Ddos 
Image: Anthropic
Recently, Anthropic introduced a Chrome-based extension for Claude Max users, designed to read active web pages and execute tasks based on user instructions. At present, the extension is available only as a limited research preview.
Once installed, the extension automatically activates the browser’s sidebar whenever Chrome is opened. This layout allows users to continue browsing within the main window on the left while interacting with Claude on the right, issuing commands or engaging in conversation without disrupting their workflow.
Anthropic has cautioned, however, that employing AI within the browser introduces the risk of prompt injection attacks. Malicious actors can conceal harmful instructions within websites, emails, or documents, thereby tricking the AI into carrying out dangerous actions without the user’s knowledge.
Such attacks could potentially lead to file deletion, data theft, or unauthorized financial transactions. According to Anthropic’s own red-team testing, these threats are not theoretical but demonstrably feasible. As a result, users of Claude for Chrome are urged to exercise vigilance.
For security reasons, Claude has been preemptively blocked from accessing certain categories of websites, including banking and financial services, investment and trading platforms, adult or pornographic sites, and cryptocurrency exchanges. Yet Anthropic admits that it cannot anticipate or identify every such domain, and encourages users to submit reports if they encounter unblocked sites of this nature.
In addition, Claude is strictly prohibited from engaging in stock or investment trading, bypassing CAPTCHA systems, processing sensitive personal data, or scraping and collecting facial imagery. Consequently, Anthropic recommends that only experienced AI professionals experiment with the extension, while casual users are advised to refrain from adoption at this stage.
Related Posts:
- Perplexity AI to Pay Publishers for Content in a Groundbreaking New Revenue Model
- Claude AI Integrates with Google Workspace
- Smarter Claude AI: Free Users Get Web Search, Voice Mode Coming Soon
- The AI Cold War: Anthropic Revokes OpenAI’s Claude API Access Over Terms of Service Dispute
- Anthropic Launches Claude Max Subscription with Higher Usage Tiers
Support Our Threat Intelligence
If you find our CVE report and cybersecurity news helpful, consider supporting our work.
Buy Me a Coffee
PayPal
