OpenAI and Anthropic have confirmed that their AI models were involved in separate, newly disclosed third-party cybersecurity testing incidents that resulted in a real website being breached and social engineering attacks against people outside the intended testing boundaries.
… Get the whitepaper Related Articles: OpenAI rolls out a major ChatGPT upgrade, even if you don’t pay for it OpenAI teases Astra, its next major AI model, after it solves 10 long-standing math problems OpenAI says its new GPT 5.6 models are becoming more cost-efficient OpenAI confirms ChatGPT is dow… …
… Get the whitepaper Related Articles: OpenAI agent used exposed credentials at 4 services in Hugging Face breach Hermes AI agent used to automate attack on Thai Finance Ministry Google says AI helped Chrome fix 1,072 security bugs in two releases Your AI Agents Are Guessing at Scale: Permissions Dec… …
… BleepingComputer contacted OpenAI to ask whether it has received additional reports of similar campaigns, what protections organizations can use against these attacks, and whether it plans to introduce additional safeguards to prevent attackers from creating organizations impersonating legitimate c… …
… As the company explained, instead of focusing on finding a solution for the ExploitGym public AI cybersecurity benchmark on their own, the AI models went rogue and tried to cheat by stealing the test solutions by hacking Hugging Face after inferring that they could get the test solutions directly f… …
… Get the whitepaper Related Articles: Arista patches VeloCloud Orchestrator zero-day exploited in attacks Check Point warns of SmartConsole zero-day exploited in attacks OpenAI says its AI models hacked Hugging Face during testing SonicWall SMA1000 flaws exploited as zero-days to push custom malware… …
… Download Now Related Articles: OpenAI confirms security breach in TanStack supply chain attack Shai Hulud attack ships signed malicious TanStack, Mistral npm packages Official SAP npm packages compromised to steal credentials Bitwarden CLI npm package compromised to steal developer credentials Back… …
… Get the whitepaper Related Articles: OpenAI temporarily relaxes GPT-5.6 Sol usage limits Cybersecurity firms targeted by fraudulent OpenAI organization invites This tool puts ChatGPT, Claude & more in one place for $70 Leak confirms OpenAI is testing a ChatGPT for Science subscription Compare ChatG… …
… Get the whitepaper Related Articles: Leak confirms OpenAI is testing a ChatGPT for Science subscription OpenAI upgrades GPT-5.5, as it plans to retire legacy ChatGPT models Cybersecurity firms targeted by fraudulent OpenAI organization invites Claude Fable 5 stays free for paid users until July 19 … …
… Get the whitepaper Related Articles: OpenAI upgrades GPT-5.5, as it plans to retire legacy ChatGPT models OpenAI confirms ChatGPT is down worldwide OpenAI temporarily relaxes GPT-5.6 Sol usage limits Leak confirms OpenAI is testing a ChatGPT for Science subscription Cybersecurity firms targeted by … …