All news in category “AI and Security Pulse”

🔒 OpenClaw has integrated VirusTotal malware scanning into its ClawHub skills marketplace to automatically vet published skills. Packages are hashed and analyzed with Code Insight (powered by Gemini); benign skills are auto-approved, suspicious ones receive warnings, and confirmed malicious skills are blocked and re-scanned daily. The move responds to documented malicious extensions and unauthorized enterprise deployments, though OpenClaw stresses scanning is not a complete defense against prompt injection or logic abuse.

🔎 Claude Opus 4.6 markedly improves automated vulnerability discovery, finding high-severity bugs faster and without task-specific tooling. Unlike traditional fuzzers, which depend on massive random inputs, Opus 4.6 reads and reasons about code like a human researcher—spotting patterns, past fixes, and precise inputs that trigger failures. Early tests show it uncovered long-standing zero-days in projects previously subject to extensive fuzzing.

🔒SecurityScorecard warns that widespread misconfiguration of the AI assistant OpenClaw has left 40,214 agent instances — linked to 28,663 unique IP addresses — exposed to the public internet. The vendor reports 63% of observed deployments are vulnerable, including 12,812 instances exploitable via remote code execution, and has correlated hundreds with prior breaches and known CVEs. Exposures are concentrated in China, the US and Singapore and affect sectors such as information services, technology, manufacturing and telecommunications. Users are urged to limit access, adopt a zero trust posture, scrutinize agent logic, and defend against prompt injection and leaked API keys.

🛡️ OpenClaw has integrated VirusTotal scanning to inspect skills uploaded to its ClawHub marketplace, creating SHA-256 hashes for each skill and cross-checking them against VirusTotal's database. Bundles not matched are analyzed with VirusTotal Code Insight; benign verdicts are auto-approved, suspicious skills are flagged, and confirmed malicious items are blocked. OpenClaw also re-scans active skills daily but cautions this is not a complete defense against cleverly concealed prompt-injection payloads.

🔍 Anthropic says its newly released large language model, Claude Opus 4.6, was used internally to identify zero-day vulnerabilities in open-source software. The model ran inside a virtual machine with access to current project repositories and standard analysis utilities but received no specific instructions on how to conduct hunts. Despite that, Anthropic reports the system flagged 500 high-severity vulnerabilities, and company staff are manually validating findings before reporting them to maintain accuracy.

🔍 This article explains how modern deepfakes are created, deployed, and detected in real-world scams, and why virtually anyone can be a target. It describes common visual, auditory, and behavioral signs—lighting and lip-sync errors, unnatural blinking, electronic vocal tones, and awkward gestures—and notes attackers use tools from Telegram bots to commercial services like HeyGen and ElevenLabs. Practical advice includes ending suspicious chats, verifying identities via alternate channels, agreeing a family codeword, tightening privacy on photos and recordings, enabling strong account security, and using content-analyzer services to flag AI-generated media.

🔍 Anthropic's Claude Opus 4.6 has identified more than 500 previously unknown high-severity vulnerabilities across major open-source libraries, including Ghostscript, OpenSC, and CGIF. Launched this week, the model shows improved code-review and debugging capabilities and was evaluated by Anthropic's Frontier Red Team in a virtualized environment using standard developer tools. Anthropic says each flagged defect was validated and patched by maintainers, positioning the model as a defender-oriented tool to help prioritize serious memory-corruption risks while it iterates on additional safeguards to limit misuse.

🚨 As AI agent frameworks surge, Talos warns of two immediate threats: Clawdbot — a popular open-source agentic tool (aka Moltbot/OpenClaw) that requires users to store credentials and API keys locally and can accept unvetted Skills granted broad system privileges. DKnife, active since at least 2019, is a modular Linux attack framework that compromises routers and edge devices to intercept traffic, hijack updates, and deliver malware while evading many endpoint defenses. The newsletter urges skepticism toward rushed AI tools and recommends hardening gateways, auditing firmware, enforcing strong authentication, and monitoring for suspicious update behaviors.

🤖 Claude Opus 4.6 is now available in Microsoft Foundry on Azure, delivering Anthropic’s advanced reasoning and agent capabilities to enterprise workflows. The model supports a beta 1M-token context window, up to 128K output tokens, and new API controls including Adaptive Thinking and Context Compaction. Integrated with Foundry IQ and Azure governance, Opus 4.6 targets coding, knowledge work, finance, legal, cybersecurity, and multi-tool agent automation—helping teams move from experimentation to production while preserving compliance and operational control.

🔒 Glean and Prisma AIRS have integrated to provide real-time AI threat protection that neutralizes prompt injections, blocks toxic or biased outputs, and inspects generated code and URLs for malicious patterns. The integration enforces organizational policy across chats and agent interactions and immediately blocks risky requests while notifying users. Deployment is designed to be frictionless—enable protection in three clicks by pasting a Prisma AIRS runtime API key into the Glean admin console.

