Your deal data.
Your control.
The questions your legal and infosec teams ask first: where did this finding come from, who approved it, and what happens to our data.
Run it in our boundary, or entirely in yours.
Private cloud
Secure SaaS
Deployment options, security architecture and integration with your existing data room are covered in a partner-led technical session as part of onboarding. Where your risk function has a standard assessment process, we work through it directly rather than asking you to accept assurances.
What happens to deal information.
Read these as deployment-dependent. Which of the controls below apply, and how they are configured, follows the deployment model and the contract agreed at onboarding. They are described here to support a security review, not as a certification, and are confirmed in writing before any deal information is loaded.
Secure document ingestion
Documents are transferred into a controlled environment scoped to the specific deal. Ingestion routes and supported formats are confirmed during technical onboarding.
Encryption in transit and at rest
Deal information is encrypted while moving and while stored. Cipher standards and key management are covered in the technical review.
Client data isolation
Each client's deal information is segregated. Evidence from one engagement does not inform another under any configuration.
Retention and deletion
Retention periods and deletion procedures are agreed as part of engagement, including what happens at the end of a deal or a licence term.
Hosting and data residency
Hosting region and data-residency requirements are confirmed before onboarding, including where a specific jurisdiction is required.
Model training
Client data is not used to train shared models. Your deal information informs your deal.
Deal teams are not one undifferentiated group.
User and role-based access
Access is granted per deal and per role, so team members see the evidence and outputs relevant to their part of the process.
Audit trails
Activity across the workflow is recorded — what was ingested, what was analysed, what was reviewed and by whom.
NDA and data-processing agreements
Engagements are covered by non-disclosure and data-processing agreements executed before any deal information is transferred.
The design question is not whether AI can be wrong. It is what happens when the evidence does not support a conclusion.
Evidence, inference and open questions stay distinct
A statement supported by a source document, an inference drawn across sources, and a gap in the record are three different objects in the output — not blended into one confident sentence.
Weak evidence is flagged, not smoothed over
Where evidence is missing, partial or contradictory, it is raised as an issue or an evidence gap rather than presented as a finding.
Findings link back to their source
Material findings and red flags connect to the supporting document, page, table or extract wherever it is available, so a reviewer can check the basis rather than take it on trust.
Every output is a draft until a named person on the deal team approves it. Interpretation, challenge and the final recommendation are human responsibilities, and the platform is built to make that review possible rather than to make it unnecessary.
What DiligenceIQ does not claim.
Overstated capability is a risk in itself. These are the boundaries, stated before a procurement process has to find them.
It does not complete diligence
It produces drafts and structured evidence for human review. It does not issue a professional opinion, sign off on a transaction or replace the deal team's judgement.
It does not replace specialist due diligence
Formal financial, legal, tax, regulatory, cyber and other specialist due diligence — and independent valuation where required — remain with qualified professionals.
It does not guarantee every risk is found
It widens document coverage and surfaces potential red flags earlier. It cannot warrant that every risk in a transaction has been identified.
It does not certify what it has not been given
Findings reflect the evidence available. Where a data room is incomplete, that is reported as an evidence gap rather than resolved by inference.
Specific controls, certifications and contractual commitments are confirmed in writing during onboarding. We would rather set them out in a technical review than assert them on a web page.
Put DiligenceIQ to the test.
Run it against a deal you have already completed, or start with one workstream on one live transaction. Compare time saved, evidence coverage, analyst effort, finding quality and IC readiness.