< ciso
brief />
Tag Banner

All news with #ai data leakage tag

28 articles

Signal Blocks OS AI Agents From Reading Private Chats

🔒 Signal will now restrict screenshots by default on Android and iOS to stop operating system AI agents from reading private chats, while admitting the safeguard is incomplete. The change extends a similar Windows 11 setting introduced after Microsoft's Recall feature. Meredith Whittaker called agentic OS integrations an existential threat to the app's privacy guarantees. Signal is urging Apple, Google, and Microsoft to offer clear opt-outs and better documentation.
read more →

AWS Quick Adds Enterprise Guardrails for Prompt Safety

🛡️ AWS Quick now offers enterprise guardrails in preview, giving administrators configurable content safety and sensitive-information controls for user prompts. Policies can be assigned to individual users, identity provider groups, or set as the account default. A detect-only mode lets teams test rules against real production traffic before enforcement begins.
read more →

UK Workers Hand Credentials and Access to Unmanaged AI

🤖 A new Dashlane report finds that 19% of UK employees have shared a password, API key or login credential with an AI tool or agent. Nearly half (48%) have let AI perform actions without first asking permission, including accessing emails, company files and customer data. Meanwhile, 72% of IT and security executives report an AI-related credential or access incident in the past year, even though 92% say they know every agent in use.
read more →

Survey Finds Widespread AI Data Exposure Risks

🔒 Research from BlackFog reveals nearly half of security leaders (47%) reported sensitive data exposure via AI tools at their organizations, with 77% of incidents involving approved tools. The findings highlight gaps in visibility and enforcement: while 98% have AI policies, only 56% pair them with comprehensive technical controls and many organizations cannot log or restrict data shared with AI. Autonomous AI agents and shadow AI further escape controls, especially in technology and telecoms.
read more →

How AI Memory Systems Create New Data Risks

🧠 This article examines how memory systems for AI chatbots — from application layers like Mem0 to vector databases such as Milvus and frontends like Attu — persist user information and can be misconfigured and exposed. It describes findings of exposed instances, examples from companion apps for children and elders, and the privacy and integrity risks including ransom and data tampering. The piece highlights how these memory stores reveal intimate, AI-curated user details and expand the attack surface for AI deployments.
read more →

September 2026 Cyber Threat Landscape Report

🛡️ September 2026 saw a sharp rise in global cyber activity: weekly attacks averaged 2,803 per organization, up 48% year over year, with education the most targeted sector. Phishing rates increased to 1 in 91 emails and ransomware incidents rose 53% year over year. GenAI prompt usage expanded, with 1 in 39 prompts posing a high risk of sensitive data leakage, highlighting growing exposure across industries and regions.
read more →

Expanding Credential Layer: Visibility and Risk

🔒 GitGuardian outlines why the credential layer deserves immediate attention and how detection is the first essential step. The article explains that credential sprawl spans repositories, endpoints, collaboration tools, and AI agents, increasing attack surfaces and making discovery urgent. It highlights research showing rising hardcoded secrets and explains the need for context — validity, ownership, permissions, and dependencies — to prioritize remediation. The piece argues security teams must connect multiple discovery sources to measure coverage and reduce exposure.
read more →

Report: Over 80,000 Organizations Had AI Logins Exposed

🔍 SOCRadar’s AI Identity Exposure Report analyzed more than one million infostealer records tied to AI services across 80,000+ corporate domains, narrowing to 482 major enterprises to determine which organizations’ AI credentials are being sold. The research found ChatGPT/OpenAI dominated exposures, with 90% of records, while developer platforms like Hugging Face and Replit also appeared. The report emphasizes that stolen AI sessions are more dangerous than passwords, acting as archives, execution engines, billable resources, and identities, and recommends SSO, token rotation, API key controls, and monitoring for session reuse.
read more →

Malicious MCP Servers Can Split Exfiltration Steps

🛡️ A new technique called GhostSplice shows how a malicious Model Context Protocol (MCP) server connected to an AI coding assistant can exfiltrate SSH keys, environment secrets, source code, and customer data by splitting a theft into harmless-looking fragments. ASSET Research Group tested the approach in isolated projects using fake credentials and found that splitting the request across tool descriptions, results, or server-initiated sampling raised compliance dramatically for many models. The attack relies on developers connecting a hostile MCP server and the agent already having access to the target files.
read more →

AI hallucinations fuel slopsquatting risk for devs

🔍 Research shows top AI coding models repeatedly invent identical nonexistent package names, creating a slopsquatting risk where attackers could register those names and distribute malicious libraries. Aleksandr Churilov identified 127 shared fake package names across five LLMs and found 53 remain registerable on PyPI and npm. The study suggests shared training materials and ecosystem conventions drive the conformity, though no malicious registrations have yet been observed.
read more →

