Kyla Guru

Kyla Guru is Head of Model Cyber Safety at Anthropic, where she architects safety mitigations for AI agents and LLMs before public deployment, tracks cyber misuse patterns, and leads pre-deployment safety evaluations. Her mission: ensuring AI systems are secure by design, not afterthought. A Stanford-trained computer scientist (BS/MS), Kyla brings battle-tested expertise from offensive security and threat intelligence roles at SpaceX, Apple, MS-ISAC, and the U.S. Government. Her thesis pioneered using LLMs for threat actor attribution and technique identification—research that now directly informs her work securing frontier AI systems. Beyond being a technologist, Kyla is a grassroots activator, reshaping cybersecurity education as Founder/CEO of the 501(c)(3) Bits N’ Bytes Cybersecurity Education Corp. and co-founder of GirlCon Conference, which unites 700+ industry professionals and students annually to close tech’s gender gap. Now in its 8th year, GirlCon has become the premier tech conference for high school students globally. A sought-after voice on AI safety and cybersecurity, Kyla has delivered keynotes at TEDxChicago, RSA Conference (USA and Singapore), and Tableau Conference, advocating for people-first security solutions. Her unique perspective—spanning nation-state threats to AI safety—positions her at the forefront of defining how we build, deploy, and defend AI systems in an increasingly complex threat landscape.