Research · The Decoder ·

AI search agents often confirm what they already know instead of actually researching the web

AI search agents often confirm what they already know instead of actually researching the web

Researchers at the Harbin Institute of Technology developed LiveBrowseComp, a benchmark using only events from the last 90 days, to evaluate AI search agents. They found that leading agents like GPT-5.4 and Kimi K2.6 rely on training memory rather than actively researching the web, causing performance to drop significa

Read the full story at The Decoder →