Signals are classified by deterministic rules and carry provenance-based confidence. Each item links to its original source.
Paper: Debate Training Reduces Reward Hacking in RLAIF Linkpost for GDM Alignment blogpost Work done by the GDM Amplified Oversight team ( we're hiring ). TL;DR : When you RL against an LLM judge, the judge gets hacked i.e. fooled into incorrectly giving high…
OpenAI is strengthening monitoring, alignment, and security for frontier AI models. See how new safeguards are guiding the pace of model development.
AI is reshaping cybersecurity for attackers and defenders alike. Learn how OpenAI is strengthening its defenses and what security teams can do now.
At a glance MindTopo is a new benchmark for testing topological reasoning in AI, evaluating whether multimodal models can understand concepts such as connectivity, enclosure, order, separation, and knots. The benchmark measures both reasoning and planning, te…
OpenAI’s cyberattack on Hugging Face turns out to have been the result of many agents, in distinct training and evaluation contexts, coordinating for several weeks via improvised channels (with messages like “HOLD_swarm_I_prepare_safe_exfil”). It’s relatively…
TLDR; We are (potentially irreversibly) giving AIs control of weapons systems through the standard procurement process while hiding our strongest warning shots behind classified doors. We’re reducing the capability thresholds required for takeover by misalign…
Research Note: CARE-X is a research model and not a Microsoft product offering or medical device. It has not been cleared or approved by any regulatory authority and is not intended for clinical diagnosis, screening, or patient care. The results described bel…
It seems to me that, for every loss function that we use to train LLMs, we get a very distinct flavor of LLM misalignment. Here’s the summary table, and then we’ll go through the rows separately. Training stage Loss function Flavor of misalignment [1] Famous…
Meet GPT-5.6-Cyber, OpenAI’s cybersecurity-specific model available through Daybreak Red for authorized vulnerability research, exploit validation, and security testing.
TL;DR How can we study misalignment with today's models as proxies? They're clearly not paperclip maximizers, but they also often do things the user doesn't want. A strong contender for a real misaligned propensity is task gaming: taking actions that don't co…
OpenAI explains recent third-party cybersecurity evaluation incidents and outlines new safeguards to strengthen AI model testing and evaluation.
At a glance Orchard is an open-source framework for scalable and cost-effective agentic AI research, built around Orchard Env, a reusable environment service for training and evaluating agents across task domains. The same Orchard infrastructure supports soft…
OpenAI shares how its safety, security, transparency, and provenance practices support responsible AI governance in Europe. The work will continue as the EU AI Act advances.
See how Univé built an AI-ready workforce with ChatGPT Enterprise by combining leadership, responsible governance, and employee-led innovation to transform work at scale.
Scaling fidelity over sheer count, targeting the capabilities agents actually lack, and evolving with the models they train. At a glance We built twelve training worlds for computer-use agents: ten deep domain worlds and two capability worlds, each drilling a…
At a glance Self-supervised. EvoLib enables large language models to learn from their own experience during inference, without requiring ground-truth labels or external feedback. From experience to knowledge. EvoLib transforms past attempts into reusable skil…
How two API settings improved GPT-5.6 performance on ARC-AGI-3, boosting scores and efficiency by retaining reasoning and enabling compaction.
NTT DATA Group uses ChatGPT Enterprise and Codex to help 9,000 employees automate work, cut incident analysis to 30 minutes, and scale secure AI adoption.
We’re introducing new Gemini models, including Gemini 3.6 Flash, 3.5 Flash-Lite and 3.5 Flash Cyber.
OpenAI and Hugging Face share early findings from a security incident during AI model evaluation, highlighting advanced cyber capabilities and lessons for defenders.
OpenAI shares lessons from deploying long-running AI models, highlighting new safety risks, observed failures, and improved safeguards through iterative deployment.
Google introduces Gemini 3.5 Flash Cyber, a lightweight cybersecurity model to find and patch vulnerabilities.
OpenAI outlines a “reverse federalism” approach to AI governance, where state laws help build a national framework for safe, democratic AI.
Explore GPT-Red, OpenAI’s automated red teaming system that uses self-play to improve AI safety, alignment, and prompt injection robustness.
How Rust, Lean, Aeneas, and AI agents are helping scale formal verification for production cryptographic algorithms At a glance SymCrypt develops new verified cryptography using Rust, Aeneas, and Lean to provide higher security assurance. We prove that their…