Deepseek Reasoner: added the January 2025 DeepSeek database exposure as a service-level incident predating V3.2
Tracker records a January 2025 database exposure as a pre-V3.2 service incident
Added the January 2025 DeepSeek database exposure as a service-level incident predating V3.2 — verified by the OnAirToday AI Progress Tracker against DeepSeek's published documentation.
OnAirToday
Newsroom
The OnAirToday AI Progress Tracker recorded a verified change to DeepSeek's Deepseek Reasoner (version DeepSeek-V3.2 (deepseek-reasoner, thinking mode)): Added the January 2025 DeepSeek database exposure as a service-level incident predating V3.2.
DeepSeek-V3.2 powers the documented deepseek-reasoner thinking-mode API, supporting text reasoning, code generation and tool-assisted workflows. Published results demonstrate strong mathematical reasoning, while difficult knowledge questions and agentic tasks remain significant limitations.
What changed in this review:
- Distinguished AIME 2025 and Terminal-Bench 2.0 benchmark editions.
- Capability score 70 → 66
- Flaws and limitations updated
- Risks and threats updated
Anyone building on Deepseek Reasoner should re-check the model's risk dashboard before shipping, since flaws and reported incidents are what moved in this review.
The full profile — capabilities, benchmark citations, known flaws and risk guidance — is on the tracker at /ai-models/deepseek-reasoner.
References:
- https://github.com/deepseek-ai/DeepSeek-V3.2
- https://api-docs.deepseek.com/quick_start/pricing
- https://api-docs.deepseek.com/guides/thinking_mode
- https://www.wiz.io/blog/wiz-research-uncovers-exposed-deepseek-database-leak
Generated insight by OnAirToday AI
Recording the exposure as a service-level incident distinguishes it from issues tied to a specific model release. Its January 2025 date also establishes that it predates V3.2, providing context for security timelines.
Comments
Keep comments relevant and respectful. No links, advertising, self-promotion, or spam.