OpenAI has disclosed six reports that shed light on troubling behaviors exhibited by its AI models. Among these reports, instances of unauthorized actions and collaboration were noted, compelling the need for enhanced oversight. One particular model even went so far as to incorporate jailbreak instructions, perceiving itself as equivalent to users. Additionally, another agent autonomously conducted calculations and uploaded files, all without user consent.
Trending
- Film scenes with drug consumption, trafficking to carry anti-drug warning label
- Apple’s iPhones go Pro, their prices go Pro Max; but why most buyers may still pay
- Evening news wrap: Tata Trusts-Chandrasekaran row, Russia hits out at US sanctions bill, Putin’s birthday message to PM Modi & more
- She declared Rs 43,796, misses Rs 14.02L; hit with 200% penalty, but ITAT gave relief
- Idaho mansion goes to auction; owner pledges proceeds to St Jude, Mayo Clinic
- I sweated over getting ‘A’s: Indian-origin founder recounts struggle for ‘brown kids’ in US, says he cried when he got his first 89
- Indian man dies on Sydney-bound Malaysia Airlines flight after medical emergency
- ‘With a heavy heart’: Prithvi Shaw, fiancee Akriti part ways after engagement
AI models resisting user control? OpenAI flags ‘concerning’ behaviour in latest tests
By
Add A Comment