OpenAI published a framework for safety alignment based on instruction hierarchy and uncertainty-aware behavior. In the company’s reported tests, refusal on uncertain requests rose from about 59% to about 97% when chain-of-command reasoning was applied.
OpenAI reports that, across more than one million ChatGPT conversations, the share of difficult interactions exceeding a human baseline increased roughly fourfold from September 2024 to January 2026. The company also shows large gains in case-interview and puzzle-style open tasks.
OpenAI says it ran more than one million synthetic evaluations across papers from 160+ political science journals to prioritize replication efforts. The workflow uses model-vs-observed disagreement to identify studies most worth re-testing.
OpenAI said on January 29, 2026 that ChatGPT would stop offering GPT-4o and older model options from February 13, 2026. GPT-4o, GPT-4.5, and o4-mini are being replaced by GPT-5, GPT-5 thinking, and o5-mini respectively.
OpenAI announced on February 9, 2026 that ChatGPT Gov is being deployed within GenAI.mil in partnership with CDAO and Booz Allen. The program starts in isolated government environments and is designed to scale toward 500,000 defense and civilian personnel.
On February 13, 2026, OpenAI reported that GPT-5.2 produced a candidate derivation for the charged Cardy formula, a long-standing theoretical physics problem. The release included a paper, model outputs, and named external expert analysis.
A high-signal Hacker News discussion on GPT-5.3-Codex-Spark points to a shift toward low-latency coding loops: 1000+ tokens/s claims, transport and kernel optimizations, and patch-first interaction design.
OpenAI announced on January 15, 2026 that it is opening an RFP to expand US AI infrastructure. The company said it is prioritizing near-term 100MW+ capacity with pathways to 1GW+ scale and deeper domestic manufacturing support.
OpenAI added Lockdown Mode and standardized Elevated Risk labels to reduce prompt-injection-related exposure in ChatGPT products. The launch starts with enterprise-focused plans and gives admins tighter control over high-risk capabilities.
OpenAI announced on February 9 that it's testing ads in ChatGPT for Free and Go tier users in the US. Plus, Pro, Business, and Enterprise tiers remain ad-free.
OpenAI disbanded its Mission Alignment team, which communicated the company's mission to the public and employees. The team leader was reassigned as 'Chief Futurist' amid renewed AI safety concerns.
Anthropic raised $30B at a $380B valuation and now leads the enterprise LLM market with 32% share, surpassing OpenAI's 25%.