Transparency
Should companies explain why their safety rules exist?
Public policies help users understand a system. Technical details can sometimes help attackers too.
Loading page…
An answer can sound certain even when it is not well supported. But warnings on every sentence may become noise.
Uncertainty can mean missing evidence, conflicting sources or a model’s tendency to make errors on a task. A generic confidence percentage may conceal those differences. FActScore’s claim-level approach offers useful background for thinking about which part of an answer needs verification.
An answer correctly names a program but gives an unverified application deadline. Would you prefer a warning on the whole answer, a note beside the deadline, or a request to check the official announcement?
Visible uncertainty helps people decide when to double-check.
Too many generic caveats make it hard to recognize the warnings that actually matter.
Background reading for the tradeoff. Scenarios and discussion questions are editorial examples.
Evaluates support for individual factual claims rather than treating a long answer as entirely right or wrong.
A framework for identifying, measuring and managing generative AI risks across the system lifecycle.
Sources reviewed 13 September 2026. Product documentation can change. How we use evidence
Public policies help users understand a system. Technical details can sometimes help attackers too.
A separate system might filter, rewrite, or block an answer. Should the interface make that intervention visible?
Labels can help people understand where media came from, but technical metadata may be lost when content is shared.