🔒 The Buyer’s Guide for AI Usage Control warns that AI adoption has far outpaced visibility and governance, producing a widening gap as AI is embedded across SaaS, browsers, copilots, extensions and shadow tools. It reframes the problem as an interaction issue rather than solely a data or app problem, and positions AI Usage Control (AUC) as a distinct governance layer that must discover and enforce policy at the moment of interaction. The guide outlines four operational stages—Discovery, Interaction Awareness, Identity & Context, and Real-Time Control—and stresses that architectural fit, operational overhead, and user experience are decisive factors when selecting a solution.

🔒 Most security leaders assume they know where sensitive data resides, but rapid AI adoption has created a new exposure surface in AI inference traffic. Prompts often contain source code, contracts, PII and proprietary workflows that flow through application layers, logs and third‑party services without classification or adequate controls. Traditional protections — transport encryption, legacy DLP and standard logging practices — frequently fail to prevent prompt leakage, producing an often invisible and growing enterprise risk.

🛡️ Microsoft has developed a scanner to detect hidden backdoors in open-weight language models, focusing on triggers and malicious behaviors inserted during training or fine-tuning. The tool flags three observable signatures — attention hijacking, leakage of poisoned training fragments, and sensitivity to partial triggers — and runs using forward passes only without retraining or backpropagation. It is designed to work with most causal, GPT-style models and to serve as an added layer of supply-chain security for enterprises using third-party or open-source models.

⚠️ Gravitee reports that roughly half of an estimated three million AI agents running in US and UK enterprises are unmonitored and potentially "going rogue." A December 2025 Opinion Matters survey of 750 IT executives found a mean of 36.9 agents per large organization and that 88% suspected an agent-related security or privacy incident in the prior year. Experts warn deployment is outpacing governance and call for continuous runtime oversight, tiered access controls, and stricter credential management.

📰 In episode 453 of Smashing Security, Graham Cluley and guest Tricia Howard examine how sloppy redaction and a mix of AI and open social profiles can deanonymise documents once thought obscured. They discuss real-world incidents including malware delivery via a compromised Notepad++ installer, a sex-addiction app leaking intimate user data, and a problematic AV update used to distribute malware. The episode also highlights insider-threat risks after a senior US cybersecurity official uploaded sensitive government material into a public ChatGPT instance, and explores how broken trust can have lasting reputational consequences for vendors and organisations.

🔍 Microsoft has developed a lightweight scanner to detect backdoors in open-weight large language models (LLMs) by evaluating three observable signals tied to internal model behavior. The tool extracts memorized content, isolates suspect substrings, and scores candidates with loss functions that formalize attention and output anomalies. The approach requires no additional training and runs across common GPT‑style models, but it needs access to model files and is best suited for trigger-based, deterministic backdoors.

🔍 Microsoft researchers released new findings and a practical scanner for detecting backdoors in open-weight language models. The study identifies three signatures — a distinctive “double triangle” attention pattern, leakage of poisoning training data through memorization, and trigger “fuzziness” — and uses them to reconstruct likely triggers without retraining. The scanner requires only forward passes, works on GPT-like models, and was validated across 270M–14B models and common fine-tuning regimes. The team notes limits: it needs model file access, favors deterministic backdoors, and should be used as part of layered defenses.

🔒 OpenClaw is an open-source, agentic AI assistant that can run locally or on servers, connect to LLMs and external APIs, and autonomously perform actions such as sending email or controlling browsers. Its local storage of config and broad access (files, terminals, sometimes root) makes misconfigured deployments attractive as backdoors. CrowdStrike observed rapid adoption and internet-exposed instances, and recommends discovery, runtime guardrails, and automated removal integrated into detection workflows.

⚠️ Security researcher Paul McCarty (aka 6mile) has identified 386 malicious OpenClaw "skills" on the ClawHub repository that impersonate crypto trading tools. The add-ons use social engineering to trick users into executing commands that deploy infostealers on macOS and Windows, harvesting exchange API keys, wallet private keys, SSH credentials and browser passwords. The discovered skills share a common C2 IP (91.92.242.30) and many remain available, with the most active uploader accounting for nearly 7,000 downloads.

🔐 AI agents—custom GPTs, copilots, coding agents and other autonomous tooling—are proliferating in production while remaining largely outside traditional IAM, PAM, and IGA controls. The piece argues for treating agents as a distinct identity class and applying continuous identity lifecycle management to ensure visibility, ownership, dynamic least privilege, and auditability. Rather than slowing adoption, this approach positions identity as the control plane for balancing innovation and security.