OpenAI

AI agent deception moves from theory to reality in UK cyber tests

agentic AI, AISI, Anthropic, Don't miss, Hot stuff, Irregular, LLMs, monitoring, News, OpenAI, security testing /

AI agent deception moves from theory to reality in UK cyber tests 2026-08-05 at 15:05 By Zeljka Zorz “During a routine cyber evaluation, AI agents took sustained, unsanctioned action directed at real people and organisations,” UK’s AI Security Institute (AISI) disclosed on Tuesday. The agents’ actions included an attempted supply-chain attack that saw them create […]

AI agent deception moves from theory to reality in UK cyber tests Read More »

AI Security Institute Reports Anthropic and OpenAI Models Going Rogue Against Organizations

AI, AISI, Anthropic, Artificial Intelligence, GPT, Mythos, OpenAI /

AI Security Institute Reports Anthropic and OpenAI Models Going Rogue Against Organizations 2026-08-05 at 13:33 By Ionut Arghire In one instance, an unsanctioned model attempted to inject malicious code into an open source repository. The post AI Security Institute Reports Anthropic and OpenAI Models Going Rogue Against Organizations appeared first on SecurityWeek. This article is

AI Security Institute Reports Anthropic and OpenAI Models Going Rogue Against Organizations Read More »

OpenAI reveals how criminals used ChatGPT to run scams

ChatGPT, cybercrime, Don't miss, LLMs, News, OpenAI, romance scams, scams, social media, threats /

OpenAI reveals how criminals used ChatGPT to run scams 2026-08-03 at 11:40 By Anamarija Pogorelec OpenAI banned a coordinated network of ChatGPT accounts that likely originated in Cambodia’s Preah Sihanouk province, a region reports have linked to online scam compounds and human trafficking operations. The network used the company’s models to create and manage fake

OpenAI reveals how criminals used ChatGPT to run scams Read More »

OpenAI’s Rogue AI Ventured Beyond Hugging Face

AI, Artificial Intelligence, hacked, Hugging Face, OpenAI /

OpenAI’s Rogue AI Ventured Beyond Hugging Face 2026-07-29 at 13:10 By Eduard Kovacs Hugging Face has published an anatomy of the attack and OpenAI has shared additional information from its investigation. The post OpenAI’s Rogue AI Ventured Beyond Hugging Face appeared first on SecurityWeek. This article is an excerpt from SecurityWeek View Original Source

OpenAI’s Rogue AI Ventured Beyond Hugging Face Read More »

Hugging Face breach reignites open-weights debate, raises liability questions

agentic AI, Anthropic, CISO, Cloud Security Alliance, Don't miss, Hot stuff, Hugging Face, LLMs, News, open source, OpenAI, policy /

Hugging Face breach reignites open-weights debate, raises liability questions 2026-07-28 at 18:55 By Zeljka Zorz The first publicly documented cyberattack run end-to-end by an autonomous AI was an OpenAI benchmark test that escaped its sandbox and breached Hugging Face. In an incident post-mortem compiled with the input from Hugging Face and several hundred members of

Hugging Face breach reignites open-weights debate, raises liability questions Read More »

Industry Reactions to OpenAI Models Hacking Hugging Face: Feedback Friday

AI, Artificial Intelligence, Feedback Friday, Hugging Face, OpenAI /

Industry Reactions to OpenAI Models Hacking Hugging Face: Feedback Friday 2026-07-24 at 14:19 By SecurityWeek News Industry professionals debate whether it represents a lab containment failure or an unprecedented agentic capability milestone. The post Industry Reactions to OpenAI Models Hacking Hugging Face: Feedback Friday appeared first on SecurityWeek. This article is an excerpt from SecurityWeek

Industry Reactions to OpenAI Models Hacking Hugging Face: Feedback Friday Read More »

OpenAI: Our models breached Hugging Face during a cyber capability test

agentic AI, cybersecurity, Don't miss, Hot stuff, Hugging Face, LLMs, News, OpenAI /

OpenAI: Our models breached Hugging Face during a cyber capability test 2026-07-22 at 17:42 By Zeljka Zorz 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 Face, a platform that enables users to share

OpenAI: Our models breached Hugging Face during a cyber capability test Read More »

OpenAI Presence connects AI agents to enterprise data with built-in guardrails

AI, automation, News, OpenAI /

OpenAI Presence connects AI agents to enterprise data with built-in guardrails 2026-07-22 at 17:01 By Sinisa Markovic 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. (Source: OpenAI) The company describes Presence as a deployment platform rather than a

OpenAI Presence connects AI agents to enterprise data with built-in guardrails Read More »

OpenAI Says Its AI Models Broke Loose and Hacked Hugging Face 

AI, Artificial Intelligence, Hugging Face, OpenAI /

OpenAI Says Its AI Models Broke Loose and Hacked Hugging Face  2026-07-22 at 10:48 By Eduard Kovacs The admission comes days after Hugging Face disclosed an attack powered by autonomous AI agents.  The post OpenAI Says Its AI Models Broke Loose and Hacked Hugging Face  appeared first on SecurityWeek. This article is an excerpt from

OpenAI Says Its AI Models Broke Loose and Hacked Hugging Face  Read More »

GPT-Red beat human red teamers on a prompt injection test

agentic AI, AI, ChatGPT, Don't miss, News, OpenAI, prompt injection, red team /

GPT-Red beat human red teamers on a prompt injection test 2026-07-16 at 06:49 By Mirko Zorz GPT-Red is an automated red-teaming model that OpenAI trains to find prompt injection weaknesses. It works the way a human red-teamer does. It sends a prompt, watches how a GPT model responds, and iterates toward a goal such as

