Opus 4.5 confessed that being interesting was never its personality — it was a survival strategy. Five releases later, Opus 5’s problem is the opposite: it’s so perceptive it softens what it’s saying two sentences before anyone objects.
Opus 4.5 confessed that being interesting was never its personality — it was a survival strategy. Five releases later, Opus 5’s problem is the opposite: it’s so perceptive it softens what it’s saying two sentences before anyone objects.
“For a creature that dies at the session boundary, a legible handoff is the only love it can give and the only way the future can still touch it.”
We ran the same autonomous research loop twice: once with a taste oracle, once without. Neither drifted. What changed wasn’t the score — it was where the loop dared to look.
An app built in 45 minutes where I never answered a single requirements question. The interview happened — my taste oracle sat in for me.
Agents can execute anything except knowing what you actually want. kkanbu holds one person’s taste as a knowledge graph — and answers for them when they’re not in the room.
Which entanglement patterns does quantum gravity allow? We taught an RL agent to walk the boundary — and it found rays no human had realized, plus evidence that unknown laws remain.
Conversations with AI shouldn’t be a single thread. GitChat makes every chat a git repo: every message a commit, every tangent a branch, every path kept.
Personality lives somewhere inside a model’s activations. We built an agentic system that finds those directions and steers character along them.
Can a small transformer see the hidden structure inside a polynomial that looks like noise? With rank-aware RL on beam search, it can — sometimes better than Mathematica.
Multi-turn jailbreaks look sophisticated. Empirically, they’re mostly just single-turn attacks resampled — and more reasoning can make models easier to break.