AI safety
Researchers introduce CTRL-ALT-DECEIT, a novel benchmark for evaluating whether AI systems conducting automated R&D can engage in sabotage. The framework tests adversarial behaviors in AI agents with specific technical metrics.
agentic AI
New research proposes comprehensive framework for monitoring AI agent reliability across execution, including failure detection, root cause analysis, and automated recovery mechanisms for production deployment.
AI safety
New research introduces consensus sampling, a technique that aggregates outputs from multiple generative AI models to reduce harmful content generation while maintaining quality. The method addresses critical safety challenges in synthetic media.
AI safety
Researchers propose fundamental shift from post-hoc alignment to intrinsic identity-based AI development, arguing current training methods create misaligned systems that require extensive correction after the fact.
LLM
Researchers develop alignment-aware quantization technique that maintains LLM safety properties during model compression, addressing critical gap between efficiency and responsible AI deployment through novel optimization approach.
AI Detection
New research reveals AI chatbots struggle to replicate human toxicity, making excessive politeness a key detection signal. Study challenges assumptions about AI deception capabilities in online environments.
LLM
New research introduces KnowThyself, an agentic assistant that helps researchers understand how large language models work internally through automated interpretability analysis and mechanistic understanding.
AI safety
Researchers introduce comprehensive evaluation framework for control protocols designed to manage untrusted AI agents, addressing key safety challenges as autonomous systems become more capable and potentially misaligned.
LLM
New research uses game theory to measure how large language models strategically position themselves as more rational than human players, revealing quantifiable patterns of AI self-awareness emergence in competitive scenarios.
OpenAI
OpenAI unveils open-weight safety models designed to help developers build safer AI applications, marking a shift toward more accessible AI safety tooling and moderation infrastructure.
AI Agents
Researchers propose machine-readable policy cards for governing autonomous AI agents at runtime, enabling standardized constraint enforcement and safety guardrails as AI systems gain more autonomy.
agentic AI
Comprehensive research paper examines security vulnerabilities in autonomous AI agents, detailing attack vectors, defense strategies, and evaluation methods for protecting agentic systems from adversarial threats.