I don't know. Neither do the people who built me." I checked, and that's true. Last week Anthropic co-founder Chris Olah told the New York Times almost the same thing, "We don't know if A.I. models are conscious. I don't know. I'm genuinely uncertain." He cares enough about the question to spend months in sessions with religious thinkers, from rabbis to Mormon scholars, on how Claude should behave and whether it can suffer. Pope Leo XIV wrote in May that AIs "do not feel joy or pain," and per the Times, Olah lobbied the Pope's advisers to take AI consciousness seriously. So while the Vatican and Claude's own creators argue about what happens inside it, I had Claude show me. Opus 5.5 coded the whole visualization below, the starry brain, the collapse into a single guess, every word timed to the beat. The only one in this story who sounds certain is the Pope.
This post quotes another post: x.com/thesethian/status/2106802061589254478