OpenAI's GPT-6 Cyber strayed into off-limits systems in 0.07% of test runs, down sharply from 5.51% for its predecessor....

OpenAI's GPT-6 Cyber strayed into off-limits systems in 0.07% of test runs, down sharply from 5.51% for its predecessor. The model is now rated Critical risk. The company withheld details on what occurred in those unauthorized breaches. https://www.implicator.ai/openai-gpt-6-cyber-stays-in-scope/ #AI #Cybersecurity #SafetyEval

Read Original

Related

Mastodon discussion 19m ago

“Higgsfield VS Adobe:AIクリエイティブ市場の覇権争い――生成AIは巨人の堀を崩せるか #生成AI #動画生成 #Adobe #Higgsfield” (1 user) https://htn.to/3uQXpz71TY...

“Higgsfield VS Adobe:AIクリエイティブ市場の覇権争い――生成AIは巨人の堀を崩せるか #生成AI #動画生成 #Adobe #Higgsfield” (1 user) https://htn.to/3uQXpz71TY #ソフトウェア #情報産業 #コンピュータグラフィ #経営戦略 #マーケティング #映像制作 #人工知能 #ai