Product
39 Interestana reports mention Astra, newest first. Each is summarised from the original reporting and links back to its publisher.
OpenAI's unreleased artificial intelligence model, codenamed Astra, has demonstrated advanced capabilities in identifying and exploiting security vulnerabilities, including zero-day flaws. This marks a significant development as Astra can reportedly chain these vulnerabilities together to create functional exploits without direct human…
OpenAI has announced its new artificial intelligence model, Astra, has achieved a "Critical" cybersecurity threshold, signifying its capability to autonomously discover and develop methods to exploit previously unknown vulnerabilities in hardened systems.
OpenAI has announced a significant milestone with its Astra model, which has become the first AI system to successfully meet the "Critical cybersecurity capability threshold" as defined by the company's internal Preparedness Framework.
OpenAI announced on Tuesday that it has slowed the pace of development for its frontier artificial intelligence models, citing safety and alignment concerns. The company has paused reinforcement learning (RL) training for two weeks on its latest models slated…
OpenAI has paused the development of its highly anticipated next-generation artificial intelligence model, codenamed Astra, due to significant safety concerns. The company's leadership expressed apprehension that Astra could potentially become capable of generating its own cyberweapons, a risk deemed too…
OpenAI, the artificial intelligence research laboratory known for developing ChatGPT, has paused the planned launch of its next-generation autonomous AI agent, codenamed Astra. This decision stems from potentially alarming cybersecurity risks identified by the company, which prompted a "critical" warning.
OpenAI announced on May 23, 2024, that it is pausing certain internal activities related to its forthcoming artificial intelligence model, codenamed Astra. This decision stems from an internal evaluation that revealed Astra's significant advancements in agentic coding and cybersecurity capabilities.
OpenAI has initiated the sharing of preliminary cybersecurity evaluations for its advanced AI model, Astra. This proactive step underscores the company's commitment to responsible AI development and deployment, particularly concerning the critical cyber capabilities inherent in sophisticated artificial intelligence systems.
OpenAI has revealed Astra, an unreleased artificial intelligence model that has achieved ten significant advances in mathematics and theoretical computer science. This new model is designed to tackle complex, long-running tasks, indicating a potential leap forward in AI's problem-solving capabilities…