Meet us at the Startup area, and at our workshop on Wednesday 7 October, 4:30 pm. Grimaldi Forum, a ten-minute pitch.
Book the workshopOur Assises workshopAGENTIC CYBER-GRC
Mindlapse is an operating system for cyber governance: signals come in, your enterprise context gives them meaning, and a governed agentic core turns them into decisions you can defend. It keeps running when nobody is logged in.
DOZENS OF FRAMEWORKS · THOUSANDS OF CONTROLS · MAPPED ONCE, REUSED EVERYWHERE
ENTERPRISE DEPLOYMENTS
Figures as of September 2026. Customer names are covered by non-disclosure agreements.
THE SHIFT
Regulators, boards and customers no longer accept a binder of policies as proof. They ask the only question that matters: is it actually true, right now?
FOUR TOOLS AND A BINDER
ONE OPERATING SYSTEM
OUTCOMES
Measured with our design-partner CISOs, figures under continuous validation.
THE OPERATING SYSTEM
Signals at the bottom, your enterprise context above them, ERNEST in the middle, and the modules your teams work in on top. Each layer feeds the next, which is why an overnight CERT advisory can end up moving a risk score before your first coffee.
01
Official threat feeds and the CVE catalog land every morning, beside your ratings, ticketing and procurement connectors. Evidence you upload is validated by AI, then reviewed by a human.
Official threat feeds Connector marketplace Ratings, ticketing, procurement
02
Your entities, business value chain, systems and classification scales are first-class objects, not tags. A knowledge graph and a retrieval layer let every answer cite where it came from, inside the access scope of whoever asked.
Scoped access Value chain Knowledge graph
03
One agentic engine proposes likelihoods, scenarios, treatments, control mappings and questionnaire answers, each specialized for the decision at hand. Ernest shows its plan step by step, cites its sources, and says plainly when the data does not support a conclusion.
Plan step by step Sourced answers Human sign-off
04
Risk Intelligence, Compliance and Third-Party Risk Management run on that same core, and the Cyber Cockpit consolidates them. Thousands of cross-framework mappings mean one answer can carry you across several regulations at once.
Every module One cockpit One shared core
ONE PLATFORM · THREE MODULES · ONE COCKPIT
Three licensable modules on the same core, and one consolidated view for the people who have to answer for all of it.
See risk as it forms
Every risk scored inherent, residual and target, on the methodology you practice, under an appetite the board governs.
Learn more
Prove it, continuously
Frameworks, controls and policies in one atlas: comply with one regulation and see exactly how far that carries you into the next.
Learn more
Trust your ecosystem
A Supplier Hub where partners enroll on a portal, remediation becomes a joint plan, and every Trust Grade is recomputed overnight.
Learn more
Consolidated posture across entities, executive summary, regulatory status on NIS2, DORA and GDPR, and board packs exported to PDF or PowerPoint straight from the live data.
Learn more
GOVERNED AUTONOMY
Agentic does not mean unattended. Four mechanisms make the autonomy accountable, and they are enforced in the product rather than promised in a slide.
The platform proposes, a person decides. Accepting, correcting or rejecting a suggestion is an explicit act, recorded with the person who made it, and every object keeps its origin: human, AI-suggested or hybrid.
Every agent works inside a bounded scope: what it may read, what it may do and how much it may consume are set per organization and per user, and enforced by the platform rather than by the prompt.
AI output is evaluated automatically before it reaches a reviewer, and every suggestion arrives with the confidence the engine puts on it.
Every AI action is traceable: who asked for what, on which basis and with which result, in a ledger your teams and your auditors can review and replay.
CYBER COLLECTIVE LAB
6 editions since December 2024: the Cyber Collective Lab brings together 30+ CISOs from large European enterprises to challenge, test and shape the platform. It isn’t a sales pitch; it’s a working session.
RECOGNIZED BY
FROM THE NEWSROOM
Events & conferences ·
Mindlapse is a candidate for the Trophées Cyber at Cybernight 2026, organised by Républik Cyber. The Grand Oral was held on 8 September, behind closed doors in front of the jury. What is left is the date that now matters: Monday 28 September, at the Théâtre Mogador in Paris, where the winners will be announced on stage.
Events & conferences ·
Mindlapse takes the stage at Les Assises de la cybersécurité in Monaco on Wednesday 7 October, 4:30pm. Ten minutes, one idea, one live demo: a critical third party enters the scope, and everything downstream recalculates.
Press releases ·
For its 8th edition, the radar published by Wavestone with Bpifrance maps 234 French cybersecurity startups. Mindlapse appears on it, in the governance, risk and compliance area.
READY WHEN YOU ARE
Thirty minutes with our team, on your use cases. NIS2, DORA, CRA or AI Act: bring your hardest questions.
Our host’s audience measurement (Vercel) uses no cookies and is not covered by this choice. Google Analytics and marketing trackers stay off until you say otherwise. Read the cookie policy
