Skip to content
noticia
Tools

The right tool
for the matter.

Most of our work runs in Nuix Discover, with Claira for AI-assisted review inside it — one platform, one review environment, nothing moving between systems. If your team is already standardized on something else, we work there instead.

AI-assisted review that shows its reasoning.

Claira is document review AI built into Nuix Discover. Ask a question in plain language, test it against a single document, then scale the same prompt across thousands with Bulk Scan. Unlike tools that hand back a result and nothing else, Claira shows its reasoning at every step, so your team can audit, verify and defend every decision — without leaving the review environment they already use.

10x faster first-pass review with Bulk Scan
10,000 documents processed per hour
50%+ reduction in review cost on a typical matter
8 coding fields populated in a single scan

Key benefits

Review up to 10x faster

Bulk Scan works through thousands of documents an hour and writes results straight into your Nuix Discover fields in real time. What took days of manual review finishes in hours, and the calls stay consistent across the whole set.

Cut review cost by half or more

Review is the single largest cost in most matters. AI-assisted workflows take the weight off first-pass, privilege screening and coding — meaningful savings on every file, without giving up defensibility.

Same platform, no disruption

Claira appears inside your Nuix Discover environment as another review tool. No data migration, no new software to learn. Your team writes prompts, tests them and scales without leaving the platform.

Built for every review task

Relevance, privilege screening, objective coding, PII identification, chronologies, summarization, translation, fact extraction, QC. Start from a template or write your own prompt for whatever the matter needs.

From prompt to production in minutes

Three steps. No data migration, no retraining, no new tool for the team to learn.

1

Write a prompt

Describe what you need in plain language — relevance criteria, coding rules, extraction targets. Pick a template or write your own, and use Case Context to give Claira the background it needs to get it right the first time.

2

Test in the Prompt Lab

Run the prompt across a sample and see the results immediately. Adjust the wording and the criteria until the output matches what a human reviewer would have produced — all before committing to the full set.

3

Scale to thousands

Launch a Bulk Scan and Claira works through up to 10,000 documents an hour, populating your Nuix Discover fields live. Watch progress as it runs, and resume if anything interrupts it.

Works across every review task

Start from a template or write your own prompt. Twelve of the things we use it for most.

Relevance review

Identify responsive documents, tie findings to specific issues and surface the most critical material first, with confidence scoring behind each call.

Privilege screening

Screen for solicitor-client, litigation, common interest and settlement privilege, with structured reasons that feed straight into the privilege log.

Objective coding

Dates, titles, authors, document types and other metadata extracted automatically — up to eight coding fields in a single Bulk Scan.

Image analysis

Photographs, screenshots and diagrams described, objects identified and relevant detail pulled out of visual content.

Handwritten documents

Handwritten notes, logs and annotations read alongside typed material, processed inside the same scan. No manual transcription.

PII identification

Names, email addresses, account numbers and other personal information detected for breach response, notification or redaction prep.

Chronology generation

Dated timelines built from thousands of documents in one scan, in a consistent format that sorts, verifies and exports cleanly.

Summarization

Concise summaries for first-pass triage across a large collection, so the team can see what deserves a closer read.

Translation

Language detected automatically, with English or French summaries produced for multilingual collections — triage before paying for certified translation.

Fact extraction

Factual statements and actions about named individuals pulled out, to build witness profiles and investigation summaries consistently.

QC review

Documents re-evaluated against structured criteria and AI output compared against human coding, surfacing discrepancies for quality control.

Agent mode

An agent scoped to your matter that runs review tasks, answers questions about the file and surfaces patterns across what has been reviewed, in a secure persistent thread grounded in case context.

For the teams that carry the volume.

The same tooling adapts to very different pressures. What changes is who is short of what — headcount, hours, or infrastructure.

  • Law firms Large-scale litigation, regulatory responses and investigations. High volumes handled defensibly, key evidence surfaced earlier, and partners confident in the process — without adding headcount or servers.
  • In-house legal teams Rising volumes, limited resources, tight deadlines. Interrogate a large collection quickly, identify PII for breach response, support internal investigations and produce chronologies, with senior-led support behind you.
  • Government agencies Access requests, commissions of inquiry, regulatory files and litigation against the Crown. High volumes handled to a documented standard, with bilingual output and traceable results that stand up to public scrutiny.

The tool is half of it.

Claira is capable on day one, but the value comes from pairing it with people who understand legal workflow, defensible review and what a Canadian court expects to see. That is what we bring to every matter it runs on.

  • Implementation and setup We handle configuration, testing and go-live inside your Nuix Discover environment. No technical burden on your team.
  • Prompt design and training Your team asks in plain language; the built-in prompt generator does the translation. We then refine and test those prompts against your matter, criteria and coding requirements.
  • Support throughout the matter Available while the file is live to answer questions, refine prompts, review output and escalate anything that needs it. Same people, same standards.
  • Defensibility assurance QC processes validate AI output against human coding and surface discrepancies, so your team can certify the review with confidence.
  • Ongoing training As the tool evolves and your team gets more confident, we keep them current on new features, prompt strategy and emerging practice in AI-assisted review.

Common questions

What does Claira cost? +

It depends on the volume being reviewed and the scope of the matter. We give a clear written estimate before any work starts, so there are no surprises on the invoice. Tell us about the file and we will price it.

Can it be used outside litigation? +

Yes. It is built for the full range of review work — investigations, regulatory responses, access requests, internal reviews and arbitration. If a team needs to find critical information across a large collection quickly and defensibly, the matter type does not much matter.

How quickly can we start? +

Because it runs inside Nuix Discover, there is no new infrastructure to build and no lengthy setup. We manage implementation and configuration, and most matters are running in a very short window. Tell us your deadline and we will tell you what is realistic.

Is our data secure? +

Yes. Documents are hosted in Canada on infrastructure certified to ISO 27001 and SOC 2 Type II. Claira operates entirely within the Nuix Discover environment and the data stays in Canada. Document content is not retained after processing, and it is never used to train or fine-tune AI models.

Is AI-assisted review defensible in Canadian courts? +

Claira is built for transparency. Every result is traceable and the reasoning behind each assessment is visible to reviewers and supervising counsel. We apply QC processes to validate the output. As with any AI-assisted review, results must be verified by the supervising legal team before being relied on — and we recommend discussing the approach for your specific matter with us before starting.

Do we need to already use Nuix Discover? +

No. If you use a different review database, we can access the AI functionality inside Nuix Discover and overlay the results directly into your existing platform. You do not need to set up or maintain any of that infrastructure — we do it. If you already use Nuix Discover, it integrates in place and your team can start immediately.

What if we have questions mid-matter? +

You get the same senior-led support we provide on every engagement. Our team is available throughout to answer questions, refine prompts, review output and resolve issues — the same people, from scoping through to production.

See it on
your own data.

A working demonstration on a sample from a real matter, with the reasoning visible, and an estimate for the file in front of you.

Work with us

Your data stays your own. No cookies, no trackers, no cookie banner — analytics are aggregate and anonymous, and the only thing your browser stores is the language you picked.

Privacy notice →
no cookies