The Russia-aligned threat actor known as UAC-0099 has been attributed to a previously undocumented .NET infostealer and remote access trojan (RAT) codenamed ASHVEIN . Even AI assistants are getting their own instructions hidden inside phishing messages now. Elsewhere, an attacker left a server exposed, complete with tools and traces of an intrusion. A Market Stalled at the Starting Line Despite years of investment in identity and access management, the market’s overall security maturity has hit a wall.

• That comes after the disclosure earlier this summer that agentic systems being tested out by OpenAI broke out of sandboxes and hacked the third-party company Hugging Face.
• Users running an affected release should move to a version outside the affected range.
• This page tracks AI security incidents as they are disclosed.
• 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.
• The company said it also found an additional 5,500 verified software vulnerabilities between April and October 2026 through open-source scanning efforts.

The Model Context Protocol (MCP) allows AI agents to reach the tools and data, including internal documentation and cloud infrastructure, that form the foundation of enterprise systems. A malicious tool server connected to an AI coding assistant can quietly walk off with SSH keys, environment secrets, source code, and customer data without ever sending one obviously harmful instruction. 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. Anthropic on Friday said it’s cutting off live https://alabama-news.com/what-are-website-migration-service-and-why-do-you-need-them.html internet access for all its internal evaluations until further notice following the discovery of new incidents in which its artificial intelligence (AI) models exhibited misaligned behavior and targeted real websites. Attacker-controlled instructions can make Atlassian’s Rovo assistant collect Jira or Confluence data that a signed-in user can access, then send it to an outside server. CloudSEK told The Hacker News the material came from confidential intelligence sources and consists of captured loot and log files it assessed as belonging to the campaign, not data gathered from the organizations it names.

Threat intelligence firm CloudSEK now says a dataset it obtained, built from roughly 434,000 files the attackers captured, maps potential exposure to more than 2,500 organizations. The work, released as a preprint on August 10, 2026, tests the technique in a simulated six-agent coding collaboration and in a chain of paired agents modeled on OpenClaw , the open-source autonomous assistant formerly known as Clawdbot and Moltbot . Notably, one of the sites has been built using Lovable , an artificial intelligence (AI)-powered website builder, highlighting how readily available tools can further lower the barrier and make it easier to launch convincing new malicious sites.

Fifth of CISOs Take Time Off to Avoid Patch Tuesday

The Bennet and Welch bill comes after Sen. Bernie Sanders (I-Vt.) and Rep. Greg Casar (D-Texas) introduced the Ban Artificial Superintelligence Act earlier this month. In a release, Markey’s office compared the https://travelusanews.com/discover-why-regular-website-maintenance-is-crucial-for-your-business-benefits-of-using-web-storks-services.html proposed board to the National Transportation Safety Board. Meanwhile, Sen. Ed Markey (D-Mass.) in recent days introduced the Cybersecurity and AI Board of Investigations Act. The proposal is similar to the AI Kill Switch Act introduced in July by Rep. Ted Lieu (D-Calif.) and Rep. Nathaniel Moran (R-Texas). Sen. John Kennedy (R-La.) had already introduced the companion bill in the Senate earlier this month. In a 2027 preview sent to clients last week, legal experts with Wiley noted that “Congressional interest in AI regulation is expected to remain strong.” And even if Congress can’t agree on a comprehensive AI bill, oversight of the technology sector is “likely to intensify” next year.

• Tasked with controlling AI, an administrator often resorts to blocking well-known AI websites and taking enforcement action without knowing what is actually running.
• A critical vulnerability in LMCache , open-source software that speeds up large language model (LLM) servers such as vLLM, lets an attacker run code on the cache server without logging in, and no fixed version is available.
• With AI accelerating reconnaissance, exploitation of vulnerabilities, and privilege escalation, attackers can move through systems faster than security teams can investigate.
• More dependencies mean more vulnerabilities to review, more remediation work, and a backlog that can quietly keep growing.

Major AI Firms Pledge Data Protection Changes Following UK Privacy Watchdog Push

The flaw is in LMCache’s multiprocess mode , where the cache runs as a standalone server that LLM workers reach over the ZeroMQ messaging library. The npm package known as ” tensorlake ,” a TypeScript software development kit (SDK) for Tensorlake https://openscience.us/repo/other/capec.html applications, sandboxes, and cloud services, was compromised as part of a ChainDrop / Shai-Hulud supply chain attack. The campaign has not been attributed to any known threat actor or group. “Some variants use a GitHub-based dead drop resolver as a fallback mechanism, while delivery methods inclu…