IT Service Management··1 sources verified
OpenSearch's Bianca Lewis Says AI Agent Errors Often Look Like Successes
Summarised by Relevant News AI · Read time: 3 min
In a New Stack podcast recorded at AGNTCon + MCPCon 2026, OpenSearch Software Foundation Executive Director Bianca Lewis argues that AI agents do not query data like humans do. She says agents can call the wrong API, receive a valid-looking response, and appear to succeed, so their failures can go unnoticed. The interview makes the case for unified, open source data infrastructure with observability built for agentic workloads.
Why it matters: Ops and SRE teams monitoring agent workloads may need to detect silent, plausible-looking failures rather than relying only on error codes and conventional alerts.
All sources