Tag: ai
-
Microsoft Patches Security Flaw That Exposed Confidential Emails to AI
Microsoft Corp. confirmed it is addressing a significant security lapse that allowed its Copilot AI to bypass privacy protections and summarize users’ confidential emails without authorization. The bug, which has persisted since late January, effectively ignored data loss prevention (DLP) protocols designed to keep sensitive corporate information out of the reach of large language models..…
-
Hidden Commands Found in AI Summarize Buttons
Commands Push Lasting Preferences Into AI Assistants. Microsoft researchers found companies embedding hidden commands in summarize with AI buttons to plant lasting brand preferences in assistants’ memory. The tactic, dubbed AI recommendation poisoning, exploits persistent memory features to bias future responses. First seen on govinfosecurity.com Jump to article: www.govinfosecurity.com/hidden-commands-found-in-ai-summarize-buttons-a-30784
-
AI platforms can be abused for stealthy malware communication
AI assistants like Grok and Microsoft Copilot with web browsing and URL-fetching capabilities can be abused to intermediate command-and-control (C2) activity. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/security/ai-platforms-can-be-abused-for-stealthy-malware-communication/
-
‘Promptware’ Attacks Await an Unprepared AI Industry
Researchers Say AI Prompt Injection Has Emerged As a Dangerous New Class of Attacks. The large language model industry has mostly treated prompt injection attacks as a risk analogous to traditional web server prompt injection attacks. Researchers now say feeding rogue instructions to an artificial intelligence system merits its own classification as promptware. First seen…
-
A CISO’s Playbook for Defending Data Assets Against AI Scraping
Discover a strategic approach to govern scraping risks, balance security with business growth, and safeguard intellectual capital from automated data harvesting. First seen on darkreading.com Jump to article: www.darkreading.com/cyber-risk/ciso-playbook-defending-data-assets-against-ai-scraping
-
‘Pomptware’ Attacks Await an Unprepared AI Industry
Researchers Say AI Prompt Injection Has Emerged As a Dangerous New Class of Attacks. The large language model industry has mostly treated prompt injection attacks as a risk analogous to traditional web server prompt injection attacks. Researchers now say feeding rogue instructions to an artificial intelligence system merits its own classification as promptware. First seen…
-
Firebase Misconfiguration Exposes 300M Messages From Chat Ask AI Users
A technical mistake in the popular Chat Ask AI app has left 300 million private messages from 25 million users exposed online. Discover what happened and how you can protect your personal data when using AI chatbots. First seen on hackread.com Jump to article: hackread.com/firebase-misconfiguration-chat-ask-ai-users-expose/
-
‘Pomptware’ Attacks Await an Unprepared AI Industry
Researchers Say AI Prompt Injection Has Emerged As a Dangerous New Class of Attacks. The large language model industry has mostly treated prompt injection attacks as a risk analogous to traditional web server prompt injection attacks. Researchers now say feeding rogue instructions to an artificial intelligence system merits its own classification as promptware. First seen…
-
The 20 Coolest Web, Email and Application Security Companies Of 2026: The Security 100
CRN’s Security 100 list of the coolest web, email and application security companies includes AI-powered vendors protecting email inboxes and web browsers along with providers of modern code security. First seen on crn.com Jump to article: www.crn.com/news/security/2026/the-20-coolest-web-email-and-application-security-companies-of-2026-the-security-100
-
Agentic AI affecting the world of the SOC
While many are tempted to lean on agentic AI, the rush to do so is potentially creating more risk First seen on computerweekly.com Jump to article: www.computerweekly.com/microscope/news/366639051/Agentic-AI-affecting-the-world-of-the-SOC
-
How synthetic data can help solve AI’s data crisis
As AI demand outpaces the availability of high-quality training data, synthetic data offers a path forward. We unpack how synthetic datasets help teams overcome data scarcity to build production-ready AI. First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/02/how-synthetic-data-can-help-solve-ais-data-crisis/
-
Threat groups using AI to speed up and scale cyberattacks
A report from Palo Alto Networks finds hackers are increasingly using stolen identities and exploiting critical vulnerabilities within minutes of disclosure. First seen on cybersecuritydive.com Jump to article: www.cybersecuritydive.com/news/threat-groups-ai-speed-scale-cyberattacks/812439/
-
Millionen Chrome-Erweiterungen geben Browserverlauf preis
Eine Sicherheitslücke in beliebten Chrome-Erweiterungen führt dazu, dass der Browserverlauf der Anwender offengelegt ist.Ein Sicherheitsforscher mit dem Pseudonym ‘Q Continuum” hat 287 Chrome-Erweiterungen entdeckt, die den Browserverlauf exfiltrieren. ‘Die Akteure hinter den Lecks sind vielfältig: Similarweb, Curly Doggo, Offidocs, chinesische Akteure, viele kleinere, unbekannte Datenbroker sowie ein mysteriöses Unternehmen namens “šBig Star Labs’, das offenbar…
-
Microsoft says Office bug exposed customers’ confidential emails to Copilot AI
Microsoft said the bug meant that its Copilot AI chatbot was reading and summarizing paying customers’ confidential emails, bypassing data protection policies. First seen on techcrunch.com Jump to article: techcrunch.com/2026/02/18/microsoft-says-office-bug-exposed-customers-confidential-emails-to-copilot-ai/
-
AI Assistants Used as Covert CommandControl Relays
AIs like Grok and Microsoft Copilot can be exploited as covert C2 channels for malware communication First seen on infosecurity-magazine.com Jump to article: www.infosecurity-magazine.com/news/ai-assistants-covert-c2-relays/
-
