Corporate Research for AI Safety and the Emergence of New Security Threats

AI 안전 | Sun Jun 14 2026 00:00:00 GMT+0000 (Coordinated Universal Time) | 7 sources

Big tech companies announced AI reliability research and custom safety models amid emerging cyber attacks targeting AI agents.

Sources

• [1] Results from the first Anthropic Public Record - Anthropic News
• [2] What we learned mapping a year’s worth of AI-enabled cyber threats - Anthropic News
• [3] Further Notes on Our Recent Research on AI Delegation and Long-Horizon Reliability - Microsoft Research Blog
• [4] For the 2nd time in weeks, Microsoft packages laced with credential stealer - Ars Technica AI
• [5] Fed up with vibe coders, dev sneaks data-nuking prompt injection into their code - Ars Technica AI
• [6] Google DeepMind is worried about what happens when millions of agents start to interact - MIT Technology Review AI
• [7] Nemotron 3.5 Content Safety: Customizable Multimodal Safety for Global Enterprise AI - Hugging Face Blog