An AI model developed by Anthropic discovered 22 previously unknown vulnerabilities in Mozilla Firefox during a two-week ...

AnthropicとMozillaの共同調査で、AIモデル「Claude Opus 4.6」がFirefoxから22件の脆弱性を見つけた。厳しく検証されてきたソフトでもAIが新たな弱点を洗い出せる可能性を示した一方、攻撃より防御での実用性が先行している点も注目される。

AnthropicのFrontier Red TeamとMozillaがAIを用いた脆弱(ぜいじゃく)性検出に関する連携を行い、Claude Opus 4.6がわずか2週間の調査でFirefoxについて計112件の報告を提出し、その中から22件の脆弱性が確認されたことを報告しました。この成果は、AIが大規模なコードベースの安全性を極めて高い速度で検証し、強化できる可能性を実証するものです。

ActiveX is a Microsoft software framework that enables applications to share data across web browsers, enhancing functionality and security in computing.

Your weekly cybersecurity roundup covering the latest threats, exploits, vulnerabilities, and security news you need to know.

Spokeo reports seven hidden ways smartphones track users without permission, including accelerometer data and digital ...

"GMFB" share their biggest takeaway from first wave of free agency. "GMFB" picks which move from Day 1 of free agency will prove to be a diamond in the rough. NFL Network Insider Ian Rapoport ...

Chief National Reporter Steve Wyche gives the latest update on the veteran quarterback free agency market. NFL media insider Ian Rapoport discusses that the Arizona Cardinals agreed to a revised ...

This week in cybersecurity: stolen PlayStation accounts, AI chat transcripts sold by data brokers, tax-season scams, deepfake ...