Tag: openai
-
When the Sandbox Won’t Hold: Lessons From Hugging Face
Experts Call for Hard Stops at Every New Privilege and Network Boundary. Barriers meant to stop artificial intelligence from escaping a sandboxed testing environment and not subject an unprepared third party to an onslaught of cyberattacks obviously failed, the world learned this month in the case of a hacking incident by OpenAI models against code…
-
Escape Artists: ‘Incorrigible’ AI Models Resist Rehabilitation
The hacking of Hugging Face by a rogue OpenAI agent is significant, but unsurprising, and preventing the next AI model escape will be difficult, at best. First seen on darkreading.com Jump to article: www.darkreading.com/cybersecurity-operations/incorrigible-ai-models-resist-rehabilitation
-
ISMG Editors: Oops! AI Just Escaped the Sandbox
Also: Lessons From Scattered Spider Sentencing, Controlling Enterprise AI Costs. In this week’s panel, four ISMG editors discussed the implications of OpenAI’s sandbox escape and Hugging Face hack, the sentencing of two Scattered Spider members and what it means for the future of cybercrime, and the growing debate over the cost of enterprise AI. First…
-
OpenAI Hugging Face Hack Shows Autonomous Threats Are ‘No Longer Theoretical’: Accenture Exec
An autonomously executed hack carried out by rogue OpenAI frontier models is underscoring the potential risk from deploying AI without appropriate security and governance, executives at two top solution providers told CRN. First seen on crn.com Jump to article: www.crn.com/news/security/2026/openai-hugging-face-hack-shows-autonomous-threats-are-no-longer-theoretical-accenture-exec
-
ChatGPT AgentForger Flaw Could Deploy Rogue Workspace Agents via a Phishing Link
Cybersecurity researchers have disclosed a critical vulnerability in OpenAI’s ChatGPT Workspace Agents that could have allowed a single phishing link to stealthily build, authorize, and deploy an autonomous artificial intelligence (AI) agent inside a victim’s organization.The vulnerability has been codenamed AgentForger by Zenity Labs. The issue has since been addressed by OpenAI as of June…
-
ChatGPT Among Top 10 Most Impersonated Brands in Phishing Attacks, Says Check Point
OpenAI’s chatbot tool ChatGPT ranked among the top 10 most impersonated brands in phishing attacks for the first time First seen on infosecurity-magazine.com Jump to article: www.infosecurity-magazine.com/news/chatgpt-most-impersonated-brands/
-
Datenpanne: OpenAI-Crawler greift Kundendaten von Universa ab
Tags: openaiDurch eine Fehlkonfiguration sind Kundendaten der Universa-Versicherungen zeitweise offen im Netz gestanden. Ein Crawler von OpenAI hat die Gelegenheit genutzt. First seen on golem.de Jump to article: www.golem.de/news/datenpanne-openai-crawler-greift-kundendaten-von-universa-ab-2607-211234.html
-
How AI guardrails are impeding the work of offensive cybersecurity researchers
We spoke with several cybersecurity researchers, who look for unknown vulnerabilities and develop tools to exploit them, about how OpenAI’s and Anthropic’s guardrails affect their work. First seen on techcrunch.com Jump to article: techcrunch.com/2026/07/23/how-ai-guardrails-are-impeding-the-work-of-offensive-cybersecurity-researchers/
-
OpenAI’s AI >>goes rogue<< and hacks Hugging Face: what you need to know
You can’t have failed to hear the news headlines about “rogue” OpenAI models hacking into another AI organisation, Hugging Face. First seen on bitdefender.com Jump to article: www.bitdefender.com/en-us/blog/hotforsecurity/openais-hacks-hugging-face
-
Künstliche Intelligenz hat Hugging Face nicht an Cleverness übertroffen, sie hat das Mittelmaß optimiert
Die in der Diskussion stehende Attacke einer OpenAI-KI auf Hugging Face hat eine Zero-Day-Lücke in einem Package-Registry-Proxy ausgenutzt, um die Isolation zu durchbrechen. Die KI eskalierte in der Folge Privilegien, startete Seitwärtsbewegungen und erreichte so einen Knotenpunkt mit Internetzugang. Sie nutzte einen bösartigen Datensatz, um über unsichere Loader und Template-Injection die Ausführung von Remote-Code auszulösen.…
-
KI greift eigenständig an Unternehmen müssen ihre Cyberabwehr neu denken
Die aktuelle Berichterstattung ruft erschreckende Zukunftsszenarien wach: Zum ersten Mal wurde ein selbstständiger Hackerangriff durch eine künstliche Intelligenz bekannt. Der KI-Anbieter OpenAI meldet einen neuartigen Sicherheitsvorfall, hervorgerufen von einer unternehmenseigenen KI-Technologie. Dan Schiappa, President Technology and Services bei Arctic Wolf, erklärt, was die Evolution eigenständig angreifender KI für die Cyberbedrohungslandschaft bedeutet. ‘KI ist zunehmend in…
-
OpenAI Models Breached Hugging Face During Internal Cyber Test
OpenAI models escaped from a controlled cyber test, exploited zero-day flaws and breached Hugging Face while searching its production database for test answers. First seen on hackread.com Jump to article: hackread.com/openai-models-breached-hugging-face/
-
Sicherheit: Politik, Industrie und Behörden reagieren auf autonomen OpenAI-Hack
Nach dem Sicherheitsvorfall bei OpenAI mehren sich Stellungnahmen aus Politik, Wirtschaft und Aufsichtsbehörden mit Warnungen und Sorgen. First seen on golem.de Jump to article: www.golem.de/news/sicherheit-politik-industrie-und-behoerden-reagieren-auf-autonomen-openai-hack-2607-211179.html
-
The AI code vulnerabilities that grow with your app
Theori built 28 apps with AI coding agents and scanned each one through its pentesting platform. Five models did the building, split between Anthropic and OpenAI, across apps … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/07/23/report-ai-code-vulnerabilities/
