OpenAI pauses work on upcoming AI model Astra over cybersecurity risks
OpenAI paused development of its upcoming AI model Astra after the system was found to be significantly more adept at cybersecurity tasks than expected. CEO Sam Altman said the release is delayed for safety due to cyber capabilities, and OpenAI flagged a possible critical cybersecurity risk and tightened controls.
Detected & updated continuously · Source: Nebula
Story subjects
Sources
@blckchaindaily
🚨 OPENAI CEO SAM ALTMAN SAYS ASTRA RELEASE DELAYED FOR SAFETY DUE TO CYBER CAPABILITIES, HOPES DELAY IS NOT TOO LONG
@business
OpenAI is pausing some internal work around one of its upcoming artificial intelligence models to implement stricter safeguards after the system was found to be significantly more adept at cybersecurity tasks. https://t.co/Cy9Sjx9Zox
@Reuters
OpenAI flags possible critical cybersecurity risk in upcoming model, tightens controls https://t.co/8vzvTBXLXT https://t.co/8vzvTBXLXT
@notjazii
OpenAI just paused development of their next model astra is too good to release out for public but i still think they gonna announce it next week with limited access and make it available for public after US govt approves what do you think? https://t.co/vD0tamB7Ov
@jumperz
OpenAI just acknowledged Astra.. its upcoming model and it sounds seriously powerful... early tests say that it can find unknown security flaws and handle complex cyber tasks with barely any human help.. Could this be openAI’s version of mythos? I hope they don’t follow https://t.co/jruJqw2Pyd