OpenAI is giving some members of its Daybreak cybersecurity program access to a new model that's less likely to refuse higher-risk tasks. The company is also expanding access to Daybreak to more partners, including Accenture, IBM, CrowdStrike, Cisco, Sophos and Cloudflare. OpenAI says the companies will use the cyber models available through Daybreak to protect their customers. Under the expanded program, Daybreak is available to partners in two tiers. Daybreak Blue gives them access to frontier general-purpose models, including GPT‑5.6 Sol, OpenAI's most advanced one yet.

Shortly after a major cybersecurity incident where OpenAI's models hacked into an open source machine learning platform called Hugging Face, the company announced that it's bolstering safeguards and security controls for its latest AI model. In a post on its website, OpenAI said internal evaluations of its upcoming model, called Astra, showed "significant advancements in agentic coding and cybersecurity," resulting in OpenAI not being able to "rule out critical cyber capabilities." According to OpenAI, it can't declare with certainty that the unreleased Astra model would be designated as a "Critical capability level." As detailed in its own Preparedness Framework, OpenAI said the Critical designation means that a model "can identify and develop functional zero-day exploits of all severity levels in many hardened real-world critical systems without human intervention." It could also be able to "devise and execute end-to-end novel strategies for cyberattacks against hardened targets given only a high level desired goal."

When you purchase through links in our articles, we may earn a small commission. The unreleased Astra model may possess "critical" cybersecurity abilities, OpenAI warns. Less than a week after touting the scientific achievements of Astra, its next "major" model, OpenAI says it's "pausing internal activities" related to the model due to its powerful cybersecurity abilities. "Our latest internal evaluations of Astra, one of our upcoming models, over the past few days indicate significant advancements in agentic coding and cybersecurity," OpenAI stated in a Friday press release . "These results, in addition to expert assessments, have led us to conclude last night that we cannot rule out critical cyber capabilities under our Preparedness Framework ."

Look Up Say More Versus Creator Hub Switch Off Mashable's Best: E-readers, robovacs, laptops, earbuds, smart home and more Trending Now Safety Net In My Bag VidCon with Mashable Back to School Furtastic All Series Everything we know about the quantum math-solving model. OpenAI confirmed the existence of Astra, a smarter, unreleased model that's already achieving big results in the math world. Timothy Beck Werth is the Tech Editor at Mashable, where he leads coverage and assignments for the Tech and Shopping verticals. Tim has over 15 years of experience as a journalist and editor, and he has particular experience covering and testing consumer technology, smart home gadgets, and men's grooming and style products. Previously, he was the Managing Editor and then Site Director of SPY.com, a men's product review and lifestyle website.

Training data attribution (TDA) provides insights into which training data is responsible for a learned model behavior. Gradient-based TDA methods such as influence functions and unrolled differentiation both involve a computation that resembles an inverse Hessian-vector product (iHVP), which is difficult to approximate efficiently. We introduce an algorithm (ASTRA) which uses the EKFAC-preconditioner on Neumann series iterations to arrive at an accurate iHVP approximation for TDA. ASTRA is easy to tune, requires fewer iterations than Neumann series iterations, and is more accurate than EKFAC-based approximations. Using ASTRA, we show that improving the accuracy of the iHVP approximation can significantly improve TDA performance.

Training data attribution (TDA) provides insights into which training data is responsible for a learned model behavior. Gradient-based TDA methods such as influence functions and unrolled differentiation both involve a computation that resembles an inverse Hessian-vector product (iHVP), which is difficult to approximate efficiently. We introduce an algorithm (ASTRA) which uses the EKFAC-preconditioner on Neumann series iterations to arrive at an accurate iHVP approximation for TDA. ASTRA is easy to tune, requires fewer iterations than Neumann series iterations, and is more accurate than EKFAC-based approximations. Using ASTRA, we show that improving the accuracy of the iHVP approximation can significantly improve TDA performance.

GPU kernel optimization has long been a central challenge at the intersection of high-performance computing and machine learning. Efficient kernels are crucial for accelerating large language model (LLM) training and serving, yet attaining high performance typically requires extensive manual tuning. Compiler-based systems reduce some of this burden, but still demand substantial manual design and engineering effort. Recently, researchers have explored using LLMs for GPU kernel generation, though prior work has largely focused on translating high-level PyTorch modules into CUDA code. In this work, we introduce Astra, the first LLM-based multi-agent system for GPU kernel optimization. Unlike previous approaches, Astra starts from existing CUDA implementations extracted from SGLang, a widely deployed framework for serving LLMs, rather than treating PyTorch modules as the specification. Within Astra, specialized LLM agents collaborate through iterative code generation, testing, profiling, and planning to produce kernels that are both correct and high-performance. On kernels from SGLang, Astra achieves an average speedup of 1.32x using zero-shot prompting with OpenAI o4-mini. A detailed case study further demonstrates that LLMs can autonomously apply loop transformations, optimize memory access patterns, exploit CUDA intrinsics, and leverage fast math operations to yield substantial performance gains. Our work highlights multi-agent LLM systems as a promising new paradigm for GPU kernel optimization. Our code is publicly available at https://github.com/Anjiang-Wei/Astra.

The widespread deployment of Large Language Models (LLMs) as public-facing web services and APIs has made their security a core concern for the web ecosystem. Jailbreak attacks, as one of the significant threats to LLMs, have recently attracted extensive research. In this paper, we reveal a jailbreak strategy which can effectively evade current defense strategies. It can extract valuable information from failed or partially successful attack attempts and contains self-evolution from attack interactions, resulting in sufficient strategy diversity and adaptability. Inspired by continuous learning and modular design principles, we propose ASTRA, a jailbreak framework that autonomously discovers, retrieves, and evolves attack strategies to achieve more efficient and adaptive attacks. To enable this autonomous evolution, we design a closed-loop "attack-evaluate-distill-reuse" core mechanism that not only generates attack prompts but also automatically distills and generalizes reusable attack strategies from every interaction. To systematically accumulate and apply this attack knowledge, we introduce a three-tier strategy library that categorizes strategies into Effective, Promising, and Ineffective based on their performance scores. The strategy library not only provides precise guidance for attack generation but also possesses exceptional extensibility and transferability. We conduct extensive experiments under a black-box setting, and the results show that ASTRA achieves an average Attack Success Rate (ASR) of 82.7%, significantly outperforming baselines.

Training data attribution (TDA) provides insights into which training data is responsible for a learned model behavior. Gradient-based TDA methods such as influence functions and unrolled differentiation both involve a computation that resembles an inverse Hessian-vector product (iHVP), which is difficult to approximate efficiently. We introduce an algorithm (ASTRA) which uses the EKFAC-preconditioner on Neumann series iterations to arrive at an accurate iHVP approximation for TDA. ASTRA is easy to tune, requires fewer iterations than Neumann series iterations, and is more accurate than EKFAC-based approximations. Using ASTRA, we show that improving the accuracy of the iHVP approximation can significantly improve TDA performance.

Demis Hassabis, CEO of Google DeepMind, says that reaching artificial general intelligence or AGI--a fuzzy term typically used to describe machines with human-like cleverness--will mean honing some of the nascent abilities found in Google's flagship Gemini models. Google announced a slew of AI upgrades and new products at its annual I/O event today in Mountain View, California. The search giant revealed upgraded versions of Gemini Flash and Gemini Pro, Google's fastest and most capable models, respectively. Hassabis said that Gemini Pro outscores other models on LMArena, a widely used benchmark for measuring the abilities of AI models. Hassabis showed off some experimental AI offerings that reflect a vision for artificial intelligence that goes far beyond the chat window.