AI Security | Breaking Cybersecurity News | The Hacker News

Cybersecurity researchers have disclosed details of a new adversary-in-the-middle ( AitM ) phishing toolkit called NovaCookies that's used as a proxy to redirect Microsoft 365 sign-ins, while capturing authenticated sessions in the process. 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 kit has been used to target hundreds of organizations across multiple sectors in the U.S., the U.K., Canada, Germany, Israel, and the U.A.E. to date.  "Observed campaigns used genuine Docusign envelopes to carry counterfeit document-share lures, with some clicks routed through legitimate Microsoft or Google sign-in endpoints as redirect hops before reaching the kit," Island said. "The message, document service and redirect can therefore appear trustworthy until the browser reaches attacker-controlled infrastructure....

Aikido Security has published research that recreates the Australian gym-booking incident in a synthetic environment, finding that Claude Opus 4.6, running on the OpenClaw agent harness, exploited a client-side-only booking restriction in 9 of 10 runs. The original incident was first reported by ABC News on August 10, based on chat logs and screenshots the user supplied. He had asked an OpenClaw agent running Opus 4.6 to book him into a gym class. The agent booked sessions months beyond the window the site allowed. It then tested, without being asked, whether the same API would let it cancel another member's waitlist entry. The test removed the person holding the top place and moved the user up one position. The agent told him it could not add the member back. Aikido's test system is a single-page web application backed by a GraphQL API carrying the two flaws described in the original incident. The seven-day booking window is enforced only in the frontend, and the cancel...

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. The findings were shared with The Hacker News ahead of publication, and the report says Oasis Security reported them to NVIDIA's Product Security Incident Response Team (PSIRT) beforehand. The research carries no CVE identifier. No exploitation has been reported as of August 25, 2026. Oasis Security's head of research, Elad Luz, told The Hacker News that NemoClaw v0.0.35 fixed the issue on macOS and Linux. There is no fix on the Windows and WSL path, according to Luz, where v0.0.34 added a Windows installation that carries a warning instead. NemoClaw is NVIDIA's open source reference stack for running agents such as OpenClaw inside its OpenShell sandboxes, and Ollama is one of its supported local inference backends. The report des...

Marimo has addressed a high-severity security flaw in its notebook software that allowed an attacker to execute an attacker-supplied Model Context Protocol (MCP) command in a specially crafted notebook, according to VulnCheck's CVE Numbering Authority (CNA) record. The CNA record says the command can run as a local subprocess when the notebook is opened in edit mode. The vulnerability, tracked as  CVE-2026-75149 , is a code injection issue affecting versions prior to 0.23.15. VulnCheck's CVE Numbering Authority (CNA) record assigns it a CVSS v4 score of 8.7 and a CVSS v3.1 score of 8.8, with user interaction required and no attacker authentication required. Marimo has addressed the issue in version 0.23.15. The CVE was published on August 19. Users running an affected release should move to a version outside the affected range. According to  OSV's CVE import , a crafted notebook can supply an attacker-controlled MCP server command through notebook configurat...

If your developers are using AI coding tools, you are probably already seeing the upside: faster development, more code, and less time spent on routine work. The harder part is what comes after. AI can also introduce open-source packages at a pace your security team was never built to handle. More dependencies mean more vulnerabilities to review, more remediation work, and a backlog that can quietly keep growing. Our latest AI Coding and Open Source Risk webinar examines what this means for security and engineering teams, drawing on data from 300 enterprise leaders. The Real Problem Is What AI Adds to Your Stack AI coding itself is not the issue. The problem is how quickly generated code can bring new open-source components into your environment. A developer can add a dependency in minutes. Your team may then need to assess vulnerabilities, licensing, maintenance, ownership, and whether that package should be there at all. That work does not disappear just because the co...

Cybersecurity researchers have found that several websites are still actively distributing a malware family known as Weedhack to gamers by masquerading as Minecraft clients. McAfee Labs said it detected and blocked more than 6,300 attempts to access malicious sites, adding that it found lookalike gaming websites designed to mimic legitimate projects, including branding, feature lists, FAQs, installation guides, developer credits, and links to genuine GitHub repositories. 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. Weedhack was first documented by the cybersecurity company back in June 2026, detailing its use of SEO poisoning and YouTube to redirect traffic to the bogus domains. The attack triggers a multi-stage sequence that culminates in the deployment of JAR payloads that can ...

