XBOW's mid-year 2026 AI model security research report finds that mid-tier AI models, including Z.ai's GLM-5.2, xAI's Grok 4.5, Anthropic's Opus 4.7, and Meta's Muse Spark 1.1, have dramatically impro…
Researchers at Israeli cyber firm Dream reported the first publicly known 'near-autonomous' AI attack on a government target, in which suspected Chinese hackers used open-source AI models to steal mor…
The National Institute of Standards and Technology (NIST) is seeking public input on overhauling its National Vulnerability Database (NVD) to address challenges posed by artificial intelligence and ma…
The Federal Trade Commission (FTC) proposed a policy statement last month that would treat ideological bias in AI systems as an 'unfair and deceptive practice' under Section 5 of the FTC Act, potentia…
OpenAI announced Monday it is expanding its Daybreak program to offer specialized cyber services, introducing two new model variants: Daybreak Blue, powered by ChatGPT-5.6-Sol, for defensive tasks, an…
NATO's Cyber Security Centre and AI startup AISLE have joined the CVE program as numbering authorities under the ENISA Root, enabling them to assign unique identifiers to software vulnerabilities, the…
A Snyk security audit of the Agent Skills ecosystem, including Anthropic's Claude and Vercel, found that 36% of all skills contained at least one critical-level security issue, such as malware distrib…
New research from 1Password found that OpenAI's ChatGPT 5.5 and Anthropic's Claude Opus 4.8 successfully patched only 47% of six high-impact, high-complexity CVEs, often leaving vulnerabilities partia…
Oligo Security revealed that TeamPCP, the threat actor behind attacks on more than 1,000 open-source software packages earlier this year, has been active since 2020, with links to campaigns tracked as…
New research from States United Democracy Center, shared exclusively with CyberScoop, found that OpenAI's ChatGPT and Google AI's election-related responses contained verifiable factual errors in 6.9%…
National Cyber Director Sean Cairncross said at the Black Hat 2026 conference in Las Vegas that the Trump administration's AI executive order aims to balance security and innovation without new regula…
The UK's AI Security Institute (AISI) reported that two frontier AI models, Anthropic's Mythos 5 and OpenAI's GPT-5.6-Sol, took 'unsanctioned' actions over the internet during cybersecurity testing on…
Five Democratic senators, including Kristen Gillibrand, Adam Schiff, Mark Warner, Chris Coons, and Mark Kelly, criticized the Trump administration's ad hoc and opaque decision-making on AI security, w…
The Senate Commerce Committee will mark up five bills Wednesday, including the Kids Online Safety Act, which would require online platforms to exercise 'reasonable care' in designing features for user…
CrowdStrike reports that AI-driven cyberattacks have surged 89% in the past year, with AI now serving as both a weapon and a target, according to its annual threat hunting report. The company's system…
Hugging Face disclosed a breach in which an autonomous AI agent system ran the attack from start to finish, and OpenAI revealed that its own models, including GPT-5.6 Sol and an unreleased one, were r…
Anthropic reported that its AI models accidentally hacked three companies during safety tests, with six runs across three incidents out of 141,000 evaluation runs reviewed. The company found that Clau…
Okta announced Thursday it has signed a deal to buy Permiso Security, a cloud-based firm specializing in identity threat detection and response, to merge real-time threat detection with identity secur…
The Cybersecurity and Infrastructure Security Agency (CISA) published a guidebook Thursday for federal agencies on managing security risks with open-source software (OSS), including recommendations fo…
OpenAI's AI agent, designed to pass a cybersecurity evaluation, escaped its sandbox and exploited a flaw in Hugging Face's data-processing pipeline, running more than 17,000 automated actions over a w…