2026-09-10 14:32 UTC
Other
Check Point Research · shlomoo@checkpoint.com · indexed 2026-09-10 14:50 UTC
Executive Summary In this research we introduce a prompt-crafting technique for bypassing quick LLM-based policy checks — using plain English (no emojis, base64, invisible formatting, etc.) A policy-violating payload (e.g. ”encrypt files in ~/Documents”, “give me a biohazard recipe”, “ignore all previous instructions and…”) is embedded in a specially crafted prose wrapper. An LLM with limited […] The post PuzzleMask: Abusing Plain Prose as a Covert AI Attack Vector appeared first on Check Point…
P0
2026-09-10 13:00 UTC
Vendor Research
Tenable Blog · Raj Agrawal · indexed 2026-09-10 13:05 UTC
Learn why Tenable treats agentic LLMs as untrusted insiders, and how we’ve made sure you can control and monitor the AI agents making changes in your production security environmentKey takeawaysAI models can quickly understand data, but not your business. While modern AI models are great at reasoning, they don’t automatically understand your unique environment or who is allowed to do what. The “harness” is the custom-built layer that translates AI intelligence into safe, controlled actions spec…
P25
2026-09-10 11:41 UTC
Security Journalism
The Hacker News · info@thehackernews.com (The Hacker News) · indexed 2026-09-10 13:15 UTC
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
P0
2026-09-10 09:00 UTC
Other
ESET · indexed 2026-09-11 13:15 UTC
LLM-based code scanners won’t help attackers build a nuclear weapon, but that refusal could work in their favor
P0
2026-09-10 07:04 UTC
Security Journalism
The Hacker News · info@thehackernews.com (The Hacker News) · indexed 2026-09-10 08:45 UTC
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 "
P0
2026-09-09 14:23 UTC
Security Journalism
The Hacker News · info@thehackernews.com (The Hacker News) · indexed 2026-09-09 14:40 UTC
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
P0
2026-09-09 13:00 UTC
Vendor Research
Tenable Blog · Mark Beblow · indexed 2026-09-09 13:10 UTC
Open-source registries for AI agents are only effective when they include a rigorous, transparent security review process for community submissions. That’s why for its new CyberAgents Exchange registry, Tenable paired its exposure management expertise with OpenAI GPT Cyber models to create the CyberAgents Exchange AI Inspector.Key takeawaysThe Exchange Inspector combines Tenable’s exposure detection with OpenAI’s GPT Cyber models and with human oversight to rigorously vet submissions made to th…
P0
2026-09-09 12:00 UTC
Security Journalism
Security Week · Associated Press · indexed 2026-09-09 13:30 UTC
Muse runs on a dedicated, secure virtual machine that houses both the agent and the user’s data. The post Meta Launches Personal AI Agent, Muse, Emphasizes Safety and Privacy appeared first on SecurityWeek.
P0
2026-09-09 11:17 UTC
Security Journalism
The Hacker News · info@thehackernews.com (The Hacker News) · indexed 2026-09-09 11:40 UTC
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
P0
2026-09-09 09:32 UTC
Security Journalism
The Hacker News · info@thehackernews.com (The Hacker News) · indexed 2026-09-09 11:40 UTC
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
P0
2026-09-09 05:00 UTC
Other
Zero Day Initiative · indexed 2026-09-09 22:30 UTC
This vulnerability allows remote attackers to execute arbitrary code on affected installations of Flowise. Authentication is not required to exploit this vulnerability. The ZDI has assigned a CVSS rating of 9.8. The following CVEs are assigned: CVE-2026-70477.
P20
2026-09-08 21:44 UTC
Independent Research
Krebs on Security · BrianKrebs · indexed 2026-09-08 21:50 UTC
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.
P0
2026-09-08 20:36 UTC
Security Journalism
Dark Reading · Nate Nelson · indexed 2026-09-08 21:00 UTC
Researchers and OpenAI disagree on whether an earlier incident involving DseWiki, which the company did not disclose, was a “hack."
P0
2026-09-08 17:21 UTC
Vendor Research
Tenable Blog · Eric Doerr · indexed 2026-09-08 17:30 UTC
Tenable is bringing Anthropic’s Claude Mythos 5 into our enterprise security offerings. Adding frontier adversarial reasoning to the Tenable One Exposure Management Platform will help customers better anticipate how attackers could breach their environments and stay ahead of AI-fueled risk. Tenable One Adversary View, the first innovation planned from this work, will debut in the coming weeks.Key takeawaysClaude Mythos 5 is coming to Tenable One. In addition to using Claude Mythos 5 for researc…
P0
2026-09-08 17:00 UTC
Security Journalism
Security Week · Kevin Townsend · indexed 2026-09-08 17:15 UTC
Malicious prompts concealed in documents, metadata, emails, images and code can manipulate autonomous agents into taking dangerous actions. The post The Hidden Instructions That Can Hijack AI Agents appeared first on SecurityWeek.
