Baseten

Reading a model's internals instead of its output cut the cost of watching for cyber misuse by about thirty times, Goodfire says

A lightweight probe reads activations at every turn and escalates only what looks wrong, which drops judge cost from $3,166 to $65 per million turns.

Your ChatGPT plan can now pay for inference inside other companies' products, and enterprise commitments can buy Figma

Requests made in someone else's app count against the usage included in your plan, with a weekly limit you set per app.

Three companies that host and serve open models will try to write a safety standard for them

The target is abliteration, the practice of surgically removing a model's refusals. The three say safety has to be built into how open models are trained and served, not bolted on.