344
Productivity & Workflow355
Automation & Workflow225
Software Development251
Marketing & Growth192
AI Infrastructure & MLOps175
Writing & Content Creation203
Data & Analytics142
Photography & Imaging156
Design & Creative170
Customer Support132
Sales & Outreach125
Voice & Speech135
Education & Learning131
Operations & Admin87
Large law firms are starting to build custom AI software, not just buy it, to protect know-how and offer services competitors cannot copy.
In short: More large law firms are building custom AI tools, not just buying standard software, to differentiate their services.
After several years of testing AI, many law firms are now deciding what comes next. A growing number want AI that can handle routine work, like drafting contracts and searching past cases, and can also help them serve clients faster.
This demand has created a busy market for legal AI suppliers, including Harvey and Legora, plus long-time providers like LexisNexis and Thomson Reuters. Research firm Grand View Research estimates the global legal tech market could reach about $70bn by 2033, up from around $31bn in 2026.
Now, some international firms are choosing to build their own AI platforms, or heavily customize what they buy. In May, Kirkland and Ellis said it set aside $500mn to create its own AI platform, instead of relying only on tools that its competitors can also use. Freshfields has partnered with Anthropic and will use its Claude AI across the business, with early access to new legal-focused features. Goodwin Procter said it will spend about $25mn a year to buy and develop its own AI.
Observers say the market is splitting. Many firms will use standard AI for everyday work. A smaller group will develop bespoke tools for complex, sensitive, and higher-value matters, like building a tailored calculator instead of using a generic spreadsheet.
About 20% of large corporate law firms are experimenting with building or personalizing legal AI, according to Brian Tang at the University of Hong Kong. If more firms try to build tools but lack in-house tech teams, new outside service providers may emerge to help with setup, security, and confidentiality.
Source: Financial Times