Mindgard

What the company sells

Mindgard sells software that maps AI systems for exposure, red-teams AI models and agents for vulnerabilities, and applies runtime protection against attacks. It was founded in 2022 on research from Lancaster University and is headquartered in Boston, with a second base in London. Mindgard is privately held, having raised a Series A funding round in 2026.

Categories

• AI model and agent security (primary category)
• AI firewalls and guardrails

Headquarters

Boston, United States

Founded

2022

Ownership

Privately held

Website

mindgard.ai

Products and services

• AI Recon & Discovery

  Maps the AI attack surface across models, agents and applications using attacker-style reconnaissance.

• AI Red Teaming

  Runs automated adversarial testing against AI agents and systems to find exploitable vulnerabilities and report on AI risk compliance.

• Runtime AI Protection & Response

  Provides continuous runtime monitoring intended to find, fix and stop attacks on AI systems before impact.

Sources

1. MindgardUndated page, accessed September 27, 2026 · Tier A: the company, a regulator or a filing
2. MindgardUndated page, accessed September 27, 2026 · Tier A: the company, a regulator or a filing
3. SecurityWeekPublished August 12, 2026 · Tier B: trade or business press

Recorded September 27, 2026 from the sources above. Descriptions are in our own words; we do not rank or recommend vendors, and no vendor pays to be listed or to change what we write. Disclosure: until the publisher confirms it has no business relationship with this company, treat one as possible (how the network works). Spot a factual error? Request a correction.