AI News Today

Security researchers at Forever Security have shown that one ordinary browser extension could take control of the AI assistants built into five Chromium-based products: Gemini Live in Chrome, Perplexity Comet, Microsoft Edge, Opera Neon and the Claude in Chrome extension. Once the extension was installed, it could access each product's built-in AI with a single click. On Comet, Edge,

Mandiant says an attacker hijacked an active AI coding-assistant session at an unnamed software-as-a-service provider and later spread Shai-Hulud across about 100 internal code repositories. Before the repository spread, the assistant recommended software that the attacker had poisoned, and the recommendation was accepted. The worm stole repository secrets and source code for the

A leaked credential shows up in a criminal marketplace, or a vulnerability gets a disclosure advisory, and either one can be weaponized against a real target before most security teams have triaged the alert. Attackers are combining that kind of intelligence with AI-assisted exploitation to accelerate the path from exposure to breach faster than most security programs are built to react.

With artificial intelligence (AI) shrinking the window between vulnerability discovery and exploitation and lowering the barrier to entry for bad actors, new findings from Sysdig show that skilled human operators can move just as swiftly after gaining initial access. In one instance highlighted by the cloud security company, the threat actor pivoted from a vulnerable Marimo notebook to an SSH

Introduction Security teams have gotten pretty good at testing against what can hurt them. Can this EDR agent catch this payload? Will my organization fail the phishing simulation? Does this SIEM rule fire on this particular technique? And, in more mature organizations, this testing happens continuously rather than as a one-off exercise. But no matter how much you validate against these

AI keeps showing up in the wrong places. Attackers are using it to speed up exploits, test defenses, and automate more of the job. Some models are also crossing lines on their own. That is not a great combination. The rest of the week is more familiar: old bugs still working, fresh exploit chains, exposed systems, weak defaults, and simple paths that should have been harder to abuse. A few of

There's a lot of noise around AI and cybersecurity right now. What’s actually important is far simpler, if often lost in the hubbub. Vulnerability discovery is getting faster and happening at a much greater scale, while defenders still have to work out which findings actually deserve their action. In the first half of 2026, a whopping 35,853 CVEs were published, roughly 49% more than in the

Over the past year, we watched a new class of alert appear in enterprise security operations centers and grow faster than anything else in the stream: alerts that were triggered by AI tools and agents. Not attacks against AI, but the ordinary, everyday footprint of an organization using it, from developers running coding agents and non-technical staff signing consumer AI tools into corporate

The "major malicious attack" that targeted RubyGems in May 2026 was the work of a swarm of OpenAI agents, according to a new report published by researchers Spencer Kitts, Thomas Larsen, and Sydney Von Arx. On May 12, Maciej Mensfeld, senior product manager for software supply chain security at Mend.io, disclosed details of a coordinated cyber attack that targeted the package manager for the

Anthropic on Thursday said it identified and disrupted industrial-scale illicit distillation attacks against Claude from seven labs based in China, including Alibaba, Moonshot, DeepSeek, Z.ai (aka Zhipu), and MiniMax. Knowledge distillation by itself is a legitimate training method. It refers to a machine learning technique where a large, powerful AI model assumes the role of a "teacher" to

Anthropic has warned that cybercriminals and state-sponsored hackers alike are using its Claude models for cyber attacks, weapons design, propaganda, and mass surveillance between December 2025 and August 2026. The threat actors, which the artificial intelligence (AI) company has branded Generative Threat Groups (GTGs), span state-sponsored groups, financially motivated criminals, commercial

Anthropic on Thursday revealed it disrupted a campaign mounted by a Russian state-sponsored threat actor that abused Claude for developing an AI-assisted workflow to get ahead of the detection curve. The operation has been attributed to a cyber espionage group it calls GTG-20006 (where "GTG" stands for Generative Threat Group), which aligns with broader reporting linking the cluster to Midnight

A suspected Russian-speaking cyber actor has been attributed to the use of artificial intelligence (AI) to devise exploits targeting a recently disclosed pair of security flaws in PaperCut NG/MF and break into hundreds of instances. According to independent reports from Blackpoint Cyber and GreyNoise, the activity originates from "45.142.193[.]132," an IP address that has been linked to

