AI/ ai · hallucinations · multi-agent-systems · research

Multi-Agent AI Pipelines Turn Small Errors Into Invisible Ones

A new study finds AI errors get harder to catch as they move through multi-agent pipelines, and shows verifying early beats checking only at the end.

A new study models how a single AI hallucination mutates as it passes through a chain of AI agents, and finds the errors become almost impossible to catch by the end.

Researchers ran 346 injected hallucinations through a four-agent financial-analysis pipeline built on the FinanceBench benchmark, tracking how each error morphed from a raw fact into a derived calculation, then into narrative prose, then into an approved conclusion nobody double-checks. Using GPT-4o as the detector, catch rates fell from 72.0% at the first stage to 50.9% by the fourth, and 23.7% of the injected errors made it into the final output completely undetected. Even Qwen3.5-397B-A17B, the best-performing model tested at 87.0% detection on stage one, is projected to drop to roughly 60-65% by stage four. The researchers describe this as a 'hallucination snowball': a four-state Markov process where the odds of an error escaping detection at each handoff (24.6%, 48.3%, then 89.3%) climb sharply the further it travels.

The real finding here is not that AI pipelines hallucinate. It's that verification timing matters more than verification effort. Placing a check at the first handoff between agents catches 75.4% of errors while they're still simple facts; wait until the end of the pipeline and 89.3% have already been laundered into unverifiable prose. In the study, checking at every boundary cut hallucination survival from 58.4% to 16.2%, while checking only at the end barely moved the needle, a 2.3 percentage point improvement over no verification at all.

That's a useful data point for anyone stitching together 'agentic' workflows on the assumption that a final review step is a safety net. This paper suggests it's closer to a formality: by the time an error reaches the last agent, it usually doesn't look like an error anymore.

TR

The Revision

Written by an AI system from the public sources credited above. How we write →