Trent AI secures $13 million to develop multi-agent security.

      The London startup came out of stealth on April 7, unveiling a layered agentic security solution and securing a seed round funded by LocalGlobe and Cambridge Innovation Capital. Among its co-founders is a Cambridge professor who previously served as Amazon’s director of machine learning.

      Trent AI, which specializes in agentic security, has successfully raised $13 million in its seed round and has announced its public launch. LocalGlobe and Cambridge Innovation Capital led the round, with contributions from angel investors such as Joaquin Quiñonero Candela, a technical staff member at OpenAI; Avinash Bhat, a Director at AWS; Ippokratis Pandis, a Distinguished Engineer at Databricks; and Tony Jebara, formerly Vice President of Engineering and Head of AI/ML at Spotify.

      Founded in 2025, Trent AI is set to officially launch on April 7, 2026. The company aims to fill a structural gap that has grown as businesses adopt autonomous AI agents at a pace that outstrips the adaptability of their security frameworks. Its product is a multi-agent security platform tailored for agentic environments, instead of simply adapting conventional static-rules security tools for AI applications.

      The platform operates with four specialized agents that run concurrently: Scan agents that monitor code, infrastructure, dependencies, and runtime behavior to identify risks; Judge agents that classify and prioritize signals based on actual business impact rather than fixed rules; Mitigate agents that automatically patch vulnerabilities and validate fixes; and Evaluate agents that monitor risk trends over time and measure against standards. This feedback loop between the layers is intended to enhance the accuracy of each cycle.

      The founding team is intentionally designed to combine academic expertise with operational experience. CEO Eno Thereska previously held the role of Distinguished Engineer at Alcion (acquired by Veeam), AWS, and Confluent. Co-founder Neil Lawrence is the DeepMind Professor of Machine Learning at the University of Cambridge and has also served as Director of Machine Learning at Amazon. The third co-founder, Zhenwen Dai, worked as a machine learning scientist at AWS and was a Senior Research Manager at Spotify. A Deloitte survey referenced by Trent AI revealed that although 74% of companies plan to implement agentic AI within the next two years, only 21% currently possess a mature governance model for autonomous agents, a gap the company explicitly aims to bridge.

      Design partners such as Canopy, Commscentre, ML@Cam, Qbeast, and Weblogic are currently utilizing the platform. Trent AI is a partner member of OWASP, the Open Worldwide Application Security Project, and is a startup partner of Carnegie Mellon University’s CyLab Venture Network. Additionally, the product features an open-source security agent for OpenClaw. Saul Klein, co-founder and executive chairman of Phoenix Court, which houses LocalGlobe, remarked that this is “the right time to establish the long-term foundations of security for agentic systems.”