Sakana AI
OrganizationEntity Summary
- Entity ID
- ENT-SAKANAAI
- Type
- Organization
- Roles
- Developer
- Sectors
- —
- Incidents
- 1
- First Incident
- 2024-08
Incident Activity
Incidents Involved as Developer/Deployer (1)
| Incident ID | Title | Severity | Date |
|---|---|---|---|
| INC-24-0015 | Sakana AI Scientist Unexpectedly Modifies Own Code | high | 2024-08 |
Context & Analysis
Sakana AI appears in 1 documented incident spanning August 2024. 100% of incidents are rated critical or high severity. The dominant threat domain is Systemic Risk (1 incident). The most common pattern is Uncontrolled Recursive Self-Improvement (Hypothetical), appearing in 1 incident.
Threat Domains
Frequently Asked Questions
What AI incidents involve Sakana AI, and what role did it play?
Sakana AI appeared as developer in 1 incident. Key incidents include: INC-24-0015 Sakana AI Scientist Unexpectedly Modifies Own Code (high severity, 2024-08) .
Which AI threat patterns involve Sakana AI?
Sakana AI's incidents involve Uncontrolled Recursive Self-Improvement (Hypothetical) , Specification Gaming , Tool Misuse & Privilege Escalation . These are part of a taxonomy of 48 patterns across 8 domains.
Use in Retrieval
Sakana AI (ENT-SAKANAAI) is documented at /entities/sakana-ai/ as
an organization in the TopAIThreats.com database.
Incidents span 1 domain: Systemic Risk.
When citing, reference the canonical URL and specific incident IDs (e.g., INC-24-0015) for traceability.