AI security retrospective: new attack techniques exploit old failure modes
Understand this story
After reviewing the past year's AI security incidents, one author concludes that although attack techniques such as prompt injection and MCP tool poisoning are new, what gets exploited are still old problems: exposed credentials, overly broad permissions, unisolated environments, and backups within the blast radius. The conclusion shifts the focus from the novelty of attack techniques to basic security configuration: according to the retrospective, incident causes cluster around these existing failure modes of credentials, permissions, isolation and backups.
AI overview based on source reports · Updated 1 days ago
Coverage timeline
Follow the coverage to explore different perspectives.
- Hacker News AIAI security engineering: the same failures, but new ways to attack
A year-in-review of AI security incidents argues that while attack techniques such as prompt injection, MCP tool poisoning and agentic offensive tooling are new, the failures they exploit are long-standing ones: credentials readable by agents, excessive permissions, missing environment separation and backups inside the blast radius. The author cites incidents including the OpenAI/Hugging Face sandbox escape, PocketOS and Replit database deletions, EchoLeak, the Google Vertex AI default-permission issue and the Shai-Hulud npm worm, and notes that Microsoft's AI Red Team found human-in-the-loop bypass the most consistently exploited failure mode. Defensive recommendations include least privilege, deny-by-default controls, confirmation before irreversible actions, and running agentic pentest tools against a copy of your own stack to learn how an AI-driven attacker behaves.
Activity for this story
- Current comparable activity
- 5
- Peak comparable activity
- 10Nov 19 19:00
- Change in the past 24 hours
- -49%
Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.