OpenAI is rolling browser and app control out to customers while Anthropic pushes Claude Cowork. The open question is not capability but which credentials the software gets, and who reviews the log.
Publishers:businessinsider.com
Reality
- Evidence38
- Adoption41
- Hype gap+32
- Incentives76
- Confidence44
Record & Replay and Record a Skill shipped weeks apart. Both bet the context agents need lives in what people do, not in what they can be bothered to write down.
Publishers:fastcompany.com
Reality
- Evidence22
- Adoption18
build1 distinct publisher RuntimeWire's teardown of current builds describes an unreleased recorder, Parka, whose action schema routes each follow-up to Cowork, Claude Code, or a human. The code to run it is absent.
Publishers:runtimewire.com
Reality
- Evidence74
- Adoption6
Anthropic says per-action human approval degraded into a rubber stamp inside its own products. The control it now spends most of its engineering on is what the agent can reach.
Publishers:anthropic.com
Reality
- Evidence55
- Adoption58