Using AI to generate or complete legal text — contracts, motions, briefs, correspondence — based on lawyer prompts or templates; lawyer reviews and edits before use.
Last reviewed: 2026/05/19
AI contract drafting and review inside Microsoft Word for transactional lawyers.
Thomson Reuters' GPT-backed legal research and drafting with Westlaw integration (relaunched as CoCounsel Legal, 2025).
Move from this definition to role-based legal AI shortlists and the selection criteria that matter for each type of legal team.
Am Law 200 and global firm workflows: accuracy at scale, security compliance, and matter-level auditability.
Reviews built for 2–20 attorney firms: collaborative workflows, mid-range budgets, limited IT overhead.
60 legal AI tools vetted for the solo lawyer: tight budget, no IT team, billable-hours pressure.
Last reviewed: 2026/05/19. Definitions are written by the LawyerAI Editorial team. Commercial relationships are disclosed and do not determine editorial scores or conclusions. See our Sponsorship & Affiliate Disclosure.
AI-assisted drafting is the use of artificial intelligence to generate or complete legal text — including contracts, motions, briefs, demand letters, and client correspondence — based on lawyer-provided prompts, templates, or uploaded precedents. The AI produces a starting draft; the lawyer reviews, edits, and takes responsibility for the final work product. The AI is a drafting accelerant, not an autonomous author. Quality and accuracy of AI drafts depend on the specificity of the prompt, the quality of the underlying model, and how well the tool has been trained on the relevant legal domain.
Drafting is among the most time-intensive legal tasks and among the most amenable to AI acceleration. A first draft of a commercial lease, employment agreement, or demand letter that previously took two to three hours can be produced in minutes with a well-crafted prompt and a capable AI tool.
The time savings compound across high-volume work. A transactional practice that produces dozens of similar agreements each month can use AI drafting to compress associate time on initial drafts, redirecting lawyer effort to negotiation, issue spotting, and client advice.
The risk is hallucination. AI models can generate plausible-sounding but legally incorrect provisions — invented statutory citations, inapplicable jurisdictional rules, or clause combinations that conflict internally. A draft that reads fluently may contain substantive errors. Every AI-generated draft requires lawyer review before filing, sending, or signing.
Drafting capability ranges from general-purpose LLM integration to purpose-built legal drafting tools. Spellbook integrates directly into Microsoft Word, suggesting clause completions and drafting from precedent libraries trained on commercial contract language, reducing the friction of switching between drafting environments.
Harvey supports long-document drafting with firm-specific style preferences and precedents, producing drafts that align with a firm's established positions rather than generic market language. CoCounsel structures drafting tasks around specific document types with defined inputs, reducing the prompt-engineering burden on the lawyer.
Tools differ significantly in how they handle jurisdiction-specific language, whether they support firm-specific precedent libraries, and how transparently they surface uncertainty. Buyers should test drafting tools on representative internal precedent types before deployment.