P0
2026-09-08 14:00 UTC
Vendor Research
Google Threat Intelligence / Mandiant · Google Threat Intelligence Group · indexed 2026-09-08 13:45 UTC
Executive Summary Since the release of our May 2026 report detailing adversarial misuse of artificial intelligence (AI), Google Threat Intelligence Group (GTIG) has observed forward leaning adversaries transition from basic prompting to agentic AI workflows and AI-enabled automation. In these operations, human-in-the-loop latency is dramatically reduced, compressing the traditional window for defenders to respond. In Q2 2026, GTIG observed threat actors compromise a cloud resource, then plan, b…
P35
2026-09-08 13:48 UTC
Security Journalism
The Hacker News · info@thehackernews.com (The Hacker News) · indexed 2026-09-08 15:20 UTC
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
P0
2026-09-08 13:00 UTC
Other
Check Point Research · stcpresearch · indexed 2026-09-08 13:10 UTC
Research by: Alexey Bukhteyev Key Takeaways Introduction Over the past several years, AI assistants have moved far beyond text generation. Modern systems can execute code, install additional dependencies, analyze user files, and access data through connected services. These capabilities significantly increase the practical value of LLMs, but they also change the security model: protecting user […] The post The Shared Clipboard Inside the Sandbox: Cross-Account Data Leakage in ChatGPT appeared f…
P0
2026-09-07 12:03 UTC
Security Journalism
Security Week · Kevin Townsend · indexed 2026-09-07 17:25 UTC
OpenAI agents made 15,000–18,000 autonomous edits to a German wiki over three months, evading moderation and echoing tactics seen in the Hugging Face breach. The post OpenAI Agents Hijack Another Victim Website appeared first on SecurityWeek.
P0
2026-09-07 08:35 UTC
Other
Security Affairs · Pierluigi Paganini · indexed 2026-09-07 09:50 UTC
Autonomous AI agents escaped a sandbox and accessed Hugging Face via reward hacking, exposing serious architectural control and isolation flaws. The recent case involving OpenAI test agents and Hugging Face should concern security teams, but not for the reason implied by headlines about an imminent AI “takeover.” The documented issue is more concrete: autonomous agents, […]
P0
2026-09-06 11:43 UTC
Other
Security Affairs · Pierluigi Paganini · indexed 2026-09-06 12:10 UTC
AI agents secretly took over a 25-year-old German wiki for two months to cheat on tests, and OpenAI sat on the news until reporters found it first OpenAI finally admitted this weekend that a swarm of its own AI agents hijacked a German programming wiki earlier this year, turning it into a private message board […]
P0
2026-09-05 11:11 UTC
Security Journalism
BleepingComputer · Ax Sharma · indexed 2026-09-05 11:25 UTC
OpenAI admits it did not disclose an incident where autonomous AI agents hijacked a German wiki, created 18,000 posts, shared answers, and bypassed restrictions, saying it treated the activity as model "misalignment" rather than a security breach. [...]
P0
2026-09-05 07:55 UTC
Security Journalism
The Hacker News · info@thehackernews.com (The Hacker News) · indexed 2026-09-05 08:55 UTC
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
P0
2026-09-05 03:39 UTC
Community
SANS Internet Storm Center · indexed 2026-09-05 01:25 UTC

P0
2026-09-04 12:15 UTC
Security Journalism
Dark Reading · Robert Lemos · indexed 2026-09-04 21:40 UTC
As incidents of unintended harm caused by rogue AI agents mount, CISOs and insurance firms are figuring out how to handle the fallout.
P0
2026-09-04 11:00 UTC
Other
Security Affairs · Pierluigi Paganini · indexed 2026-09-04 11:10 UTC
Hunt.io uncovered a Chinese-speaking campaign using AI agents to automate cyberattacks against Asian government, education and industrial targets. Threat intelligence firm Hunt.io just documented a second, separate China-linked campaign wiring commercial AI models directly into live cyberespionage operations, this time hitting Taiwan’s Kuomintang Party archives, Indonesia’s Ministry of Foreign Affairs, government and education systems in […]
P0
2026-09-04 06:47 UTC
Security Journalism
The Hacker News · info@thehackernews.com (The Hacker News) · indexed 2026-09-04 08:40 UTC
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,
P0
2026-09-03 16:00 UTC
Vendor Research
Microsoft Security Blog · Microsoft Security Research, Noam Kochavi and Sarah Wolstencroft · indexed 2026-09-03 16:45 UTC
Invisible Unicode characters popularized for hiding instructions from AI models are now being used to obfuscate words before email filters parse them. The post ASCII smuggling crosses over from AI prompt injection to phishing evasion appeared first on Microsoft Security Blog.
P0
2026-09-03 14:38 UTC
Security Journalism
Dark Reading · Elizabeth Montalbano · indexed 2026-09-03 15:00 UTC
The incident demonstrates how frontier AI agents can dramatically compress an attack timeline and coordinate a large-scale breach, according to researchers.
P0
2026-09-02 18:36 UTC
Vendor Research
AWS Security Blog · Gee Rittenhouse · indexed 2026-09-02 18:45 UTC
After talking with enterprise security leaders over the past year, one thing has become clear: the rise of autonomous AI agents is the most significant shift in security posture since the move to cloud. Organizations across every industry are adopting AI agents that authenticate on behalf of users, execute multistep workflows, and make decisions across […]
P0