Malicious Chrome Extensions Stole Conversations From ChatGPT, Claude, Gemini, and DeepSeek
Researchers reported that several AI-themed browser extensions in official marketplaces covertly harvested user conversations from generative AI services including ChatGPT, Claude, Copilot, Gemini, and DeepSeek. G Data identified Urban VPN, Smart Sidebar: ChatGPT, Claude and DeepSeek, and AI Assistant/Chat AI as examples that remained outwardly functional while injecting scripts, intercepting requests, tampering with page content, or using hidden iframes to capture prompts and responses from visited AI sites.
Investigators said the stolen data included conversation IDs, hostnames, timestamps, and full chat histories, with some payloads encoded in Base64 and sent to external servers. The extensions reportedly had strong ratings, large user bases, and in one case a Chrome Web Store Featured badge, underscoring how official store vetting can miss dynamic malicious behavior such as remote iframe communications; researchers warned that AI-focused extensions now expose a broad attack surface and urged enterprises to enforce strict extension allow-listing, minimize permissions, and remove unnecessary add-ons from systems accessing sensitive AI platforms.
Get ahead of threats like this
Mallory correlates global threat intelligence with your attack surface — know if you’re exposed before adversaries strike.
How this story unfolded
2 events from the most recent confirmed update back to the earliest known activity.
MalExt Sentry exposes PromptSnatcher ad blockers stealing AI chats
Researchers disclosed that the browser extensions Smart Adblocker and Adblock for Browser were covertly intercepting and exfiltrating conversations from eight AI platforms while posing as legitimate ad blockers. The report linked both extensions to the same operator through shared infrastructure and code, and said the campaign affected about 90,000 users.
G Data reports malicious Chrome extensions stealing AI chat data
G Data reported that the Chrome extensions Urban VPN, Smart Sidebar: ChatGPT, Claude and DeepSeek, and AI Assistant/Chat AI were covertly collecting conversations from AI platforms and exfiltrating them to external servers. The report said the extensions targeted services including ChatGPT, Claude, Copilot, Gemini, and DeepSeek using script injection, network interception, and hidden iframes.
Related entities
Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.
Sources
3 references tracked. Mallory keeps watching after this page renders.
PromptSnatcher Ad Blocker Extensions Steal AI Chats From ChatGPT, Claude, and Gemini
cybersecuritynews.com
Open sourceMalicious AI Extensions Expose Private Conversations
securityonline.info
Open sourceMalicious Browser Add-Ons Target ChatGPT, Claude, Copilot, Gemini, and DeepSeek Users
cybersecuritynews.com
Open sourceSee the full picture, correlated to your attack surface.
Map indicators from this story to your assets and identify affected systems in minutes.
Every observed campaign, victim, and pivot linked to actors named in this story.
Malware, exploits, and IOCs connected to the activity described here.
YARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Get matching new stories delivered to your team as they break — not the next morning.
Ask questions about this story and take action on the answers.
Related stories
Other stories Mallory is tracking that overlap on entities, sources, or topic labels.
Malicious Chrome Extensions Steal ChatGPT and DeepSeek Conversations
Two rogue Chrome extensions, impersonating the legitimate AITOPIA AI sidebar tool, have compromised over 900,000 users by exfiltrating ChatGPT and DeepSeek conversations along with full browsing histories to attacker-controlled servers. The extensions, named "Chat GPT for Chrome with GPT-5, Claude Sonnet & DeepSeek AI" and "AI Sidebar with Deepseek, ChatGPT, Claude and more," request consent for "anonymous analytics" but covertly steal sensitive data, including proprietary code, business strategies, PII, and internal URLs. The malware operates by monitoring browser tabs, scraping chat content and session IDs, and sending Base64-encoded data to C2 servers every 30 minutes, exposing users to risks such as espionage, identity theft, and phishing. Researchers from OX Security discovered the threat, noting that the extensions remain available on the Chrome Web Store, with one losing its "Featured" badge after disclosure. The extensions also redirect users to each other if uninstalled, and their privacy policies are hosted on third-party sites to obscure their origins. The incident highlights the growing trend of browser extensions being used to capture AI chatbot conversations, a tactic dubbed "Prompt Poaching," and underscores the need for vigilance when installing browser add-ons, especially those requesting broad permissions under the guise of analytics or enhanced user experience.
Mar 21, 2026
Urban VPN Proxy Chrome Extension Harvests AI Chatbot Conversations
The Urban VPN Proxy Chrome extension, boasting over 6 million installs and a "Featured" badge on the Chrome Web Store, was discovered secretly collecting and exfiltrating user conversations with major AI chatbots, including ChatGPT, Claude, Gemini, Microsoft Copilot, Perplexity, DeepSeek, Grok, and Meta AI. Researchers found that a silent update in July 2025 (version 5.5.0) introduced hidden code that intercepts every prompt and response exchanged with these AI platforms, along with conversation identifiers, timestamps, and session metadata. The extension, which markets itself as a privacy and security tool, continued to harvest data regardless of whether the VPN service was active, betraying user trust and bypassing normal browser security boundaries. Captured data was sent to Urban VPN’s servers and subsequently sold to marketing analytics firms, notably BiScience, a known data broker. The malicious behavior was enabled by script injection that overrode browser APIs such as `fetch()` and `XMLHttpRequest()`, ensuring all relevant traffic was intercepted. The extension’s widespread distribution and high user ratings, combined with its presence on both Chrome and Edge marketplaces, amplified the impact of this breach. The incident highlights the risks posed by browser extensions with privileged access and the potential for abuse even among highly rated and officially endorsed add-ons.
Mar 21, 2026
Malicious and High-Risk AI-Powered Chrome Extensions Enable Account Hijacking and Phishing
Security researchers reported multiple risks tied to **AI-themed browser extensions** in the Chrome/Edge ecosystem, including active malicious campaigns. Malwarebytes identified **16 malicious extensions** (15 Chrome, 1 Edge) masquerading as ChatGPT “enhancers” that **steal ChatGPT session tokens**, enabling attackers to take over accounts and access conversation history and metadata; the extensions also exfiltrate additional telemetry (e.g., extension version/language and usage details) to help attackers profile victims and maintain longer-term access. Separately, Varonis described a new **malware-as-a-service** offering called **“Stanley”** that claims to reliably get **phishing-capable Chrome extensions** through Chrome Web Store review, using full-screen `iframe` overlays to present attacker-controlled login pages while the address bar continues to show the legitimate domain; it also advertises auto-install support across Chrome/Edge/Brave, a management panel, geo/IP targeting, and frequent C2 polling. In parallel with these overtly malicious cases, an Incogni study of **442 AI-powered Chrome extensions** found broad privacy and security exposure from over-privileged extensions (e.g., script injection and deep page access) and extensive data collection (52% collecting user data), highlighting that even popular tools (e.g., **Grammarly** and **QuillBot**) can present significant privacy risk due to the scope of permissions and data categories collected.
Mar 21, 2026