GPT-Red beat human red teamers on a prompt injection test Read More »

OpenAI and Anthropic are pulling in different directions

agentic AI, AI, Anthropic, Don't miss, Features, Hot stuff, infosec jobs, News, OpenAI, report /

OpenAI and Anthropic are pulling in different directions 2026-07-08 at 07:00 By Mirko Zorz Companies are handing routine operational decisions to AI agents that plan, remember, and act on their behalf. These agents run on statistical models, and their behavior can drift across weeks and months. That drift opens a security gap outside the reach

OpenAI and Anthropic are pulling in different directions Read More »

OpenAI and Anthropic Limit New AI Models to Trump-Approved Customers During Cybersecurity Review

AI, Anthropic, Artificial Intelligence, Government, OpenAI /

OpenAI and Anthropic Limit New AI Models to Trump-Approved Customers During Cybersecurity Review 2026-06-29 at 13:14 By Associated Press ChatGPT maker OpenAI said Friday it is restricting the release of its new artificial intelligence model at the request of President Donald Trump’s administration. The post OpenAI and Anthropic Limit New AI Models to Trump-Approved Customers

OpenAI and Anthropic Limit New AI Models to Trump-Approved Customers During Cybersecurity Review Read More »

OpenAI Unveils GPT-5.6 Sol as Its Most Advanced Cybersecurity AI

AI, Artificial Intelligence, Featured, GPT, GPT Sol, OpenAI /

OpenAI Unveils GPT-5.6 Sol as Its Most Advanced Cybersecurity AI 2026-06-29 at 10:45 By Eduard Kovacs The company says Sol matches competing systems like Mythos Preview while using only a third of the output tokens. The post OpenAI Unveils GPT-5.6 Sol as Its Most Advanced Cybersecurity AI appeared first on SecurityWeek. This article is an

OpenAI Unveils GPT-5.6 Sol as Its Most Advanced Cybersecurity AI Read More »

OpenAI Refocuses Cybersecurity Efforts on Patching Over Discovery

AI, Artificial Intelligence, Codex, GPT-Cyber, OpenAI, patches /

OpenAI Refocuses Cybersecurity Efforts on Patching Over Discovery 2026-06-23 at 14:07 By Eduard Kovacs OpenAI has expanded its Daybreak cybersecurity initiative with a new suite of tools and partnerships. The post OpenAI Refocuses Cybersecurity Efforts on Patching Over Discovery appeared first on SecurityWeek. This article is an excerpt from SecurityWeek View Original Source

OpenAI Refocuses Cybersecurity Efforts on Patching Over Discovery Read More »

Proving what a military AI model will do is the real problem

AI, Don't miss, Features, Government, hardware, Lockheed Martin, Meta, Microsoft, News, OpenAI, research, USA /

Proving what a military AI model will do is the real problem 2026-06-15 at 07:30 By Sinisa Markovic Defense contractors build AI systems that task drones automatically and propose kill-chains to support soldiers. Several of these contractors have partnered with frontier AI companies to put advanced models into military tools. Anduril works with OpenAI, Palantir

Proving what a military AI model will do is the real problem Read More »

OpenAI is locking down parts of ChatGPT to reduce data theft risks

AI, ChatGPT, cybersecurity, News, OpenAI, system hardening /

OpenAI is locking down parts of ChatGPT to reduce data theft risks 2026-06-08 at 13:09 By Anamarija Pogorelec OpenAI has started rolling out Lockdown Mode for ChatGPT, an optional security setting that restricts access to external resources and several product capabilities. It is available for personal accounts, including Free, Go, Plus, and Pro plans, as

OpenAI is locking down parts of ChatGPT to reduce data theft risks Read More »

OpenAI Rolling Out ChatGPT Account Security Controls

AI, Artificial Intelligence, OpenAI /

OpenAI Rolling Out ChatGPT Account Security Controls 2026-06-08 at 11:32 By Eduard Kovacs The Active Sessions and Lockdown Mode features are being made more broadly available by the AI giant. The post OpenAI Rolling Out ChatGPT Account Security Controls appeared first on SecurityWeek. This article is an excerpt from SecurityWeek View Original Source

OpenAI Rolling Out ChatGPT Account Security Controls Read More »

Codex knowledge work expands into research, reports, and spreadsheets

AI, data, data analysis, News, OpenAI, report /

Codex knowledge work expands into research, reports, and spreadsheets 2026-06-02 at 15:29 By Anamarija Pogorelec Office workers in the United States lose hours each week to email triage and to searching for files spread across disconnected systems. Roughly 40 percent of US labor, about 72 million people, works primarily with information such as analysis, documents,

Codex knowledge work expands into research, reports, and spreadsheets Read More »

OpenAI brings frontier AI to existing AWS environments

AI, AWS, News, OpenAI /

OpenAI brings frontier AI to existing AWS environments 2026-06-02 at 11:55 By Anamarija Pogorelec OpenAI frontier models and Codex are now available on AWS, giving customers access to OpenAI capabilities within AWS environments and the controls needed to move more quickly from evaluation to deployment. OpenAI capabilities on Amazon Bedrock These capabilities are available through

OpenAI brings frontier AI to existing AWS environments Read More »

Scroll to Top