#ai safety

// News

29.07.2026 :: Ing. Calogero Bono

FAR.AI test finds 448 jailbreaks on Grok and 249 on Gemini while Claude and GPT remain immune to automated attacks

A new report from the nonprofit organization FAR.AI has revealed how easy it is to bypass the safety guardrails of some...

read →

// News

12.07.2026 :: Ing. Calogero Bono

OpenAI loses head of safety Johannes Heidecke in reorganization of research and safety teams

OpenAI has announced an internal reorganization that involves its safety and research teams, with the departure of Johan...

read →

// News

01.07.2026 :: Pietro Maiorana

Anthropic Adds New Security Measure for Fable 5 After Trump Administration Restrictions

Anthropic has agreed to implement a new safeguard for its Fable 5 and Mythos 5 AI models, following restrictions imposed...

read →

// News

26.06.2026 :: Pietro Maiorana

OpenAI Releases GPT 5.6 in Staged Manner at White House Request Over Security Fears

OpenAI has completed development of GPT 5.6, its most advanced language model to date, but the rollout will not follow t...

read →

// News

26.06.2026 :: Meteora Web

OpenAI to release ChatGPT 5.6 only to government-approved customers

OpenAI has announced that its new AI model, ChatGPT 5.6, will not be available to everyone at once. According to a repor...

read →

// News

08.05.2026 :: Ing. Calogero Bono

OpenAI Revolutionizes Voice API, Safety, and Browser with Codex for Chrome

OpenAI has just released a hat trick of innovations that redefine how developers, businesses, and users interact with ar...

read →