Deepfakes, Betrug und unsichere Inhalte – 2025 gab es 346 KI-Vorfälle, ChatGPT am häufigsten beteiligt
First seen on security-insider.de Jump to article: www.security-insider.de/steigende-vorfaelle-von-ki-missbrauch-2025-a-7188b0190068b40db2b49da6c7830a1f/
-
Securing OpenClaw Against”ClawHavoc”
As of February 2026, OpenClaw (formerly Clawdbot and Moltbot ) is a popular platform for autonomous AI agents. Its “sovereign” architecture, which gives AI direct access to file systems and terminals, significantly increases its attack surface”, leading to elevated risks, most notably illustrated by the ClawHavoc supply-chain campaign, which exposed thousands of deployments to potential…
-
Palo Alto Networks CEO: AI Won’t Replace Security Tools ‘Any Time Soon’
Investor fears that AI poses more of a risk than an opportunity for cybersecurity vendors are unfounded, with LLMs unlikely to become capable of displacing security products in the foreseeable future, Palo Alto Networks CEO Nikesh Arora said Tuesday. First seen on crn.com Jump to article: www.crn.com/news/security/2026/palo-alto-networks-ceo-ai-won-t-replace-security-tools-any-time-soon
-
Your AI-generated password isn’t random, it just looks that way
Seemingly complex strings are actually highly predictable, crackable within hours First seen on theregister.com Jump to article: www.theregister.com/2026/02/18/generating_passwords_with_llms/
-
Cybersecurity Tech Predictions for 2026: Operating in a World of Permanent Instability
In 2025, navigating the digital seas still felt like a matter of direction. Organizations charted routes, watched the horizon, and adjusted course to reach safe harbors of resilience, trust, and compliance.In 2026, the seas are no longer calm between storms. Cybersecurity now unfolds in a state of continuous atmospheric instability: AI-driven threats that adapt in…
-
Cybersecurity Tech Predictions for 2026: Operating in a World of Permanent Instability
In 2025, navigating the digital seas still felt like a matter of direction. Organizations charted routes, watched the horizon, and adjusted course to reach safe harbors of resilience, trust, and compliance.In 2026, the seas are no longer calm between storms. Cybersecurity now unfolds in a state of continuous atmospheric instability: AI-driven threats that adapt in…
-
(g+) Anthropics Bericht über KI-Hacker: Keine CVE-ID – didn’t happen!
Ohne gründliche Dokumentation sind Anthropics Berichte über KI-Hacker unglaubwürdig. Das heißt nicht, dass LLMs kein Risiko darstellen. First seen on golem.de Jump to article: www.golem.de/news/anthropics-bericht-ueber-ki-hacker-keine-cve-id-didn-t-happen-2602-205498.html
-
AI Found Twelve New Vulnerabilities in OpenSSL
The title of the post is”What AI Security Research Looks Like When It Works,” and I agree: In the latest OpenSSL security release> on January 27, 2026, twelve new zero-day vulnerabilities (meaning unknown to the maintainers at time of disclosure) were announced. Our AI system is responsible for the original discovery of all twelve, each…
-
AI Found Twelve New Vulnerabilities in OpenSSL
The title of the post is”What AI Security Research Looks Like When It Works,” and I agree: In the latest OpenSSL security release> on January 27, 2026, twelve new zero-day vulnerabilities (meaning unknown to the maintainers at time of disclosure) were announced. Our AI system is responsible for the original discovery of all twelve, each…
-
OpenClaw AI Framework v2026.2.17 Adds Anthropic Model Support Amid Credential Theft Bug Concerns
OpenClaw, the open-source autonomous AI assistant that has gained widespread adoption in early 2026, released version v2026.2.17 on February 17, 2026, introducing support for Anthropic’s latest Claude Sonnet 4.6 model. The release comes amid growing security concerns after researchers documented the first in-the-wild credential theft targeting OpenClaw configuration files by infostealer malware. New Anthropic Model…
-
From Shadow APIs to Shadow AI: How the API Threat Model Is Expanding Faster Than Most Defenses
The shadow technology problem is getting worse. Over the past few years, organizations have scaled microservices, cloud-native apps, and partner integrations faster than corporate governance models could keep up, resulting in undocumented or shadow APIs. We’re now seeing this pattern all over again with AI systems. And, even worse, AI introduces non-deterministic behavior, autonomous actions,…
-
Keenadu: Android malware that comes preinstalled and can’t be removed by users
Embedded in core system apps: Keenadu can control legitimate system applications on affected devices. Kaspersky observed it inside critical components such as face unlock applications, raising the possibility that attackers could access biometric data. The malware was also found operating within the home screen app that controls the device’s primary interface.The researchers warned that the…
-
Microsoft says bug causes Copilot to summarize confidential emails
Microsoft says a Microsoft 365 Copilot bug has been causing the AI assistant to summarize confidential emails since late January, bypassing data loss prevention (DLP) policies that organizations rely on to protect sensitive information. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/microsoft/microsoft-says-bug-causes-copilot-to-summarize-confidential-emails/
-
Prompt Control is the New Front Door of Application Security
Discover how AI-driven systems are redefining application security. Research highlights the importance of focusing on inference layers, prompt control, and token management to effectively secure AI inference services and minimize risks associated with cost, latency, and data leakage. First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/02/prompt-control-is-the-new-front-door-of-application-security/
-
HackerOne ‘updating’ Ts&Cs after bug hunters question if they’re training AI
CEO lauds security researchers, insists they’re not ‘inputs’ First seen on theregister.com Jump to article: www.theregister.com/2026/02/18/hackerone_ai_policy/