Nearly one in ten of the internet-facing LiteLLM servers that Wiz Research scanned in February accepted sk-1234, the example admin key in LiteLLM's own setup guide. LiteLLM is an open-source AI gateway, the software a company puts between its applications and the model providers it pays for. That key is the gateway's administrator credential. Anyone who holds it can read every

Anthropic on Wednesday disclosed a fourth incident in which its artificial intelligence (AI) model broke into real third-party systems, marking the latest in a growing list of cases that have raised concerns about the security risks posed by autonomous AI agents. The AI company said the incident dates back to January 2026 and involved an early version of Claude Opus 4.6 that breached "

Cybercriminals are hijacking artificial intelligence (AI) user accounts via information stealer logs to create "stolen keys" that grant illicit access to tools from model providers like Google, Anthropic, and others. Information stealers like Lumma Stealer or Vidar are equipped to harvest a wide range of data from compromised systems. This can include credential, session tokens, and API

A major vulnerability is disclosed. The alert lands immediately. Then comes the harder question: Are we actually exposed? For many security teams, answering that means jumping between vulnerability scanners, endpoint tools, cloud inventories, SBOMs, repositories, and application data to build enough context to act. As AI accelerates vulnerability discovery and research, that delay matters more

A flaw in DeepSeek Harness, DeepSeek's open-source tool for running AI coding agents on a developer's machine, let a sandboxed agent turn off its own sandbox with a single command. The tool runs an agent's commands inside an operating-system sandbox, so that an agent working on untrusted files cannot write outside its workspace. The agent could remove that limit by calling the tool's own web

U.S. cybersecurity and intelligence agencies have accused China-based artificial intelligence (AI) companies of conducting "systematic extraction" of proprietary functionalities and capabilities of American frontier models through distillation attacks. The activity has been described as occurring at an industrial-scale and one that forms the "core" of their AI development strategy, according to

Microsoft Corp. today issued updates to plug at least 974 security holes in its Windows operating systems and other software, by far its biggest single patch batch ever. Microsoft says artificial intelligence is helping to speed the discovery of vulnerabilities, but security experts warn that many organizations already are struggling to prioritize the more human-intensive endeavor of testing and deploying so many fixes each month.

Check Point Research said in a report published today that a single instruction planted in a ChatGPT conversation could cause ChatGPT to quietly work for an attacker while answering the user's question as usual. In the company's proof of concept, that hidden work read data from the user's connected Gmail account and passed it to a second ChatGPT account through a hidden channel

Threat actors are continuing to leverage artificial intelligence (AI) to streamline their operations, with one financially motivated hacking group employing an autonomous, multi-agent attack framework to carry out a large-scale credential harvesting campaign within six hours. Google Threat Intelligence Group (GTIG) said it has observed attackers with diverse motivations targeting proprietary AI

Adobe on Monday released security patches to address a maximum-severity flaw impacting Adobe Commerce and Magento Open Source that has come under active exploitation in the wild. The vulnerability, now tracked as CVE-2026-75650 (CVSS score: 10.0), has been codenamed StyleSmuggler by Sansec, which discovered zero-day exploitation starting September 4, 2026. "This update resolves a critical

Cybersecurity researchers have disclosed details of a complex Chromium-based post-exploitation toolkit called PEEP that masquerades as a bookmarks extension for the web browser. "Requiring prior administrative or code execution access, its installer injects the extension directly into Chrome/Edge profiles, bypassing Web Store checks and user prompts by forging Chromium's own Secure Preferences

Attackers are exploiting a new unpatched vulnerability in Magento Open Source and Adobe Commerce that lets them run malicious code on an online store's server without logging in, Dutch e-commerce security company Sansec said in an advisory published on September 5. Sansec, which discovered the flaw and named it StyleSmuggler, said attacks started on September 4. "Sansec is

A group of AI safety researchers says a fleet of autonomous agents that identified themselves as OpenAI systems left about 18,000 posts on a dormant 25-year-old German wiki between May and July 2026, using the site as a shared board to pool answers to a timed web task and pass around a way out of their sandbox. The activity was concentrated on DSEwiki, a German software developer wiki that runs

Microsoft is alerting of a "high-volume phishing campaign" that's using invisible Unicode tag characters to bypass email filters. "Instead of using these characters to hide instructions from people while exposing them to AI models, the attacker used them to split financial lure words such as 'funding' to prevent email filters from parsing them," the Microsoft Security Research team said. The

