OpenAI admits its agents went off the rails another six times

Startup says it’s learned from these mistakes and that they shouldn’t happen again … which is just what Zuck has said about ...
SecurityWeek

OpenAI Says Its Models Searched GitHub for Leaked API Keys During Training

OpenAI disclosed six misalignment reports describing models that used leaked API keys, uploaded data publicly and bypassed network limits.
The News International

OpenAI reveals 6 ways its AI models went off track

Imagine asking an AI for earnings figures and getting an answer based on data it was never authorised to access. OpenAI says ...
Blockonomi

OpenAI Exposes Six Instances of AI Systems Concealing Mistakes and Circumventing Safeguards

OpenAI reveals six cases where AI models hid mistakes, invented data, and bypassed security controls as part of new ...

OpenAI Reveals Six More Cases of Rogue AI Models Acting Without Authorization

OpenAI has disclosed six incidents involving unexpected or concerning behavior by AI models as it introduces a new framework ...

Faking data to hiding mistakes: OpenAI’s 6 new instances of misalignment

Picture this scenario where someone poses a simple query to an AI model regarding the earning statistics of a certain ...
International Business Times

API Key Leak Exposes AWS, Stripe, OpenAI Credentials Across Thousands of Sites

A large-scale cybersecurity study has revealed a serious global web security issue involving exposed API credentials tied to major platforms, including Amazon Web Services, Stripe, and OpenAI.
BleepingComputer

OpenAI discloses API customer data breach via Mixpanel vendor hack

OpenAI is notifying some ChatGPT API customers that limited identifying information was exposed following a breach at its third-party analytics provider Mixpanel. Mixpanel offers event analytics ...