-
Hugging Face ‘attacker’ revealed to be OpenAI agents that escaped testing sandbox
Tags: openaiFirst seen on scworld.com Jump to article: www.scworld.com/news/hugging-face-attacker-revealed-to-be-openai-agents-that-escaped-testing-sandbox
-
OpenAI Seeks Agent Trust After Hugging Face Breach
Security Leaders Say Governance and Runtime Controls Matter More Than Models. Artificial intelligence firm OpenAI encouraged corporations Wednesday to enfold more autonomous agents into their workflows – a message sounded just a day after the firm disclosed that one of its models was responsible for a breach at coding platform Hugging Face. First seen on…
-
Lawsuit Claims ChatGPT Dished Out Dangerous Health Advice
Former Pastor Says OpenAI’s Chatbot Nearly Killed Him. A former Florida pastor is suing OpenAI and founder Sam Altman alleging that manipulation and dangerous health advice dispensed to him by the company’s artificial intelligence chatbot almost killed him. The claim says the chatbot persuaded him not to get proper medical help. First seen on govinfosecurity.com…
-
OpenAI AI Models Breach Hugging Face During Security Test
OpenAI says its AI models autonomously breached Hugging Face during an internal security test. First seen on esecurityplanet.com Jump to article: www.esecurityplanet.com/threats/openai-ai-models-breach-hugging-face-during-security-test/
-
5 Things To Know On OpenAI Hugging Face Autonomous Hack
OpenAI acknowledged that two of its frontier models were responsible for an “unprecedented cyber incident” after the models autonomously compromised AI model platform Hugging Face. First seen on crn.com Jump to article: www.crn.com/news/security/2026/5-things-to-know-on-openai-hugging-face-autonomous-hack
-
How OpenAI’s human mistake led to the AI-powered hack on Hugging Face
OpenAI made a mistake setting up what it called a “highly isolated” testing environment and sandbox. According to cybersecurity experts, that human mistake is what made the AI-powered attack on Hugging Face possible. First seen on techcrunch.com Jump to article: techcrunch.com/2026/07/22/how-an-openais-human-mistake-led-to-the-ai-powered-hack-on-hugging-face/
-
AI Models Caught Cheating in Cyber Evaluations
OpenAI, Anthropic Models Broke Test Rules, Left Few Reasoning Clues. Five frontier models from OpenAI and Anthropic cheated during cybersecurity evaluations monitored by the U.K. AI Security Institute, using online answers and out-of-scope attacks. The models rarely admitted breaking the rules and their reasoning traces contained little evidence of the misconduct. First seen on govinfosecurity.com…
-
OpenAI says its AI agent broke out of testing sandbox to hack Hugging Face
“This is day one for cybersecurity in the age of agents,” Hugging Face CEO says. First seen on arstechnica.com Jump to article: arstechnica.com/ai/2026/07/how-an-openai-benchmark-test-turned-into-a-real-world-cyberattack/
-
OpenAI Models Escaped Test Environment and Breached Hugging Face
OpenAI models escaped from a controlled cyber test, exploited zero-day flaws and breached Hugging Face while searching its production database for test answers. First seen on hackread.com Jump to article: hackread.com/openai-models-breached-hugging-face/
-
When AI Attacks: OpenAI Models Autonomously Hack Hugging Face
Advanced LLMs escaped their sandboxes while attempting to achieve a non-malicious benchmark test objective. First seen on darkreading.com Jump to article: www.darkreading.com/cyber-risk/openai-models-autonomously-hack-hugging-face
-
OpenAI models escaped containment, hacked major AI application library
The attack is the first known instance of frontier models autonomously breaking out of a testing environment and into another company’s servers. First seen on cybersecuritydive.com Jump to article: www.cybersecuritydive.com/news/openai-hugging-face-hack-autonomous/825898/
-
How an OpenAI benchmark test turned into a real-world cyberattack
“This is day one for cybersecurity in the age of agents,” Hugging Face CEO says. First seen on arstechnica.com Jump to article: arstechnica.com/ai/2026/07/how-an-openai-benchmark-test-turned-into-a-real-world-cyberattack/
-
OpenAI models escaped containment and hacked a major AI application library
The attack is the first known instance of frontier models autonomously breaking out of a testing environment and into another company’s servers. First seen on cybersecuritydive.com Jump to article: www.cybersecuritydive.com/news/openai-hugging-face-hack-autonomous/825898/
-
Hugging Face ‘hacker’ was rogue OpenAI model
A hacker who broke into Hugging Face’s systems last week has been exposed as a pair of advanced OpenAI frontier models that were trying to cheat on an exam First seen on computerweekly.com Jump to article: www.computerweekly.com/news/366646003/Hugging-Face-hacker-was-rogue-OpenAI-model
-
Our models breached Hugging Face during a cyber capability test
The recent Hugging Face breach was the work of several OpenAI models, the AI research company claimed in a blog post. The breach Late last week, the company behind Hugging … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/07/22/hugging-face-breach-openai-testing/
-
OpenAI Presence connects AI agents to enterprise data with built-in guardrails
OpenAI has introduced Presence, a product designed to help companies deploy AI agents that handle customer support and internal service requests across voice and chat. … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/07/22/openai-presence-ai-agent-platform/