What Mylo AI can and can't do

Mylo AI is the assistant built into every Mylo Prime plan, at no per-question charge. It’s most useful when you treat it like a very fast junior colleague who has read your firm’s matters — great at retrieval, drafting, and summarizing; always subject to your review.

What it’s good at

• Answering questions about your matters. “What’s the status of the Herrera matter?” “When is the next deadline on anything for this client?” Mylo answers from your firm’s own data.
• Drafting. Letters, memos, and — on litigation plans — pleadings, built on your matter’s facts and your firm’s letterhead. Everything lands as a draft for your review.
• Summarizing. Long transcripts, document sets, and intake calls, condensed to what you need to know.
• Finding things. Documents, emails, time entries, and calendar items across the firm, from one question.

What it deliberately won’t do

• Send things without you. Client-bound email drafts wait for a human to approve them.
• Practice law. Mylo drafts and researches; a licensed attorney at your firm reviews, signs, and remains responsible. The platform is built around that boundary, not despite it.
• Guess citations. Legal citations in Mylo’s drafts run through verification — see Verifying citations with Mylo Cite.

Getting better answers

Ask specific questions with matter or client names in them. “Summarize yesterday’s consultation with Dana Cole” beats “what happened yesterday.” If Mylo says it can’t find something, it usually means the underlying record doesn’t exist yet — the assistant only knows what’s in your workspace.

When Mylo is wrong

It happens — AI output is a draft, not an oracle. If Mylo gives an answer that’s wrong about your own firm’s data (a wrong balance, a missed matter), that’s a bug we want to hear about: report it with the question you asked and roughly when you asked it.