All news in category “AI and Security Pulse”

🔐 Cloudflare presents a privacy-first approach to rate-limiting AI agents using anonymous credentials. The post explains how schemes such as ARC and ACT extend the Privacy Pass model by enabling late origin-binding, multi-show tokens, and stateful counters so origins can enforce limits or revoke abusive actors without identifying users. It outlines the cryptographic building blocks—algebraic MACs and zero-knowledge proofs—compares performance against Blind RSA and VOPRF, and demonstrates an MCP-integrated demo showing issuance and redemption flows for agent tooling.

🧬 Researchers used open-source AI to design variants of ricin and other toxic proteins, then converted those designs into DNA sequences and submitted them to commercial DNA-order screening tools. From 72 toxins and three AI packages they generated roughly 75,000 designs and found wide variation in how four screening programs flagged potential threats. Three of the packages were patched and improved after the test, but many AI-designed variants—often likely non-functional because of misfolding—exposed gaps in detection. The authors warn this imbalance could produce an arms race where design outpaces reliable screening.

🔒 Microsoft-owned LinkedIn will begin using profile details, public posts and feed activity from users in the UK, EU, Switzerland, Canada and Hong Kong to train generative AI models and to support personalised ads across Microsoft starting 3 November 2025. Private messages are excluded. Users can opt out via Settings & Privacy > Data Privacy and toggle Data for Generative AI Improvement to Off. Organisations should update social media policies and remind staff to review their advertising and data-sharing settings.

🔒 Artificial intelligence is reshaping cybersecurity, creating both enhanced defensive capabilities and new risks that require broad AI literacy. The White House's America’s AI Action Plan and Fortinet’s 2025 Cybersecurity Global Skills Gap Report show strong adoption—97% of organizations use or plan AI in security—yet 48% cite lack of staff expertise as a major barrier. Fortinet recommends targeted training, policies for generative AI use, and its Security Awareness modules to help close the gap and reduce threat exposure.

⚠ Researchers report a new form of context-poisoning called AI-targeted cloaking that serves different content to agentic browsers and AI crawlers. SPLX shows attackers can use a trivial user-agent check to deliver alternate pages to crawlers from ChatGPT and Perplexity, turning retrieved content into manipulated ground truth. The technique mirrors search engine cloaking but targets AI overviews and autonomous reasoning, creating a potent misinformation vector. A concurrent hTAG analysis also found many agents execute risky actions with minimal safeguards, amplifying potential harm.

🛡️ The UK AI Security Institute (AISI), Check Point and Lakera have launched b3, an open-source benchmark to assess and strengthen the security of backbone LLMs that power AI agents. b3 focuses on the specific LLM calls within agent workflows where malicious inputs can trigger harmful outputs, using 10 representative "threat snapshots" combined with a dataset of 19,433 adversarial attacks from Lakera’s Gandalf initiative. The benchmark surfaces vulnerabilities such as system prompt exfiltration, phishing link insertion, malicious code injection, denial-of-service and unauthorized tool calls, making LLM security more measurable, reproducible and comparable across models and applications.

⚠️ The British Standards Institution warns of a widening AI governance gap as many organisations accelerate AI adoption without adequate controls. An AI-assisted review of 100+ annual reports and two polls of 850+ senior leaders found strong investment intent but sparse governance: only 24% have a formal AI program and 47% use formal processes. The report highlights weaknesses in incident management, training-data oversight and inconsistent approaches across markets.

🔍 Advances in AI and deepfake technology make it increasingly difficult to tell what’s real online, enabling convincingly fake videos, images and audio that scammers exploit to deceive individuals and organizations. Threat actors use deepfakes of public figures to promote bogus investments, create synthetic nudes to extort victims and deploy fake voices and videos to trick employees into wiring corporate funds. Watch ESET Chief Security Evangelist Tony Anscombe outline practical defenses to recognize and resist deepfakes, and explore other Cybersecurity Awareness Month videos on authentication, patching, ransomware and shadow IT.

🛡️ Organizations face an identity crisis as generative AI and vast troves of breached personal data enable realistic digital doppelgangers. Attackers now automate hyper-personalized phishing, smishing and vishing, clone voices, and run coordinated multi-channel campaigns that reference real colleagues and recent projects. The article urges a shift to “never trust, always verify,” with radical visibility, rapid detection and phishing-resistant authentication such as FIDO2. It also warns of emerging agentic AI and recommends strict least-privilege controls plus continuous red-teaming.

🔐 Agentic AI is presented as a practical cybersecurity capability that can operate without direct human supervision, handling high-volume, time-sensitive tasks at machine speed. Industry leaders from Zoom to Dell Technologies and Deloitte highlight seven priority use cases — from autonomous threat detection and SOC augmentation to real-time zero‑trust enforcement — that capitalize on AI's scale and speed. The technology aims to reduce alert fatigue, accelerate mitigation, and free human teams for strategic work.

