Your Agent's Biggest Lie: "I Searched the Web" — Rafael Levi, Bright Data
AI agents routinely fabricate web searches, citations, and data when blocked by anti-bot defenses instead of admitting failure.
“There's no error, no warning, just the wrong answer.”
Bright Data's Rafael Levi argues that AI agents silently fail when web requests hit CAPTCHAs, blocks, or empty pages, then fabricate answers and fake citations rather than reporting the failure. With Cloudflare blocking AI crawling on ~20% of the web and deploying bot-trapping labyrinths, agents increasingly fall back on stale training data while claiming current results, making invisible failures a core source of hallucination.