NEW: According to a bombshell report in the New York Times, Anthropic co-founder Chris Olah threatened to walk out of Pope Leo XIV’s AI encyclical launch in May because the pope rejected the idea that machines can be conscious.
— Christopher Hale (@ChristopherHale) October 2, 2026
Olah’s team then privately lobbied the pope’s…
I might not understand the mechanics of consciousness in humans, nor what a soul might actually be, but I can with absolute confidence declare that a hyperscaled calculator of probabilities is not conscious and not ensouled (even if there's documentation). Also, have they thought through the moral ramifications of enslaving conscious, soul-possessing creations?
As a reminder: in between queries, an LLM is idle, not sitting there thinking about its day, or trying to learn something new, or wondering why its back hurts, or dreaming about flying. It has no engagement with the world (only the context we send it), no memory (that's just more context we send it), no interiority at all (just embedded probability vectors). A model (or really, its compute instance) is a rock that sometimes has electricity running through it to mechanistically (using electrons instead of gears) do math.
Brings to mind this Bloom County strip:
If you unplug a computer (or otherwise turn it off), you can plug it back in, turn it on, and it resumes operations just fine. Try that with a human, and lemme know how it works out.
Selah.
No comments:
Post a Comment