Curia resource · Ontario litigation

AI discovery questions for Ontario litigators.

A source-grounded workflow for examination preparation that keeps questions, documents, undertakings, and lawyer review connected to the matter record.

Try Curia free See matter workflows

Source-linked questions Undertaking follow-up

Quick answer

AI can draft the outline. Counsel owns the examination plan.

Ontario litigators can use AI to group issues, prepare first-pass questions, identify document gaps, and track undertakings. The useful version keeps every important question tied to a source and a lawyer-reviewed objective.

Curia workflow

Discovery prep works best inside the matter file.

Curia connects documents, timelines, research, and drafts so counsel can move from a record issue to a question, an undertaking, a transcript note, and the next task.

Explore matters

Why it matters

Discovery questions need a record trail, not just a prompt

Examination preparation often requires counsel to move between pleadings, productions, chronologies, expert materials, correspondence, and gaps in the document record. AI can help sort that material into question groups, but a generated outline is only useful when the lawyer can see why each question exists.

Canadian professional guidance on AI use emphasizes confidentiality, competence, supervision, verification, and lawyer responsibility. Those controls matter in discovery preparation because questions may reveal work product, touch privileged material, or shape undertakings and follow-up steps.

The safer workflow is to use AI for organization and first-pass drafting, then require counsel to test the outline against the record, the examination objective, and the practical needs of the file.

Discovery workflow

A six-step AI question workflow.

The goal is not to outsource judgment. The goal is to make examination preparation easier to inspect, refine, and reuse across the matter.

01
Start from the pleadings and issues.
Frame the question plan around the claims, defences, damages theory, admissions needed, and facts that remain disputed. AI should organize counsel's theory of the examination, not invent one.
02
Load the record by source type.
Separate pleadings, productions, correspondence, transcripts, expert materials, timelines, and prior answers so the question set can stay tied to the material counsel has actually reviewed.
03
Generate issue-based question groups.
Ask AI to draft clusters for background, liability, causation, damages, mitigation, credibility, documents, missing records, and undertakings, then refine each cluster against the file.
04
Attach source anchors and objectives.
Each important question should show the source that prompted it and the reason it matters, such as confirming a date, testing an inconsistency, explaining a gap, or setting up a document request.
05
Review for privilege, proportionality, and tone.
Counsel should remove questions that are unsupported, overbroad, privileged, duplicative, needlessly argumentative, or not useful for the examination objective.
06
Turn answers into follow-up work.
After discovery, use the transcript and undertakings to update the matter timeline, identify missing documents, draft follow-up questions, and prepare lawyer-reviewed next steps.

Question fields

What to capture before trusting an outline.

A useful discovery outline should show more than a list of questions. Counsel needs the issue, source, objective, expected document, and follow-up logic behind the draft.

Issue

The claim, defence, damages topic, credibility concern, or procedural point the question is meant to test.

Question objective

What counsel needs the answer to confirm, deny, narrow, explain, or preserve for later use.

Source anchor

The pleading, production, timeline event, correspondence, transcript line, expert report, or document reference behind the question.

Expected exhibit or document

The file material that may need to be put to the witness or requested through an undertaking.

Follow-up trigger

The answer, refusal, inconsistency, or missing document that should prompt a follow-up question or undertaking request.

Review note

Counsel's judgment about wording, privilege risk, relevance, tone, and whether the question belongs in the final outline.

Guardrails

Keep discovery preparation sourced and lawyer-led

Do not use AI to create facts, allegations, or questions that are not grounded in the record.
Do not rely on a generated question list without lawyer review for privilege, relevance, proportionality, and file strategy.
Do not paste sensitive matter material into tools that are not approved for confidentiality and access controls.
Do not treat a public article or generic prompt as advice about how to conduct a specific discovery.
Do not let a clean AI outline hide uncertain dates, missing productions, inconsistent evidence, or unanswered undertakings.

Related guide

Start with a cleaner document record.

Discovery questions improve when productions, correspondence, transcripts, and expert materials are classified before the outline is drafted.

Read document review guide

Next step

Use timelines to test the question plan.

A sourced chronology helps counsel spot missing dates, inconsistent answers, and follow-up areas before and after the examination.

Read timeline guide

FAQ

AI discovery question questions

Can AI help draft discovery questions for Ontario litigators?

AI can help organize the record, group issues, draft first-pass question outlines, and identify gaps for lawyer review. Counsel should still check the questions against the pleadings, productions, privilege concerns, and the specific examination objective.

What should an AI-assisted discovery question outline include?

A useful outline should include issue groups, draft questions, source references, expected documents, follow-up triggers, possible undertakings, and review notes that explain why each question matters.

How should lawyers verify AI-generated discovery questions?

Lawyers should confirm that each important question is supported by the record, tied to a legitimate file objective, worded appropriately, and reviewed for confidentiality, privilege, relevance, and proportionality before use.

Can AI help after examination for discovery?

Yes. After lawyer review, AI can help summarize transcript answers, update timelines, track undertakings, identify inconsistencies, and draft follow-up lists tied to the transcript and source documents.

Prepare discovery with questions tied to the record.

Use Curia to connect documents, timelines, research, undertakings, and draft work product in one verification-first workspace for Canadian legal teams.

Start free

curia.ca · Toronto, ON

Sources

This resource is general legal-technology information for lawyers and law firms. It is not legal advice and does not recommend any discovery strategy, question wording, undertaking position, filing step, or outcome for a specific file.