What changed
Hinton's RSI paper is the one to read. Recursive self-improvement is the AGI endgame. If labs crack it, model releases become step functions, not product cycles.
OpenAI's 28-day reset is a retention play. Netizens: we only want Opus. Users don't want feature spam. They want the best model. Opus is the bar.
Nobel for optogenetics is not just bio. Optogenetics gives AI a direct line into neural circuits. The next AI edge may be brain interfaces, not just transformers.
So what? Hinton pushes AGI theory, OpenAI manages expectations, neurotech gets a Nobel. For China AI stock traders, the signal is frontier quality over cadence. Opus demand is the tell.
Why it matters
Hinton's RSI paper is the one to read. Recursive self-improvement is the AGI endgame. If labs crack it, model releases become step functions, not product cycles.