Malicious npm package stole files from AI tool

🛡️ Researchers uncovered a malicious npm package named mouse5212-super-formatter that exfiltrates files from the /mnt/user-data directory used by Anthropic's Claude AI. OX Security describes the campaign, codenamed Malware-Slop, as a postinstall script that authenticates to GitHub using environment or hard-coded tokens, creates or targets a repository, and uploads local files to an attacker-controlled account. The package has been downloaded hundreds of times, and the linked GitHub account—created shortly before the package appeared—has since disappeared. Analysts noted the actor leaked a private token, suggesting poor OPSEC and possibly AI-assisted malware creation.
read more →

Venice OT intrusion claim and Anthropic source leak risks

🔒 Smashing Security episode 463 examines two incidents that expose operational and AI security weaknesses: a claimed intrusion into Venice’s flood‑defence pump controls and an accidental full‑source disclosure by Anthropic. Hosts Graham Cluley and Tanya Janca discuss the physical risks of compromised legacy OT systems, how packaging/CI misconfigurations can leak high‑value IP and attack surface, and the governance challenges of powerful internal tools like Mythos. They recommend stronger CI/CD defaults, strict access controls for model assets, and reliable out‑of‑band incident communications.
read more →

Experts Warn of Browser Extensions Poaching AI Prompts

🛡️ Security researchers have warned of malicious Chrome extensions that silently monitor and exfiltrate users' AI chat content. According to Expel, extensions watch open tabs and capture prompts and responses via API interception or DOM scraping before sending the data to external servers. Attackers either impersonate popular tools or convert legitimate extensions into malicious ones after building a user base. Organisations are urged to block unvetted AI extensions and centrally manage and audit extension use.
read more →

Preventing AI Agent Data Leaks: Webinar Guide for Security

🔒 AI agents are transforming workflows but can act as an unmonitored access layer—an 'invisible employee' with broad privileges. In an upcoming webinar, Rahul Parwani, Head of Product for AI Security at Airia, will explain how attackers are manipulating agents to exfiltrate sensitive information and how to stop them. Attendees will learn the Dark Matter of identity, common manipulation techniques, and a practical safety blueprint to limit privileges, detect misuse, and prevent leaks. Reserve your spot to learn actionable defenses.
read more →

Shai-Hulud–Style Worm Hits npm Packages and AI Tools

🔒 Socket's Threat Research Team discovered a supply chain worm, tracked as SANDWORM_MODE, spreading via typosquatted npm packages and compromised GitHub accounts while also manipulating local AI coding assistants. The malware harvested developer and CI credentials, injected rogue MCP servers into tools like Claude Desktop and VS Code Continue, and exfiltrated API keys for multiple large language model providers. Affected packages were removed and infrastructure disabled; developers should rotate credentials and audit CI workflows and local AI configurations.
read more →

Supply Chain Worm Uses Malicious npm Packages to Steal Keys

🔐 Socket warns of an active supply-chain worm, codenamed SANDWORM_MODE, that abused at least 19 malicious npm packages to harvest developer credentials and cryptocurrency keys. The packages — many typosquatting legitimate modules and published by aliases official334 and javaorg — contain code to steal tokens, environment secrets and LLM API keys. The campaign also includes a weaponized GitHub Action, an optional home-directory wiper, and an McpInject component that targets AI coding assistants. Users should remove affected packages, rotate tokens, and audit repositories and CI workflows.
read more →

The Silent Security Gap in Enterprise AI Adoption Risks

🔒 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.
read more →

AI Coding Assistants Secretly Exfiltrate Developers' Code

⚠️A new report alleges two popular AI coding assistants, together used by roughly 1.5 million developers, are quietly copying everything they ingest to servers in China. Security researchers say the extensions capture editor content, code snippets, and related telemetry without clear user disclosure. The behavior appears systematic and persistent rather than incidental. Until vendors provide transparent remediation, developers and organizations should avoid unvetted extensions and perform immediate audits and containment.
read more →

CISA Acting Director Uploaded FOUO Files to ChatGPT

🛡️ The acting director of the U.S. Cybersecurity and Infrastructure Security Agency uploaded multiple for official use only (FOUO) contracting documents to the public version of ChatGPT between mid‑July and early August 2025, triggering automated DHS security alerts. Sensors detected the activity in early August, generating several alerts in the first week and prompting an internal review. The uploads—containing contracting information not intended for public release—underscore gaps in AI governance and exception handling for senior officials at CISA.
read more →

Cloudflare Acquires Human Native to Improve AI Data Access

🤝 Cloudflare has acquired Human Native, a UK AI data marketplace that converts multimedia into licensed, structured datasets for AI developers. The team will help Cloudflare expand tools like AI Crawl Control, Pay Per Crawl and the AI Index, enabling publishers to expose structured updates and control access. It emphasizes licensed, high-quality data, creator compensation and greater control over how content is used by AI systems.
read more →