OpenAI has launched Astra for Law, a legal configuration of GPT-6 Astra with a specialist search index, legal instructions and governance controls.

This is not a contract lifecycle management platform. It does, however, move a powerful set of research, analysis and drafting capabilities underneath the platforms that sell contract intelligence.

That changes the competitive question for CLM.

What OpenAI actually launched

Astra for Law combines OpenAI's frontier model with a legal search index and instructions designed for professional legal work. OpenAI says API customers including Harvey and Legora will be able to build on it, while selected law firms can access it through a Trusted Access programme before broader API availability.

The legal search index covers US case law, statutes, regulations, court rules and administrative decisions. OpenAI also launched 26 ecosystem plugins connecting ChatGPT with specialist legal tools, including Relativity and Clio.

For contract work, the more important examples are not basic clause summaries. Sullivan & Cromwell has built an agreement analyser that combines negotiating playbooks and precedents to identify risks across provisions, propose redlines and draft advice. Ropes & Gray has developed a diligence system that traces findings back to source material and identifies contracts requiring notice or consent.

The CLM feature moat is getting thinner

For years, CLM vendors could differentiate through extraction, search, drafting, redlining and conversational access to agreements. Those capabilities are increasingly available from the underlying model and tool ecosystem.

That does not make CLM redundant. It raises the bar.

A credible contract platform still needs to connect the intelligence to:

  • Authoritative contract records and permissions.

  • Approval policies, deviations and escalation paths.

  • Supplier, spend and operational context.

  • Obligations after signature.

  • A defensible audit trail showing the sources and decisions behind the output.

My view is that the value is moving away from generating contract text and towards knowing which context applies, what action follows and whether that action was completed.

Buyers should not confuse a benchmark improvement with reliability

OpenAI reports that Astra for Law passed the overall correctness check on 54% of 200 questions in a private validation set, compared with 38.7% for GPT-6 Astra using web search alone. That is a material relative improvement, but it is also a reminder that specialist configuration does not remove the need for verification.

Buyers should test performance against their jurisdictions, playbooks, agreement types and risk tolerances. They should also ask where the legal content comes from, how current it is, which outputs carry citations and how confidential information is retained or reviewed.

The bottom line

OpenAI is giving legal technology companies a stronger foundation on which to build. That benefits CLM vendors that can turn intelligence into governed enterprise action. It creates a problem for products whose differentiation is little more than a chat interface over a contract repository.

The model is becoming infrastructure. The workflow, context and evidence are where contract platforms now have to earn their place.

Sources