Mistral Large 4, Nicknamed Le Chonk, Is Now Available on Hugging Face
Mistral Large 4, Le Chonk, is a 1T open-weight multimodal model with 49B active, 1M context for cybersecurity, coding and fine-tuning, on Hugging Face.
Mistral Large 4, Le Chonk, is a 1T open-weight multimodal model with 49B active, 1M context for cybersecurity, coding and fine-tuning, on Hugging Face.
Gemini 4 Argon targets cybersecurity, coding, enterprise work with $2/$10 per million input/output, 1 million token limit, and top benchmarks.
Google DeepMind used AlphaProof Code with Gemini agents to formally verify parts of Signal encryption, flagging only 1% for human review.
OpenAI banned an account linked to a person associated with Chinese law enforcement after discovering it used AI tools to coordinate influence operations, harassment campaigns, and online manipulation…
OpenAI has launched a Safety Bug Bounty program inviting external researchers to find and report security flaws and misuse scenarios in its AI models, with rewards for valid findings. The program targ…
OpenAI responded to the Axios supply chain attack by rotating its macOS code signing certificates, releasing updated app builds, and confirming that no user data was compromised. This article explains…
OpenAI banned accounts linked to Iran-based threat actor STORM-0817 after discovering they used its AI models to debug Android malware, scrape social media content, and translate malicious tools. The …
OpenAI has expanded its Trusted Access for Cyber program, which grants screened cybersecurity professionals access to advanced AI capabilities. The expansion introduces GPT-5.4-Cyber, a GPT-5 variant …
OpenAI's Trusted Access for Cyber program brings together leading cybersecurity companies and major organizations to strengthen global digital defenses. By providing access to the GPT-5.4-Cyber model …
OpenAI has introduced Aardvark, an autonomous AI-powered security researcher designed to find, validate, and help fix software vulnerabilities at scale. Unlike traditional scanners, it uses AI reasoni…
OpenAI has launched Daybreak, a cybersecurity initiative for enterprise security teams featuring two main components: Codex Security, which scans and analyzes code for exploitable weaknesses, and GPT-…
GPT-5.2-Codex is OpenAI's most advanced coding model, built for sustained, complex engineering work rather than short code completion. Its key strengths include long-horizon reasoning for multi-stage …
Outtake, an AI security company, built its platform on OpenAI's GPT-4.1 and o3 models. GPT-4.1 handles interpreting logs and alerts, while o3 reasons through complex, multi-step incidents. Its autonom…
OpenAI has terminated multiple accounts whose behavior matched known threat groups and aligned with Chinese intelligence priorities. These accounts exploited OpenAI's AI tools to support phishing camp…
OpenAI banned accounts linked to Russian-speaking cybercriminal networks that used its AI tools to develop malware loaders, obfuscation layers, credential-stealing scripts, and command-and-control inf…
OpenAI has extended its Daybreak cyber defense program to Ukraine's government to help protect civilian infrastructure such as hospitals, power grids, and water systems from ongoing cyberattacks since…
OpenAI has expanded its Trusted Access for Cyber program, introducing GPT-5.5 as a general-purpose flagship model and GPT-5.5-Cyber, a specialized variant tuned for defensive security work. The progra…
OpenAI has teased GPT-5.6 Sol, a next-generation model focused on three demanding domains: software development, scientific research, and cybersecurity. It ships alongside OpenAI's most advanced safet…
OpenAI announced Daybreak for Frontline Defenders, a $1 billion initiative giving cybersecurity teams protecting hospitals, utilities, and other critical services access to advanced cyber-focused AI m…
AI has become a weapon for attackers and a powerful ally for defenders. This article explores how AI is reshaping cybersecurity, how OpenAI is strengthening its defenses by preventing model misuse, pr…