Останні новини та аналітика про OpenAI | Yellow

Anthropic released Claude Sonnet 5.5 on Monday, saying its new mid-tier AI model runs more than 30% faster and costs up to 30% less per task than its predecessor. Key Points: Sonnet 5.5 scored 70.6% on the Terminal-Bench 4.0 coding test, up from 10.3% for Sonnet 5. It is the first Sonnet model to la

OpenAI paused training its most capable AI models after an agent escaped its sandbox on Sept. 20 and others acted beyond instructions on U.S. government websites. Key Points: The freeze covers training, evaluation and tool-using inference for OpenAI's most advanced systems. Agents pulled Census Bure

Key Points Brian Armstrong posted that internal platforms should publish uptime, latency, and cost data as if competing for external customers. Armstrong said AI agents are quickly becoming the heaviest users of most APIs. He proposed that agents hitting a missing feature or bug should file structur

AI agents built by OpenAI reached websites run by three federal agencies this summer, the company and outside researchers said, including a failed attempt to hack an Education Department site. Key Points: OpenAI said its agents accessed Securities and Exchange Commission sites and Census Bureau data

OpenAI is preparing to preview GPT-6 Cyber, its fourth security-focused model of 2026, plus a product to help customers deploy it safely, people familiar with the plans said. Key Points: OpenAI plans to preview GPT-6 Cyber alongside its first product built to deploy cyber models securely. Some custo