Legal AI market accelerates as major vendors, firms and investors expand their bets
The latest wave of generative AI activity in the legal sector reflects how quickly the market has changed since late 2022, with developments in 2026 underscoring the pace at which vendors, law firms and investors are now moving.
Recent milestones include Thomson Reuters’ acquisition of Casetext, which was among the earliest major strategic generative AI deals in legal services, and which at the time drew significant attention across the legal technology market. Three years on, that transaction is being viewed as an early marker in a much broader shift rather than an isolated event.
This year has seen several large technology companies move into the legal vertical at scale, including OpenAI, Google, Anthropic and Microsoft. Other technology groups, including Palantir and Perplexity, have also been cited as part of the widening field of interest in legal applications of artificial intelligence.
Thomson Reuters has also unveiled its own large language model, Thomson, while Kirkland & Ellis, an Am Law 100 firm, has said it will build its own AI system. The moves point to a market in which both established legal information providers and major law firms are increasingly seeking to develop proprietary tools rather than rely solely on external products.
Investment activity has also continued to rise. Harvey has reportedly reached a valuation of more than $15 billion, while Legora, which raised about $10 million in May 2024, has since accumulated more than $1.4 billion in total investment over roughly two and a half years. The figures illustrate the scale of capital now flowing into legal AI businesses.
At the same time, the number of AI native law firms has grown from a small handful in 2025 to several dozen this year, with further launches expected. The broader legal innovation ecosystem has also expanded, with legal technology companies adding new features, innovation and legal operations teams pushing adoption internally, and more lawyers using AI tools in day to day practice.
Looking ahead, the next phase is expected to be shaped by frontier model developers, as well as emerging approaches such as agentic systems and vibe coding. While the sector is still far from a fully agentic legal environment, the foundations are being laid. For legal professionals, the practical implication is clear, AI capability is moving from experimentation towards operational planning, with further change likely over the next 12 months.