Ricardo Prieto · AI safety, security and community
AI safety,grounded in security.
I'm Ricardo Prieto, a cybersecurity practitioner focusing my career on AI safety. I want to help make increasingly capable AI systems useful, secure and subject to meaningful human control.
PhiloCyber is where I share my writing, talks and resources, and document how I am contributing to AI safety: through security, programs, community and communication.
Useful AI • Safety • Human control
Security experience • Community building • Public work
At a glance
- years in offensive, application and AI security
- 7+
- AI Security functions built from scratch
- 2
- members in the security community I co-founded
- ~2,000
- community events with ThreatX Security
- 34
- public talks, from Ekoparty to universities
- 7
AISB London 2026 • BlueDot AGI Strategy • Apart Research judge • AI risk on radio and in the press
01Direction
London made the purpose clear.
I was already working on the security of AI applications when I attended the AI Security Bootcamp (AISB) in London. The experience made me think more deeply about the safety of increasingly capable systems, the evidence behind deployment decisions and how people can retain meaningful control. I came back believing I had found the work I want to dedicate my life to.
PhiloCyber is changing with that decision. The security experience that brought me here is the foundation I want to build on.
Read the full story02Work
Selected work
Writing, talks and practical resources, with a clear distinction between what is available and what is still being developed.
- AvailableArgusTMA local-first multi-agent tool that turns architecture context into a threat-model draft you can inspect and challenge.
- Public draftAI Red Teaming Field GuideA field reference organized by attack surface for authorized testing of AI-enabled systems.
- Talk
October 2025MCP Security for Enterprise OrganizationsA public presentation of practical MCP security work, delivered at Ekoparty 2025 — AI Resilience Hub village.
03Writing
Selected writing
Three pieces selected from the archive for their connection to the work I'm carrying forward.
04Now
What I'm focusing on now
I'm looking for my first role or programme in AI safety, in whichever branch I can be most useful: security, programs, field-building, operations or research. Meanwhile I keep the public resources going, strengthen my machine-learning foundations and bring AI risk into public discussion in Argentina.
05Contact
Have a joint project, role, interview or useful piece of feedback in mind?
I'm open to AI safety roles, programmes and fellowships (research, programs, field-building, operations or security), plus speaking, press, community conversations and feedback on the guide and tools. PhiloCyber is no longer taking new consultancy engagements.
Contact me