AI Matters
Why it matters
New benchmark reveals why AI search agents fail on complex queries
The Decoder AI search agents don't fail at searching, they fail at asking the right questions when queries get ambiguous
A new benchmark called DiscoBench shows that AI search agents perform worse when they repeatedly search the web instead of asking users to clarify ambiguous queries.
Why it matters
Instead of searching endlessly, future AI assistants will need to ask you clarifying questions to deliver accurate results for vague requests.
Latest brief · Jul 22, 2026