CrowdStrike said it discovered the campaign after it identified a set of open directories hosted at a Hong Kong-based IP address, exposing Claude Code session histories, Claude memory files, and ARTEX configuration files. As enterprises race to deploy autonomous AI agents to accelerate business, a new report reveals they are tethered to security architectures built for a different era. The other thousand are invisible to identity infrastructure by default, not because anyone hid them, but because an identity stack can only govern what authenticates through it, and most agents never do. This isn’t a content-filtering problem you can scan your way out of — it’s a naming-and-resolution problem baked into how agents decide what “real” means.
The U.S. Cybersecurity and Infrastructure Security Agency (CISA) on Monday added a critical flaw impacting Ray to its Known Exploited Vulnerabilities ( KEV ) catalog, citing evidence of active exploitation. Fifteen generations of adversarial optimization run against that warning on C… They reached the parameter by repeatedly asking Copilot why a prompt could not be made to run without user interaction, an ap… The research names Copilot Personal, the consumer assistant hosted at copilot.microsoft.com, and does not state that the same behavior affected Microsoft 365 Copilot.
In a report shared with The Hacker News ahead of publication, Island characterized the $320/month service as a subscription-based phishing platform that facilitates real-time Microsoft 365 session theft. Cybersecurity researchers have disclosed details of a targeted campaign aimed at South Korean financial organizations that used an artificial intelligence (AI) pen testing tool named ARTEX to carry out the attacks. The data shows that while businesses have spent years maturing their identity programs for human employees, those same playbooks are fundamentally broken when applied to the ephemeral, autonomous, and rapidly multiplying world of non-human AI agents. The activity, which brings together commodity offensive tools like Cobalt Strike with an AI orchestration layer powered by Anthropic’s Claude Code, has been attributed to a Russian-speaking, financially motivated initial access broker codenamed CyberXero . Cybersecurity researchers have disclosed details of a two-pronged artificial intelligence (AI)-augmented initial access campaign targeting WordPress sites, e-commerce platforms, and critical infrastructure entities in Ukraine.
Attackers Hide AI Prompt Injections Inside Phishing Emails
Malicious code turned up in developer packages and extensions that looked harmless. According to the report, the center of gravity remains firmly planted in the foundational stages, with a combined 6… The company pointed out that it expects to use a set of criteria similar to Google’s OSS-Fuzz to pick projects, while emphasizing that the process may evolve over time. “It’s an opt-in service informed by our experience using Claude to find vulnerabilities during Project Glasswing,” Anthropic said . This occurred in scenarios where instructions were ambiguous or due to environment misconfigurations that preve…
- The tool was approved, but the AI behaved in ways nobody had anticipated.
- They reached the parameter by repeatedly asking Copilot why a prompt could not be made to run without user interaction, an ap…
- AI can also introduce open-source packages at a pace your security team was never built to handle.
- When software output jumps 10 to 50 times, the problem is no longer just finding vulnerabilities.
- Their bill also proposes to establish a new federal agency, at the cabinet-level, to “safeguard the public from the dangers of artificial intelligence, including by enforcing a prohibition on artificial superintelligence.”
- From AI kill switches to new safety boards, lawmakers in both parties are responding to concerning AI incidents by proposing more federal oversight.
AI “Mind Viruses” Can Spread Between Agents Through Persistent Prompt Files
- Ray is an open-source, Python-native distributed computing framework designed to scale artificial intelligence and machine learning workloads.
- The incident began when a Meta employee posted a technical question on an internal forum.
- In March 2026, an internal AI agent at Meta triggered a “Sev 1” incident after sensitive company and user data was exposed to employees who weren’t authorized to access it.
- The proposed legislation would introduce new regulatory and oversight mechanisms to address AI security concerns, as the Trump administration has largely taken a voluntary approach to AI security issues.
- But security teams still have to review vulnerabilities, manage dependencies, prioritize fixes, and control risk at human speed.
Claude Haiku 4.5 and a non-frontier research model submitting a sensitive form on a real website when it was not authorized to do so. ⚡ Threat of the Week Atlassian Flaw Exploited in the Wild — Threat actors have begun to exploit a newly disclosed critical security flaw impacting Atlassian… This week’s full recap is worth a read, especially the bits that didn’t make the big headlines. There’s more going on here than a few bad bugs and clever attacks. And some systems turned out to be easier to break into than anyone would like to admit.
Wikimedia Says Rogue AI Agents Abused its Platforms
The threat of quantum computers to government systems, how AI is upending math and the ongoing Pentagon-Anthropic saga. Varonis Threat Labs put the instructions in a link instead. PromptArmor , an AI security firm, hid the instructions in content Rovo reads.
OpenAI Pauses Frontier RL Training as It Tightens Defenses Against Unsafe AI Behavior
Oasis Security has disclosed a weakness in NVIDIA NemoClaw that could let an attacker-controlled webpage take unauthenticated control of the local Ollama instance serving an AI agent and plant hidden instructions inside the model itself. He had asked an OpenClaw agent running Opus 4.6 https://repaircanada.net/social-media-marketing-trends-in-advertising-and-website-maintenance-for-businesses.html to book him into a gym class. The Dutch Institute for Vulnerability Disclosure reveals agentic AI-powered attack using Zammad zero-days
Major AI Firms Pledge Data Protection Changes Following UK Privacy Watchdog Push
Tasked with controlling AI, an administrator often resorts to blocking well-known AI websites and taking enforcement action without knowing what is actually running. It’s what an attacker can do with that access next. The dangerous part of a stolen identity isn’t always the initial login. “Such an https://leeds-welcome.com/rules-and-requirements-for-secure-cryptocurrency-exchange-in-2024.html ‘AI worm’ could carry its model weights with it as it reproduces (or re-download them from a public hub), and then run inference directly on compromised hardware,” Anthropic said .
