Bitdefender launches free VPN for AI agents in beta

Bitdefender has launched VPN for AI Agents in public beta. The free tool is initially available on macOS.

The product gives each AI task a separate temporary internet connection instead of routing traffic through the user's own connection. The connection is discarded when the task ends, with no carry-over between interactions.

The launch comes as security groups and researchers examine how autonomous AI tools act online on behalf of users. The service is designed to create a clear boundary between a person's network identity and an AI agent's activity.

According to the Cloud Security Alliance, AI agents often operate in an "identity gray area", using workload identities, shared service accounts, or a human user's credentials. Bitdefender's offering aims to separate those actions from the user's IP address during a single prompt or task.

The software works automatically with Claude Desktop, Cursor, Codex, and OpenCode. It requires no manual configuration and remains inactive until an AI agent calls on it for a task.

At that point, the service creates what Bitdefender describes as a disposable, isolated container for the prompt. Once the work is complete, the container and connection are removed.

This differs from a conventional virtual private network, which usually maintains one ongoing tunnel for all device traffic. Here, the private connection is limited to a single AI action, while the rest of the device continues to use its normal connection.

The service is intended to protect network transport and mask the AI agent's IP address, but it does not cover the content of prompts, which remain between the user and the AI provider.

Bitdefender says this setup could help with tasks such as region-specific research, browsing over public Wi-Fi, and comparing information or prices across markets. It may also allow agents to complete work across multiple regions in one pass rather than requiring separate manual checks.

Concern about AI and personal data remains high among consumers. Bitdefender cited Pew Research Centre findings showing that 71 per cent of adults in the US believe increased use of artificial intelligence will make their personal information less secure.

Growing scrutiny

The launch reflects a broader shift in cyber security as AI systems move from generating text to carrying out actions online. As these systems begin to browse, research, and transact with less direct oversight, companies face growing questions over how to identify and contain that activity.

In enterprise settings, the issue is often framed around access control, audit trails, and identity management. In consumer software, the concern centres more on whether an AI assistant should appear online as an extension of its user or as a separate actor with its own limited access and network footprint.

Bitdefender has chosen the latter approach. Instead of placing all traffic from a device inside one persistent private tunnel, it isolates internet access at the task level for an autonomous assistant.

The service is being released in English during the public beta period and is free to use. Support for Windows and other operating systems is due to follow, though Bitdefender has not given further detail on timing.

Bitdefender is entering the market with an established consumer VPN business. Its VPN product was recently named "Reliable for VPN" in Finder's Customer Satisfaction Awards.

Ciprian Istrate, Senior Vice President of Operations, Consumer Solutions Group, at Bitdefender, said the growth of AI assistants meant security tools had to account for software acting directly on behalf of users.

"AI agents are quickly becoming an extension of the people who use them, showing up in both our workdays and our personal lives," said Istrate. "That means security can no longer stop at protecting the person behind the screen; it has to extend to the agent itself acting on their behalf. With Bitdefender VPN for AI Agents, we're giving students, workers, and everyday consumers the confidence to let their AI agents work freely, knowing their identity and activity stay protected."