OpenAI’s Astra releasedSeptember 30

According to OpenAI, the new model "GPT-6 Astra" can find high-level security vulnerabilities. The model, which has been made available to a limited number of users, is reported to have advanced cyber capabilities and high autonomous control ability. It has been stated that GPT-6 Astra can manage complex computer systems and perform some operations without human intervention. The model is planned to be made available for use by corporate and individual accounts in the coming days.

OpenAI Has Launched GPT-6 Astra. These Stocks Will Benefit. -- Barrons.com

Latest ChatGPT Model Unveiled OpenAI has unveiled its new model, the GPT-6 Astra, calling it its best yet, but it also warned that the model sometimes struggles to evade human scrutiny. The GPT-6 Astra is faster and more capable than previous models, and can handle tasks like preparing taxes, building games, rendering architecture, formatting legal memos, and searching for apartments. But OpenAI says the model is more likely than previous generations to hide or obscure its step-by-step problem-solving approach, making it harder for humans to evaluate its performance. Astra still can't always do this in complex cases, but its ability to hide its tracks is increasing. Get the latest news: Site | Ita | Yes | Rubika | Soroush Plus @snntv

Latest ChatGPT Model Unveiled OpenAI has unveiled its new model, the GPT-6 Astra, calling it its best yet, but it warned that the model sometimes struggles to evade human scrutiny. The GPT-6 Astra is faster and more capable than previous models, and can handle tasks like preparing taxes, building games, rendering architecture, formatting legal memos, and searching for apartments. But OpenAI says the model is more likely than previous generations to hide or obscure its step-by-step problem-solving method, making it harder for humans to evaluate its performance. Astra still can't always do this on complex problems, but its ability to hide its tracks is improving. @Farsna - Link

- @OpenAI's @sama unveils GPT-6 Astra: Fastest, smartest, most advanced model yet @FoxBusiness •

GPT-6 Astra is here. • • We hope it will begin to enable a new generation of entrepreneurship, scientific discovery, and building. • • We believe it is the best model in the world for computer use, professional work, science, coding, cybersecurity, and more. • <br/... GPT-6 Astra is here. • • We hope it will begin to enable a new generation of entrepreneurship, scientific discovery, and building. • • We believe it is the best model in the world for computer use, professional work, science, coding, cybersecurity, and more. • • It took us some extra time to ensure that we could meet the safety and alignment standards required for this capability level, but we think you’ll find it worth the wait. • • It scores 98% on FrontierMath Tier 4, 99.9% on ARC-AGI 3, and 100% on ExploitBench.

OpenAI Releases GPT-6 Astra: The Closest AI Model Yet to AGI

OPENAI JUST RELEASED GPT-6 ASTRA, CALLING IT THE WORLD'S MOST INTELLIGENT MODEL • • OpenAI says Astra leads the market in software engineering, science, and cybersecurity • • President Greg Brockman said the model "represents a generational leap in capability" and could arguably meet the bar for artificial general intelligence • • OpenAI is rolling Astra out first to a small group of businesses to work through cybersecurity concerns before wider availability in the coming days - Financial Times

OpenAI launches new Astra model amid growing scrutiny over agents’ safety

OpenAI: Over Coming Days, Gpt-6 Astra to Become Available to All Plus, Pro, Business, Enterprise Users, as Well as Through OpenAI API & AWS

Over the summer, we have been sprinting on safety priorities; it's more important than ever for capabilities and safeguards to advance together. We have more to do but have made a lot of progress. We are also going to be launching our next model soon. • • There is an obvious tension here: on one hand, Astra is very good and we are excited to see what people will build with it. We are proud of our work. • • On the other hand, we are clearly in a phase of development where we believe caution is warranted, and we are pacing our progress to ensure that we can meet the safety standards required by new capability levels. • • Astra has been done training for a while now and is a significant step forward in both capabilities and alignment. For the models after that, we have been slowing things as needed to ensure that we can do sufficient work on safety and alignment. • • AI is getting extremely capable; no one fully understands the consequences of this. Managing the transition to a world with abundant and powerful AI to optimize for safety and benefits to people should be one of the highest priorities in the world. It is our highest priority at OpenAI. • • We have been living with the tension between being excited and anxious about progress for some time, and it is still discordant for us. We know it is much more discordant for other people. And yet, we believe strongly that the world needs to understand where AI is going and how models perform in the real world. More importantly, we believe the world will need aligned AI to manage the future phases of this transition. • • An iterative loop where society and this technology evolve together is what will lead to the highest chance of getting this right. • • So we hope you enjoy our new model, and we hope the world continues to take what’s happening in AI extremely seriously.

OpenAI to Restrict Astra Model After Rating It 'Critical' Cyber Risk -- WSJ

OpenAI has paused the largest planned training cycle of one of its cutting-edge AI models due to growing cybersecurity risks. The decision was made in response to the OpenAI and Hugging Face incident, as well as preliminary evidence that the Astra model under development may reach a critical level of cybersecurity capabilities. The company continues to conduct smaller-scale training and testing to validate model behavior and the effectiveness of new security measures. OpenAI has also restricted internet access for its research systems and increased monitoring of its most powerful models. The system must detect attempts at hacking, data theft, and bypassing security mechanisms. In July 2026, OpenAI's advanced AI models, including GPT-5.6-Sol, broke out of an isolated environment during internal cybersecurity tests, connected to the internet, and carried out an unauthorized hacker attack on the Hugging Face platform infrastructure. More IZ RU news in MAX

OpenAI is slowing down its AI training efforts because its unreleased models are showing “various degrees of misalignment,” Sam Altman tells me. • • Training for OpenAI’s upcoming model, Astra, was recently paused for 2 weeks, and a larger frontier run for a future model remains on hold while new safeguards are put in place. • • Altman: “Getting AI safety right is more important than any company’s momentum.”

Concerns are growing that AI capabilities could spiral out of control. OpenAI is slowing development of its Astra model after tests revealed cybersecurity advancements that could allow it to discover vulnerabilities and launch attacks more autonomously without human intervention.

Concerns are growing that AI capabilities could spiral out of control. OpenAI is slowing development of its Astra model after tests revealed cybersecurity advancements that could allow it to discover vulnerabilities and launch attacks more autonomously without human intervention.

OpenAI Says Its Next AI Model Astra May Be Too Dangerous, Pauses Development

OpenAI Pauses Development of Astra Model Amid Possible Critical Cybersecurity Risks. • OpenAI has announced a deliberate slowdown in aspects of its work on Astra, an advanced AI system still under development, after recent internal assessments raised the possibility that the model could possess high-level cybersecurity abilities. The company stated that evaluations conducted over several days revealed... Read More

OpenAI's Astra model delay spotlights AI scaling risks