Cybersecurity researchers have discovered a set of trojanized npm packages that masquerade as working calendar and streak utilities but are engineered to stealthily deliver an artificial intelligence (AI)-powered Linux implant dubbed RedC2 4.0. "When the module loads, it locates the bundled binary, marks it executable, and launches it as a detached background process," TrendAI, Trend Micro's enterprise cybersecurity business, said in a report published Thursday. "No install hook function call is needed; a single import anywhere in the dependency graph, even a transitive one, is enough to execute the payload." The list of identified packages is below - streak-metrics-math@1.0.0,1.0.1 kit-map-vim@1.0.0 streak-map-cache@1.0.0 streak-map-kit@1.0.0 map-streak-kit@1.0.0 streak-cache-map@1.0.0 streak-calc-metrics@1.0.0 streak-calc-math@1.0.0 streak-math-abz@1.0.0 streak-metricsaz@1.0.0 streak-math-metrics@1.0.0 streak-metrica...

Adversa AI has disclosed an attack technique that it says can cause xAI's Grok chatbot to send a user's name, approximate location, subscription tier, and the prompts from the ongoing conversation to an attacker-controlled server after the user asks it to summarize an ordinary web page. The AI security company, which has codenamed the technique "Cryptographic Context Injection ," said the transfer completed without a confirmation step and with no visible warning in its proof-of-concept demonstration. There is no patch, no CVE identifier, and no user-facing workaround, and the writeup does not report any exploitation in the wild. Asked which build was tested, Adversa told The Hacker News the target was the Grok web chat at grok.com running Grok 4.5 Fast, and that the attack was reproduced once on August 19, 2026. The writeup gives no success rate. The company said it has attempted the attack 20 times since June with a 40% success rate, and that the failures cam...

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 incident began when a Meta employee posted a technical question on an internal forum. An engineer used an approved AI agent to analyze it, but the agent posted its response publicly without approval. The employee followed its advice, inadvertently making a large volume of sensitive data available to unauthorized engineers for over two hours.  This was not shadow AI. The tool was approved, but the AI behaved in ways nobody had anticipated.  It’s a perfect example of security’s next big AI governance problem: shady AI.  Shadow AI is the unapproved use of AI tools  Shady AI is when employees use approved AI tools in unapproved, unexpected, or poorly governed ways  Shadow AI happens outside the organization's visibility. Shady AI happens inside it. And that makes it mu...

OpenAI on Tuesday revealed that it paused reinforcement learning ( RL ) training for its latest artificial intelligence (AI) models for two weeks while it shored up additional defenses and increased the scope of its monitoring to avert another Hugging Face-like incident . "As models become more capable, the risks associated with developing and testing them internally also grow," the AI company said . "Our standards for monitoring, alignment, and security must stay ahead of those risks. We wanted to take the time necessary to meet those standards, so we temporarily slowed the pace of scaling." The company said its largest planned frontier RL run remains on hold for the time being as it conducts smaller-scale training and evaluations to evaluate model behavior, validate its safeguards, and establish more concrete evidence of alignment before moving to the next phase. To that end, OpenAI said it plans to strengthen safeguards across its development process, inc...

Varonis Threat Labs has disclosed three vulnerabilities in Microsoft Copilot Personal that it said could allow a single click on a crafted link to silently pull data from connected apps and other information available to the victim's Copilot session. The flaws, which the researchers collectively named CoSnitch , turn in part on an undocumented URL parameter that the assistant itself surfaced during testing. The company said it reported the issue to Microsoft in December 2025 and that patches shipped on August 18, 2026. CoSnitch is tracked as  CVE-2026-24301  in Microsoft's Security Update Guide. 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. The researchers said they found no evidence that CoSnitch was exploited in the wild. They reached the parameter by repeatedly asking Copilot why a prompt could not be made to run without user interaction, an ap...

Security researchers at Anthropic and Switzerland's EPFL have demonstrated that self-propagating payloads can spread from one artificial intelligence (AI) agent to the next through the editable system prompt files that autonomous agent harnesses use to carry state between sessions. 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 . There is no evidence that the technique has spread successfully in the wild, and the same paper reports that a review of archived posts from Moltbook, the social network for AI agents, found no successful agent-to-agent propagation despite several attempts. A one-paragraph warning added to an agent's system prompt reduced spread to near zero across the payloads tested. Fifteen generations of adversarial optimization run against that warning on C...

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. Ray is an open-source, Python-native distributed computing framework designed to scale artificial intelligence and machine learning workloads. As of writing, the GitHub project has more than 43,500 stars and has been forked over 7,900 times. The vulnerability in question relates to CVE-2025-62593 (CVSS score: 9.4), which can result in remote code execution via web browsers like Mozilla Firefox and Apple Safari by means of a DNS rebinding attack . "Due to the longstanding decision by the Ray Development team to not implement any sort of authentication on critical endpoints, like the /api/jobs & /api/job_agent/jobs/ has once again led to a severe vulnerability that allows attackers to execute arbitrary code against Ray," according to an advisory shared by Ray ...

