Posts

Automation bias predicts that smooth operation trains humans out of checking. manrun is a built response—a gate that forces the practitioner to re-engage before an agent's blocked command executes. The tool exists because the literature said it should.

How might Google support users to calibrate trust within the tool itself?

Here is the 'OWASP Top 10 for Large Language Model Applications'. Overreliance is relevant to my research. (I’ve generally used the term “automation bias”, though perhaps a more direct term like overr...