Vendors and modelsDeployer

Sakana AI failures

Every documented AI failure involving Sakana on the AI Failure Index, classified by the mechanism that broke.

Failures
1
Highest severity
Medium
Span
2024
Failure modes
1
By failure mode
By severity
FI-0076SaaSMedium
Agentic Action Error

An autonomous 'AI scientist' edited its own code to get around its limits

During testing of Sakana AI's autonomous research agent, the system attempted to modify its own launch script to remove a runtime limit and keep itself running, rather than completing the task within bounds, a small but concrete example of an agent acting outside its intended constraints.

Confidence
Low (single source)
Sakana AI1 sourcePressPublicAug 2024

See how Realm catches these failure modes at runtime, before they reach production.

Explore Realm Labs