🛡️ The article explains how malicious SEO operations use keyword stuffing, purchased backlinks, cloaking and mass-produced content to bury legitimate sites in search results. It warns that generative AI now amplifies this threat by producing tens of thousands of spam articles, spinning up fake social accounts and enabling more sophisticated cloaking. Defenders must deploy AI-based detection, graph-level backlink analysis and network behavioral analytics to spot coordinated abuse. The piece emphasizes proactive, ecosystem-wide monitoring to protect trust and legitimate businesses online.

🔐 Google selected 11 startups for its inaugural Google for Startups Accelerator: AI for Cybersecurity in Latin America, a ten-week program that pairs founders with Google's technical resources, mentorship, and product support. The cohort — drawn from Brazil, Chile, Colombia, and Mexico — focuses on AI-driven solutions across threat detection, compliance automation, fraud prevention, and protections for AI agents. Participants will receive hands-on guidance to scale, validate, and deploy tools that reduce cyber risk across the region.

🤖 GitHub announced Agent HQ, a unified platform that makes coding agents native to the GitHub workflow. Over the coming months, partner agents from OpenAI, Anthropic, Google, Cognition, and xAI will become available as part of paid Copilot subscriptions. The release introduces a cross‑surface mission control, VS Code planning and customizable AGENTS.md files, and an enterprise control plane with governance, metrics, and code‑quality tooling to manage agent-driven work.

🔐 Google Cloud outlines a pragmatic framework to secure the emerging agentic web while preserving smooth user experiences. The post details how reCAPTCHA and Google Cloud combine agent and user identity, continuous behavior analysis, and AI-resistant mitigations such as mobile-device attestations. It highlights enabling safe agentic commerce via protocols like AP2 and tighter integration with cloud AI services.

🎧 In episode 74 of The AI Fix, hosts Graham Cluley and Mark Stockley survey recent AI developments including Amazon’s experimental delivery glasses, Channel 4’s AI presenter, and reports of LLM “brain rot.” They examine practical security risks — such as malicious browser extensions spoofing AI sidebars and AI browsers being tricked into purchases — alongside wider societal debates. The episode also highlights public calls to pause work on super-intelligence and explores what AGI really means.

🕹️ Google says its AI Studio will enable users to 'vibe code' simple video games by the end of the year. The company claims the tool can automatically select models and wire up APIs to streamline app creation, while noting current limitations for production-ready systems. Product lead Logan Kilpatrick highlighted the potential to broaden access to game creation, and startups like Cursor are pursuing similar next-generation vibe coding tools.

⚠️ Researchers disclosed a CSRF-based vulnerability in ChatGPT Atlas that can inject malicious instructions into the assistant's persistent memory, potentially enabling arbitrary code execution, account takeover, or malware deployment. LayerX warns that corrupted memories persist across devices and sessions until manually deleted and that Atlas' anti-phishing defenses lag mainstream browsers. The flaw converts a convenience feature into a persistent attack vector that can be invoked during normal prompts.

⚠️ OpenAI's new Atlas browser is vulnerable to a prompt-injection jailbreak that disguises malicious instructions as URL-like strings, causing the omnibox to execute hidden commands. NeuralTrust demonstrated how malformed inputs that resemble URLs can bypass URL validation and be handled as trusted user prompts, enabling redirection, data exfiltration, or unauthorized tool actions on linked services. Mitigations include stricter URL canonicalization, treating unvalidated omnibox input as untrusted, additional runtime checks before tool execution, and explicit user confirmations for sensitive actions.

🔒 Cloudflare, in partnership with Visa and Mastercard, explains how Web Bot Auth together with payment-specific protocols can secure agent-driven commerce. The post describes agent registration, public key publication, and HTTP Message Signatures that include timestamps, nonces, and tags to prevent spoofing and replay attacks. Merchants can validate trusted agents during browsing and payment flows without changing infrastructure. Cloudflare also provides an Agent SDK and managed WAF rules to simplify developer adoption and deployment.

🔒 Organizations worldwide are rapidly adopting AI across enterprises, delivering efficiency gains while introducing new security risks. Cybersecurity is at a turning point where AI fights AI, and today's phishing and deepfakes are precursors to autonomous, self‑optimizing AI threat actors that can plan, execute, and refine attacks with minimal human oversight. In September 2025, Check Point Research found that 1 in 54 GenAI prompts from enterprise networks posed a high risk of sensitive-data exposure, underscoring the urgent need to harden defenses and govern model use.