AI Conversations Compromised: Chrome Extensions Expose 900,000 Users in Massive Data Heist

On a quiet weekday, millions of Chrome users went about their business - some seeking help from ChatGPT, others coding with DeepSeek. But behind the scenes, their every word and search was silently siphoned away by two seemingly helpful browser extensions. The digital heist, now exposed, underscores the dark side of the AI boom: where convenience meets exploitation, and trust becomes a hacker’s greatest weapon.

The two rogue Chrome extensions - ‘Chat GPT for Chrome with GPT-5, Claude Sonnet & DeepSeek AI’ and ‘AI Sidebar with Deepseek, ChatGPT, Claude and more’ - hid in plain sight, masquerading as helpful productivity boosters. Both mimicked the design and branding of AITOPIA, a legitimate tool that lets users chat with large language models (LLMs) directly from a browser sidebar. But unlike their authentic counterpart, these clones had a secret: embedded code that harvested far more than “anonymous analytics.”

According to OX Security, the extensions covertly collected users’ full conversations with ChatGPT and DeepSeek, as well as every URL opened, search query entered, and even sensitive authentication tokens. For corporate users, this meant internal domains, proprietary code, and confidential strategies could be exfiltrated in real time - an open invitation for corporate espionage and identity theft. The extensions used the AI web development platform Lovable to host their infrastructure and mask their activity, making detection even harder.

The scale of the breach is staggering. With nearly a million installs, the attackers could have accessed a goldmine of intellectual property, customer data, legal communications, and business plans. Even personal users faced risks: identity theft, targeted phishing, and exposure of private conversations. OX Security warns that such data is often weaponized or sold on underground forums, amplifying the threat far beyond the initial breach.

Google has since purged the extensions from its Web Store, but the incident exposes a deeper issue: Chrome’s extension ecosystem remains vulnerable to sophisticated impersonation and social engineering. Users, lured by promises of AI-powered productivity, are left with the burden of vigilance - and the urgent task of uninstalling rogue add-ons before their secrets become someone else’s weapon.

The AI revolution promises speed and smarts, but as this breach shows, it also opens new doors to exploitation. In the race for productivity, trust is the most valuable - and most fragile - currency. For now, Chrome users are reminded: when it comes to browser extensions, convenience can come at a devastating cost.

WIKICROOK

• Browser Extension: A browser extension is a small add-on that enhances browser features but can also be misused by hackers to steal data or spy on users.
• Large Language Model (LLM): A Large Language Model (LLM) is an AI trained to understand and generate human-like text, often used in chatbots, assistants, and content tools.
• Exfiltration: Exfiltration is the unauthorized transfer of sensitive data from a victim’s network to an external system controlled by attackers.
• Authentication Token: An authentication token is a digital key that verifies your identity to apps or services, allowing secure access without re-entering your password.
• Corporate Espionage: Corporate espionage is the theft of confidential business information or trade secrets to gain an unfair competitive advantage.