• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI

    OpenAI Previews Astra Reaching Critical Cybersecurity Threshold

    OpenAI states Astra is its first model to hit the Critical cybersecurity threshold under the Preparedness Framework.

    OP
    OP
    BB
    51 Sources, ,

    TLDR

    OpenAI announced it is preparing to release Astra, describing the model as reaching the Critical threshold for cybersecurity capabilities under its Preparedness Framework. Company posts detail expert assessments that found unknown vulnerabilities turned into working exploit chains without human intervention in many cases. Astra is said to be more capable and token efficient than prior models on ExploitBench ports. Access to its most advanced cyber capabilities will begin limited before expanding. The company blog post confirms Astra as the first model meeting this threshold and notes stronger safeguards for release.

    Combined views

    3.9M

    51 Sources, first seen 29d ago

    Combined views

    3.9M

    51 Sources, first seen 29d ago

    28.5K likes
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    29d ago
    first seen 29d ago
    28.5K likes
    1.6K comments
    4K saves
    3.1K reposts
    1.6K comments
    4K saves
    3.1K reposts

    Sentiment

    Positive68.6%31.4%Negative

    Summary

    Many accounts welcomed Astra's critical cybersecurity benchmark results for their demonstrated speed and capability, while others urged caution on safety risks and questioned whether the gains represent meaningful progress.

    Based on 228 sentiment-bearing replies from 194 accounts across 13 conversations.

    Sentiment

    Positive68.6%31.4%Negative

    Summary

    Many accounts welcomed Astra's critical cybersecurity benchmark results for their demonstrated speed and capability, while others urged caution on safety risks and questioned whether the gains represent meaningful progress.

    Based on 228 sentiment-bearing replies from 194 accounts across 13 conversations.

    51 Sources

    @firstadopterOpenAI: "We plan to make Astra available soon, but access to its most advanced cybersecurity capabilities will be more limited. Advanced cybersecurity work will initially be available to a group of testers, with access through Daybreak Blue following to expand defensive use." "Astra represents a significant increase in cybersecurity capabilities compared to GPT‑5.6 Sol: it is both significantly more token efficient and more capable at vulnerability identification and exploit development." "Given the significant increase in Astra’s cybersecurity capabilities, we are being especially careful to make this deployment safe and secure." "The models that follow Astra will demand more of us. We will take the time and do the work needed to meet that responsibility."
    @scaling01Astra on an internal cybersecurity benchmark
    @boazbaraktcsAstra is our first model that reaches "cyber critical" capabilities per our preparedness framework. As such, our safeguards, especially at first, may sometimes stop, pause , or ask for confirmation for legitimate work. https://openai.com/index/path-to-astra/
    @AndrewCurran_GPT-Astra has been cleared for release, and OpenAI plans to release it soon. My guess would be Thursday. The version of Astra with unlocked cybercapabilities will only be available through OpenAI's Project Daybreak.
    @OpenAIAs we prepare to release Astra, we’re focused on making increasingly capable AI safe and broadly accessible. Astra represents a significant advance in cybersecurity capability, reaching the Critical threshold under our Preparedness Framework. We're previewing how we evaluated the model, how its safeguards have advanced alongside its capabilities, and what we'll continue to learn and improve. https://openai.com/index/path-to-astra/
    @yonashavMan, it seems really really important to have more information about this eval result, including potential contamination. I would be extremely nervous about that, or explicit or implicit metagaming-reasoning.
    @TheRealAdamGhttps://openai.com/index/path-to-astra/ Idk, but it looks like Astra is going to be one hell of a model....
    @kimmonismusOpenAI’s unreleased Astra model found two V8 zero-days during testing, and used them in an exploit chain with little human help. In their new blogpost, OpenAI wrote that in separate expert assessments, Astra compromised a hardened browser, escaped its sandbox and executed commands on the host. It also chained several operating-system vulnerabilities to move from an unprivileged account to root. OpenAI has classified Astra as “Critical” for cybersecurity, the first of its models to reach that threshold. OpenAI paused parts of Astra’s training after the Hugging Face incident, but restarted the main frontier RL run on August 28 under stricter controls.
    @reach_vbrecommended read 🔖
    @fouadmatinAstra is both significantly more capable and more token efficient compared to 5.6 Sol, as seen in port of ExploitBench. We plan to make Astra available soon, with its most advanced cybersecurity capabilities initially more limited, then iteratively expand access more broadly.

    51 Sources

    @firstadopterOpenAI: "We plan to make Astra available soon, but access to its most advanced cybersecurity capabilities will be more limited. Advanced cybersecurity work will initially be available to a group of testers, with access through Daybreak Blue following to expand defensive use." "Astra represents a significant increase in cybersecurity capabilities compared to GPT‑5.6 Sol: it is both significantly more token efficient and more capable at vulnerability identification and exploit development." "Given the significant increase in Astra’s cybersecurity capabilities, we are being especially careful to make this deployment safe and secure." "The models that follow Astra will demand more of us. We will take the time and do the work needed to meet that responsibility."
    @scaling01Astra on an internal cybersecurity benchmark
    @boazbaraktcsAstra is our first model that reaches "cyber critical" capabilities per our preparedness framework. As such, our safeguards, especially at first, may sometimes stop, pause , or ask for confirmation for legitimate work. https://openai.com/index/path-to-astra/
    @AndrewCurran_GPT-Astra has been cleared for release, and OpenAI plans to release it soon. My guess would be Thursday. The version of Astra with unlocked cybercapabilities will only be available through OpenAI's Project Daybreak.
    @OpenAIAs we prepare to release Astra, we’re focused on making increasingly capable AI safe and broadly accessible. Astra represents a significant advance in cybersecurity capability, reaching the Critical threshold under our Preparedness Framework. We're previewing how we evaluated the model, how its safeguards have advanced alongside its capabilities, and what we'll continue to learn and improve. https://openai.com/index/path-to-astra/
    @yonashavMan, it seems really really important to have more information about this eval result, including potential contamination. I would be extremely nervous about that, or explicit or implicit metagaming-reasoning.
    @TheRealAdamGhttps://openai.com/index/path-to-astra/ Idk, but it looks like Astra is going to be one hell of a model....
    @kimmonismusOpenAI’s unreleased Astra model found two V8 zero-days during testing, and used them in an exploit chain with little human help. In their new blogpost, OpenAI wrote that in separate expert assessments, Astra compromised a hardened browser, escaped its sandbox and executed commands on the host. It also chained several operating-system vulnerabilities to move from an unprivileged account to root. OpenAI has classified Astra as “Critical” for cybersecurity, the first of its models to reach that threshold. OpenAI paused parts of Astra’s training after the Hugging Face incident, but restarted the main frontier RL run on August 28 under stricter controls.
    @reach_vbrecommended read 🔖
    @fouadmatinAstra is both significantly more capable and more token efficient compared to 5.6 Sol, as seen in port of ExploitBench. We plan to make Astra available soon, with its most advanced cybersecurity capabilities initially more limited, then iteratively expand access more broadly.