Perplexity launches Computer for Counsel — an AI agent for legal teams

Perplexity has launched Computer for Counsel — a specialized version of its Perplexity Computer agent, adapted for the needs of legal teams and law firms. The new system routes requests through more than 20 language models and delivers results with verifiable links to sources — which is critical for the legal industry.

What is Computer for Counsel

Computer for Counsel is an extension of the Perplexity Computer product, created specifically for the specifics of legal work. Unlike the base version, designed for a wide range of professional tasks, Counsel focuses on the needs of lawyers: analyzing case law, drafting and reviewing contracts, researching regulatory frameworks, preparing legal positions and memoranda. Architecturally, this is a 'multi-model agent layer': the system is not tied to a single language model, but dynamically routes each request to the most appropriate model from a pool of 20+ systems from different providers.

This approach allows the use of the strengths of different models — some handle the analysis of long legal documents better, others work more accurately with precedent search, and others generate texts in the required professional style.

Integrations and Ecosystem

The key advantage of the product is built-in integration with tools that lawyers already use every day:

• Midpage — a specialized AI platform for legal research and working with court cases
• MCP-connectors — an open protocol for connecting to external data sources, corporate document repositories, and repositories
• Microsoft 365 — Word, Outlook, Teams, SharePoint, Excel
• Internal firm knowledge bases — through the same MCP-connectors, the agent has access to accumulated internal documentation

The agent does not require constant switching between applications. A lawyer formulates a task in one interface, and Computer for Counsel independently accesses the necessary systems, gathers information, and forms the final response with references.

The Hallucination Problem and Its Solution

The main barrier to AI in legal practice is the unreliability of generated content. Language models can confidently cite non-existent court cases, invent statute numbers, or misquote actual documents. In a field where an error costs a lawyer's reputation and client money, this is unacceptable. Computer for Counsel solves the problem through the principle of mandatory citation: each agent response contains links to specific primary sources that a specialist can open and verify independently. The result is not just 'neural network output,' but a verifiable statement with an evidence base suitable for inclusion in an official document or legal position.

'Cited results that lawyers can verify' — that's exactly how

Perplexity formulates the key difference of Counsel from other AI tools.

Why Perplexity is Entering Legal

The legal market is one of the most conservative, but also one of the largest in terms of text and document work. Large firms and corporate legal departments spend significant resources on due diligence, legal analysis, and contract review — tasks that lend themselves well to automation with a reliable and verifiable tool. Perplexity already positions itself as an AI search engine with an emphasis on verified sources. Moving into legal is a logical development of this positioning: where the cost of error is particularly high, transparency and verifiability of results become not an additional advantage, but a basic requirement for trust in the product.

What This Means

AI agents in legal practice are transitioning from pilot projects to real working tools. Computer for Counsel demonstrates a formula that can work: multi-model routing plus strict citation plus integration with the corporate stack. How quickly large firms will move real work processes to agents, rather than just experiments, will be shown in the coming year.