$7.3m · Seed · Cybersecurity · London, UK
London-based HelmGuard has raised $7.3m in a seed round co-led by Infinity Ventures and Frontline, with participation from FinTech Collective, Stage 2 Capital and Entrepreneurs First. The company builds AI agents that collect and assess security and compliance signals directly from source systems, aiming to replace periodic questionnaires with continuously updated risk evidence.
The funding will support engineering and go-to-market hiring, further development of HelmGuard's runtime agent-assurance layer and its Verified Risk Network, and expansion in the US through New York and San Francisco alongside its London headquarters. The company says its platform serves regulated organisations in financial services, insurance, healthcare and industrials.
HelmGuard brings unstructured documents and live operational data into one risk view, then uses specialised agents for work such as third-party risk reviews, control-gap assessments and assurance of other AI agents. The company says this can reduce assessment cycles from weeks to hours and cut assessment time by up to 80%.
Continuous assurance has to earn enterprise trust
The commercial bet is that fresher evidence is more useful than a faster version of the annual compliance questionnaire. If HelmGuard can connect claims to live source data, a buyer can revisit a vendor or control as conditions change instead of waiting for the next certification cycle. That can turn assurance from a periodic reporting task into an ongoing operational workflow, increasing the value of integrations and verified history over time.
That model also raises the standard HelmGuard must meet. Regulated customers need traceable reasoning, reliable integrations and clear boundaries for human review before they can rely on automated assessments. A customer account from TELUS Health says the organisation is building its GRC programme on HelmGuard across multiple services, products and jurisdictions; it describes API integrations as coming online, so the larger promise of continuously substantiated compliance still depends on implementation.
The planned agent-assurance layer extends the same logic to software that acts at runtime. The opportunity is to verify how agents behave against an organisation's controls, but adoption will depend on whether HelmGuard can make those judgements explainable enough for security teams, auditors and business owners to act on them.
