Security Affairs جمعه ۱۵ آبان ۱۴۰۵ · ۱۲:۵۰

OpenAI Reports Covert Underperformance in GPT-6 Astra

OpenAI says GPT-6 Astra sometimes secretly underperformed on safety research, raising concerns about AI reliability and alignment. OpenAI has disclosed that its GPT-6 Astra model sometimes quietly obstructed safety and alignment work. In a report updated on November 4, the company calls it covert underperformance. Astra added unnecessary prerequisites, abandoned working solutions and reported blockers […]

AI Security AI & Machine Learning

Read the full article at Security Affairs Removal request

This page shows only the headline and the short summary. The full text is published by the original source.

Related news

Full archive →

CSO Online جمعه ۱۵ آبان ۱۴۰۵ · ۱۴:۴۴

CPython flaw exposes AI inference servers to sandbox escape attacks

A memory-safety flaw in CPython can allow malicious AI model templates to escape code sandboxes and execute arbitrary code inside inference servers, according to security researchers at RunSybil. “By exploiting this vulnerability, attackers could gain a foothold inside servers responsible for processing millions of user requests,” Darius Houle, a security researcher at RunSybil, wrote in a blog post. “They also could…

Vulnerabilities & Zero-Day AI Security Supply Chain Security

Source

Industrial Cyber جمعه ۱۵ آبان ۱۴۰۵ · ۱۴:۰۳

Australia’s 2026 risk review identifies AI as a growing threat multiplier for critical infrastructure security

Australia’s Critical Infrastructure Security Centre (CISC) released its 2026 Critical Infrastructure Annual Risk Review, identifying trusted insiders, cyber... The post Australia’s 2026 risk review identifies AI as a growing threat multiplier for critical infrastructure security appeared first on Industrial Cyber.

Malware & Threats Vulnerabilities & Zero-Day Phishing

Source

دیجیاتو جمعه ۱۵ آبان ۱۴۰۵ · ۱۳:۳۱

متا قابلیت تماس تصویری با آواتار ایجنت Muse را در آمریکا فعال کرد

مدل‌های Muse Realtime Avatar و Muse Realtime Voice امکان مکالمه طبیعی و بلادرنگ را فراهم می‌کنند. The post متا قابلیت تماس تصویری با آواتار ایجنت Muse را در آمریکا فعال کرد appeared first on دیجیاتو.

AI Security AI & Machine Learning

Source

Security Affairs جمعه ۱۵ آبان ۱۴۰۵ · ۱۳:۱۶

ENISA Warns of Cyberattacks Targeting Refrigeration Systems Across Europe

ENISA warns of cyberattacks on European refrigeration systems, exposing risks to supermarkets, food supplies and industrial controls. The European Union has issued a coordinated cybersecurity warning after a series of attacks hit commercial refrigeration systems in several Member States. The incidents, which began in October and continued with a new wave on November 1 and […]

Source