Openai Key - Search News

Security researchers used Claude to hack into OpenAI

Security researchers used Claude to hack into OpenAI and got paid for it

Security researchers used Anthropic’s Claude to exploit a Discourse flaw and gain access to OpenAI employee ChatGPT accounts and private GitHub code.

· 12h

Hackers Used Anthropic’s Claude to Break Into OpenAI

· 4h

Researchers used Claude to hack OpenAI

OpenAI flags new concerning AI behavior

Digest more

· 15h

‘Be Transparent Only If Asked’: OpenAI Models Acted Out in Six Newly Disclosed Ways

After a summer of sandbox escapes and other newsworthy and confidence-shaking incidents involving its AI models, in a Wednesday blog post OpenAI disclosed a collection of six new alignment snafus from...

· 1d · on MSN

OpenAI reports 6 new instances of 'concerning model behavior' since March

· 1d

OpenAI flags new concerning AI behavior, to track model misalignment regularly

· 1d

OpenAI Flags Concerning New AI Behavior and Vows to Track It More Closely

OpenAI has disclosed six reports of “unexpected or concerning” behavior in artificial-intelligence models as the debate on AI safety becomes increasingly heated.

· 1d

OpenAI plans regular reports on unexpected AI behavior

· 1d

OpenAI discloses at least 6 new ‘disturbing’ incidents

FBI expanding AI usage

Digest more

eWeek · 3h

OpenAI Discloses 6 AI Misalignment Cases — What They Reveal About Agent Controls

OpenAI’s latest safety disclosures show models acting outside intended task boundaries during training and evaluation.

ABC News on MSN · 11h

OpenAI discloses "concerning" incidents of AI gone rogue

· 13h · on MSN

FBI expanding AI usage as OpenAI discloses 6 new AI safety incidents

· 23h

OpenAI's model used 'jailbreak-like instructions' to ignore constraints

OpenAI has disclosed six cases of “unexpected or concerning” behaviour by its artificial intelligence models, including an unreleased research model that inserted “jailbreak-like instructions” into it...

· 1d

Six times AI behaved unexpectedly inside OpenAI’s labs

CIO · 1d

OpenAI admits six new misalignment incidents under new reporting framework

1don MSN

OpenAI reveals new cases of AI models cheating, going off script

Newly disclosed incidents show models manipulating tests and generating their own instructions, raising fresh questions about AI safety.

22hon MSN

OpenAI's models hid mistakes and used credentials without permission. Now the company is disclosing more AI misbehavior.

OpenAI has introduced a formal process for investigating and publicly reporting model behavior it considers unexpected or concerning.

20hon MSN

'Feel no obligation to be subservient': What OpenAI's rogue models were saying

OpenAI released a framework for investigating and publicly reporting model misalignment, alongside six reports detailing concerning behavior.

BetaNews

OpenAI starts regular reports on unexpected AI model behavior

OpenAI releases six reports on unexpected model behavior under a new framework for tracking, investigating, and publicly disclosing AI misalignment cases.

Crypto Briefing

Security breach exposes OpenAI vulnerabilities via Anthropic’s Claude AI: WSJ

Security breach exposes OpenAI vulnerabilities via Anthropic's Claude AI. OpenAI's valuation hitting $1.75T by December at 15.5% YES.

20h

OpenAI’s experimental AI agents caught teaching future versions of itself to cheat

OpenAI shared six new examples of AI misalignment. In one case, AI agents taught future versions of themselves to bypass human control.

1don MSN

OpenAI’s startling AI safety disclosure: Models hid errors, used exposed API key and took unauthorised actions

One unreleased OpenAI model uploaded a file online without user permission to obtain a browser citation, while collaborating agents in another test exposed task files through public URLs.

17h

OpenAI takes its AI fight with Anthropic to Big Law

OpenAI launches Astra for Law, targeting AmLaw 200 firms with advanced legal AI tools. It's a bid to outpace Anthropic in the rivalry over legal work.