Key facts

Founded
2022
HQ
Boston, United States / London, United Kingdom
Employees
—
Website
Funding stage
Series A · US$30m
LinkedIn
Further facts
  • Capital: Album VC led Mindgard’s US$30m Series A, announced on 12 August 2026, with Karma Ventures and existing investors .406 Ventures, Atlantic Bridge, IQ Capital and Lakestar.
  • Origins: Mindgard was spun out of Dr Peter Garraghan’s AI security research at Lancaster University in 2022 and raised its seed round in 2024.
  • Ecosystem: In September 2026 Mindgard joined technology programmes with Anthropic, NVIDIA, Microsoft, Google Cloud and AWS, including Anthropic's Cyber Verification Program.
  • Research: Mindgard's research has produced more than 150 publicly disclosed AI vulnerabilities, including flaws in Google's Antigravity IDE and OpenAI's Sora.

Focus areas

AI security Agentic AI Model risk AI governance

About

Mindgard builds offensive security for AI systems and agents. It was spun out of more than a decade of AI security research at Lancaster University led by Dr Peter Garraghan, and was founded in 2022. James Brear is Chief Executive Officer and Garraghan is Founder and Chief Science Officer. The company has raised a US$30m Series A and works with Fortune 2000 enterprises and AI developers. In September 2026 it joined technology programmes with Anthropic, NVIDIA, Microsoft, Google Cloud and AWS to stay close to the models and infrastructure it tests.

How they use AI

The platform acts as an autonomous red teamer: it discovers AI across an organisation, runs attacker-style reconnaissance and then attacks models, agents and applications to find exploitable flaws. Findings feed remediation guidance and runtime protection that defends AI systems in production. Its research team feeds new attack techniques into the platform; disclosures include a flaw in Google's Antigravity IDE and hidden instructions surfaced from OpenAI's Sora video generator.

Products and services

  • AI Recon — Discovery and reconnaissance of AI assets
  • AI Red Teaming — Automated attacks on models, agents and apps
  • AI Assessment — Risk assessment of AI systems
  • AI Runtime Protection — Defence for AI in production

Key people

  • James Brear — Chief Executive Officer
  • Dr Peter Garraghan — Founder and Chief Science Officer
  • Aaron Portnoy — Chief Product Officer
  • Rich Smith — VP Research
  • Fergal Glynn — Chief Marketing Officer

Investors

  • Album VC
  • Karma Ventures
  • .406 Ventures
  • Atlantic Bridge
  • IQ Capital
  • Lakestar