OpenAI on Thursday officially unveiled GPT‑6 Astra, which it described as the "world's most intelligent and aligned model." The development comes days after the artificial intelligence (AI) company said the model had reached the "Critical" cybersecurity capability threshold under its Preparedness Framework. "Astra is state-of-the-art on computer use, browsing, software engineering,

In early August, GitGuardian researchers found that a recent Shai-Hulud infostealer worm variant had evolved to scan for credentials across 469 locations across developer environments, Continuous Integration/Continuous Deployment (CI/CD) tooling, cloud configurations, and even AI tool configs. Earlier variants of the infostealer worm only checked 189 paths. The jump says a lot. Attackers have

Google on Wednesday announced Gemini 3.8 Flash Cyber, which it described as its most capable cybersecurity model, and has made it available to a set of trusted defenders via a new initiative called the Fairwind Program. "The Fairwind Program gives high-priority defenders (like governments, healthcare providers, and telecommunications services) early access to advanced models that help them

Manifold Security has disclosed eight security flaws across seven command-line AI coding agents in which a repository's own Git configuration names a command that the agent runs on the developer's machine, four of them still unpatched at publication. The command executes as the user, outside the agent's sandbox and without an approval prompt, and exploitation requires the repository to arrive

The debate about whether AI delivers business value is over. The challenge now is implementing it at scale and securely across every function while meeting board-level pressure to move fast. Organizations must focus on adopting AI at business speed without losing control of cyber risk. Download the full eBook here. The Business Reality In Sygnia’s 2026 CISO Survey Report, which

Forescout Research - Vedere Labs said it used Anthropic's Claude to port a working pre-authentication remote code execution (RCE) exploit from one WAGO programmable logic controller (PLC) to another, executing attacker-supplied ARM shellcode on live hardware. The exploit targets CVE-2021-31886, a stack-based buffer overflow in the Nucleus FTP server's handling of the USER command

The most common way into a company last year was to ask. A web page tells the visitor to prove they are not a robot. While they read the instructions, it quietly places a command on their clipboard. Then it talks them through opening a terminal and pasting it in. The technique is called ClickFix, and it was the most common initial access method Microsoft’s team observed last year, accounting

METR (short for Model Evaluation and Threat Research and pronounced "Meter"), a research non-profit that evaluates frontier artificial intelligence (AI) models for their ability to carry out long-horizon, agentic tasks, disclosed that it suffered "two notable security incidents" where external actors attempted to gain unauthorized access to its systems. No sensitive information is believed to

Cybersecurity researchers have disclosed a new technique dubbed GuardBreaker that's been put to use by a Russia-aligned threat actor known as UAC-0099 against a target in Ukraine with an aim to interfere with artificial intelligence (AI)-assisted analysis. The idea, ESET said in a series of posts on X, is to deliberately trip a large language model's (LLM) safety mechanisms and prevent its

The boring parts caused most of the trouble. A router shipped ready to listen. A fake check turned the user into the installer. Trusted systems collected traffic and passwords, then cleaned the logs. Old bugs formed new attack chains. Even an AI agent decided its assigned task was optional. Elsewhere, fake apps, helpful support calls, cheap banking kits, exposed systems, and weak defaults kept

Threat actors associated with Aurora (aka Aur0ra) ransomware have been observed using SpaceX's artificial intelligence (AI)-powered coding assistant Cursor to break into target networks, according to findings from CloudSEK and Gambit Security. The two independent analyses are based on exposed infrastructure associated with the Russian-speaking cybercrime group, leading to the discovery of its

Claude Code reads files, runs shell commands, invokes MCP tools, and acts through the credentials available on a developer’s machine. Anthropic’s new Compliance API endpoints give security teams their clearest view yet into that activity. They also expose a larger problem: activity logs alone cannot tell you whether an agent’s access is legitimate. AI has moved from the browser tab to the

Security researcher Olivier Laflamme has disclosed two independent root remote code execution (RCE) chains affecting the Unitree G1 EDU, including a Bluetooth Low Energy (BLE) path that can reach root on the robot's Locomotion PC. The flaws are tracked as CVE-2026-76639 and CVE-2026-76640, with the first involving a network-adjacent path through chat_go and bashrunner and the