BlueDot Stories

Quitting FinTech for AI Safety — Milos Borenovic

In this episode, Dewi speaks with Milos Borenovic, Chief Product Officer and Partnerships Lead at Lucid Computing. Milos took one of the most unconventional paths into AI safety — a childhood shaped by war and scarcity in Serbia, twenty years as a volunteer mountain rescuer, and a PhD in electrical engineering. He shares how he decided to step down as CEO to devote himself to AI safety, how Blue Dot's courses became his entrance door into the field, and what it means to build transparency and trust into the compute layer that powers frontier AI.  Check out Blue Dot's courses here — https://bluedot.org/courses A podcast by BlueDot Impact [https://bluedot.org/].

Quitting OpenAI When Everyone Else Wanted In — Ashyana-Jasmine Kachra

In this episode, Dewi speaks with Ashyana, an AI ethics and safety manager at Google DeepMind, who took one of the most unconventional paths into frontier AI safety — a politics degree, 700 job rejections, and a research project with 90 women across the world that revealed digital harms hiding in plain sight. Ashyana shares how a pre-LSE research project reframed her entire career, what it was like working inside OpenAI on red-teaming GPT-4 and DALL-E 3, and the moment she decided to walk away from one of the most sought-after jobs in tech. She explains why she held two jobs simultaneously at multiple points in her career, how teaching on Blue Dot's AI governance course reshaped her thinking, and why understanding AI risk requires moving fluently between the technical and the societal. This conversation digs into what AI safety actually looks like from inside a frontier lab, why a politics degree might be more useful than you think, and how staying relentless, even through hundreds of rejections, can lead somewhere nobody mapped out for you. Check out Blue Dot's courses here — https://bluedot.org/courses A podcast by BlueDot Impact [https://bluedot.org/].

Can Insurance Make AI Systems Safer? — Daniel Reti & Luca Ambroz

In this episode, Dewi speaks with Daniel and Luca, two quantitative analysts who left Bank of America to build Exona Labs — a startup helping insurance companies price AI risks. Daniel shares his path from applied maths in Hungary through finance to AI safety, while Luca talks about measuring the Higgs boson at CERN before pivoting to the corporate world. They explain the moment AI coding assistants made them realize this technology was genuinely transformative, the historical parallel of steam boiler insurance (and why AI moves faster), and how Blue Dot's incubator helped them build risk modelling tools that incentivize safer AI deployment through market mechanisms. This conversation digs into why insurance could be faster than regulation, how financial incentives can drive safety at scale, and what it takes to build a startup in an industry that still prices everything by counting past failures. Check out Daniel and Luca's new AI Insurance Startup here - https://exonalab.com/ Want to take one of our courses?  Check them out here — https://bluedot.org/courses A podcast by BlueDot Impact [https://bluedot.org/].

Why AI Safety Needs Founders – Ryan Kidd

In this episode, Dewi speaks with Ryan Kidd, an Australian quantum physicist who pivoted from academia to become co-executive director of MATS (ML Alignment & Theory Scholars), one of the field's most ambitious AI safety talent pipelines.  Ryan shares his journey from Brisbane to Berkeley—through depression, long COVID, and a PhD he almost didn't finish—and how discovering effective altruism during his doctorate redirected his career entirely. He talks about the turning point when he realized his impact would be greater building infrastructure than publishing papers, the qualities that make someone a great researcher, and why the field desperately needs more "amplifiers" (people who can bridge technical work with operations and strategy). Ryan also reflects on MATS' explosive growth from 15 scholars to 175 fellows, the organization's plans to 10x again, and his vision for a coordinated, careful AI rollout that preserves human agency and meaning.  This conversation digs into the realities of field-building at scale, the undervalued role of founders in AI safety, and what it takes to accelerate an entire generation of researchers toward building systems we can actually trust. A podcast by BlueDot Impact [https://bluedot.org/].

AI Safety Needs More Generalists – David Conrad

In this episode, Dewi speaks with David Conrad, an Australian entrepreneur who went from running a nationwide chain of premium pet resorts to helping build Europe’s AI policy talent pipeline. David shares how burnout, the pandemic, and a growing sense of responsibility pulled him away from the startup world and toward the fast-moving landscape of AI governance. He talks through the uncertainty of entering a new field, the challenge of finding meaningful work in AI policy without a technical background, and how early mentors, courses, and community helped him find his footing. David also reflects on why he ultimately chose field-building over direct policy work, how Talos aims to scale Europe’s AI governance workforce, and the impact their fellows are already having through placements in think tanks, government bodies, and EU institutions. This conversation digs into the realities of switching careers into AI safety, the need for entrepreneurial people in governance roles, and the growing importance of building teams that can steer advanced AI toward a better future. A podcast by BlueDot Impact [https://bluedot.org/].