I’d be careful about reading too much into these numbers. The test only looks at cases where the model doesn’t know the answer, so it doesn’t show how often users will actually see hallucinations.
I’ve always felt a bit uneasy about agents getting direct access to keys. Once they have them, it’s hard to know where those keys might go.
This feels cleaner to me. The agent does not need to see the real secrets. You still have to trust another layer, but that layer feels easier to control and reason about.
I think Google is using this to put pressure on OpenAI, while also getting some extra upside—like a possible path to acquire Anthropic later.
And honestly, this could turn out to be bad news for OpenAI.