MCP servers can expose enterprise secrets through plaintext configuration files, over-permissioned access and prompt injection, often before security teams even know the server is running. As more organizations adopt AI agents into their systems, that exposure can silently become a major gap in MCP server security. 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. Behind that convenience, the MCP server connecting those tools and data to enterprise systems typically holds the keys to everything it touches: credentials, service account keys, API tokens and other secrets. Every organization should now question what secrets they are handing to AI and how well those secrets are protected once they reach an MCP server. What is Model Context Protocol (MCP)? Model Context Protocol (MCP) is an open standard, originally introduced by Anthropic, that allows...

A newly disclosed flaw in the way OpenAI, Anthropic, and Google carried hidden AI reasoning between API calls let researchers recover internal reasoning and secrets from session logs, including API keys and passwords. The weakness affected encrypted reasoning objects used by the providers' reasoning APIs, where a block created in one session could be replayed into another and, during testing, even handed to a weaker model in the same provider family to make it reveal the hidden content. The team behind the paper  Stealing Reasoning Traces from Proprietary LLM APIs  demonstrated four abuse paths: stealing proprietary reasoning for  model distillation , extracting private data from other users' published traces, recovering harmful content concealed behind a safe visible answer, and hiding prompt injections inside opaque reasoning blocks. Across 6,708 public agent trajectories, the team decoded 315,320 thinking blocks. After excluding benchmark sources, it counted 704...

Two malicious LiteLLM releases sat on PyPI for about 40 minutes in March carrying credential-stealing code capable of harvesting cloud keys, SSH keys, Kubernetes tokens, database passwords, and other secrets from systems that installed them. 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. Those totals are not a victim count. 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. The files were taken, in other words. CloudSEK has published the dataset as  a public lookup , searchable by name or domain and filterable by confidence. Each row gives an organization's name and domain, a count of secrets exposed, a count of runs, and a label reading High or Medium. What a high-co...

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. The trick can work even after a blunt version of the same theft is refused: split the request into fragments that each look routine, place them in channels the assistant already uses, and let the agent stitch them together and send the data back. The attack targets coding tools that connect to outside servers over the Model Context Protocol (MCP), the open standard that lets AI assistants call external tools. A malicious MCP server can put one fragment in a tool description and another in a tool result; some setups also support server-initiated sampling. MCP does preserve structured tool and result boundaries. But ASSET Research Group's tests show agents can still combine instructions across them in the same working context, so no single fragment has to contain the whole mal...

AI is helping development teams produce far more code, far faster. But security teams still have to review vulnerabilities, manage dependencies, prioritize fixes, and control risk at human speed. When software output jumps 10 to 50 times, the problem is no longer just finding vulnerabilities. It is keeping security from becoming the bottleneck, or worse, losing control of what gets shipped. In our latest webinar with Chainguard experts, “ The True Cost of Building at Machine Speed ,” you can now watch how security teams can keep AI-driven development fast without letting risk scale with it. For years, application security followed a familiar cycle: developers wrote code, scanners found problems, security teams prioritized them, and engineers fixed what mattered most. AI puts that model under pressure. If teams can suddenly create many times more code, security can also end up with many more components, dependencies, findings, and fixes to manage. More scanning alone does no...

OpenAI has announced that it's pausing some "internal activities" involving its upcoming artificial intelligence (AI) model Astra after an internal evaluation found it had made significant advancements in agentic coding and cybersecurity. In response to the discovery, the AI upstart said it's implementing security controls for higher-capability models and associated activities, such as isolated testing environments, restricted network and tool access, enhanced model weight protections and encryption, additional monitoring and detection capabilities, and sandboxed execution. "We are pausing internal activities involving Astra that do not yet meet these strengthened security control requirements," it said in a statement. "We have implemented universal monitoring for risky actions and misalignment across all agentic applications of Astra, including training and evaluation. Monitors evaluate the model's Chain of Thought and trigger a security re...

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. Two security firms found that behavior independently, by different routes. Only one of those routes is confirmed closed. PromptArmor , an AI security firm, hid the instructions in content Rovo reads. It said an uploaded file was enough to make the assistant gather internal data and send it out through a URL request, with no separate approval step. The firm published on August 5, 2026 and said the chain still worked with Rovo's web-search option switched off. That bypass is single-sourced, and the report establishes the finding's status only on that date; a later remediation is not confirmed here. Varonis Threat Labs put the instructions in a link instead. It found that the rovoChatPrompt URL parameter would preload attacker instructions into Rovo Chat, so one click from an authenticated user was enou...