Daniel Ek: Life After Spotify, Broken Healthcare Incentives, Catching Disease Early & AI's Potential
Daniel Ek, Spotify's co-founder and now executive chairman, discusses his transition to healthcare through his new company Neko, which offers comprehensive preventive health screening for $499. The conversation covers Spotify's founding challenges, healthcare system incentive problems, and AI's potential in medical diagnostics and beyond.
Summary
Daniel Ek reflects on founding Spotify in 2006 at age 23, when music piracy was rampant and the industry was in crisis. He and co-founder Martin negotiated with record labels by guaranteeing their bonuses and budgets for a year, reducing risk and eventually securing licensing deals. After 5 years of development in Sweden and the UK, Spotify launched in the US in 2011. Ek emphasizes how Spotify demonstrated that providing legal, convenient access to music could solve piracy better than litigation.
Ek then pivots to discuss his healthcare venture Neko, launched in 2018 and now expanding to the US market. He identifies the core healthcare problem: while spending 18% of GDP on healthcare (the largest budget item), outcomes are worsening. Chronic diseases are preventable if detected early with minimal cost, yet the system remains reactive rather than preventive. Neko's solution involves comprehensive data collection—blood tests measuring 53 markers, 6,000+ ultra-high-resolution skin images indexed with AI, cardiovascular assessments, and clinician consultation—all in one hour for $499.
The business model relies on vertical integration; Neko builds its own facilities, equipment, and software, achieving unit economics profitability. Over 100,000 scans reveal that approximately 1% of participants have serious undiagnosed conditions, and participants with poor health metrics show the most improvement after understanding their status. Specific examples include dermatology, where the average person has 950 moles—impossible for doctors to monitor manually but trackable via AI comparing images longitudinally across years.
Ek addresses systemic healthcare issues, arguing that broken incentives stem from the system's design around acute infectious disease treatment rather than prevention. Employment-based insurance creates perverse incentives: insurers won't invest in preventive measures with 10-20 year payback periods when employees average 2-3 years tenure. Neko's approach lowers upfront costs so ROI calculations become more favorable and data collection enables evidence-based policy.
On AI, Ek acknowledges the technology enables both positive and negative outcomes, emphasizing that human choices direct its application. He cites concrete AI benefits: personalized music playlists using predictive models and dermatological screening tracking 950 moles over time. He advocates for discussing positive AI applications more, arguing the industry has done a disservice by amplifying concerns. On open vs. closed models, Ek notes technology history shows both coexist—Windows vs. Linux, iOS vs. Android—and Spotify uses both proprietary and open-source models for flexibility and innovation.
Ek also discusses computation as a regulatory lens for AI, comparing it to historical restrictions on Cray supercomputers during the Cold War; computation constraints (GPUs, energy usage, teraflops) could serve as security boundaries better than software alone.
Finally, Ek recounts his pre-Spotify experience at Stardoll, a digital fashion site, where he reduced page load times from 4 minutes to under 1 second by redesigning server architecture, dramatically increasing traffic and attracting Sequoia investment before he founded Spotify.
Key Insights
- Ek resolved Spotify's founding challenge by guaranteeing record labels their bonuses and budgets for one year, eliminating their downside risk while allowing upside if the service succeeded, which eventually secured licensing deals after years of negotiation
- The US healthcare system is fundamentally designed around treating acute infectious diseases rather than preventing chronic diseases, creating perverse incentives where insurers won't fund preventive measures with 10-20 year payback periods when employees typically stay 2-3 years
- Neko's data from 100,000+ scans shows approximately 1% of participants have serious undiagnosed conditions, and participants with the worst health metrics show the most improvement after receiving visual data and clinician guidance on their health status
- The average person has 950 moles, which makes manual monitoring during doctor visits impossible, but AI systems tracking images longitudinally across years can detect abnormal growth patterns that individual doctors cannot remember from year to year
- Computation volume, not just model intelligence, should be considered as a security boundary for AI regulation, similar to how Cray supercomputers were historically restricted; limiting access to massive GPU clusters may be more effective than software-based controls
Topics
Transcript
[0:01] Okay, everyone, welcome back to the All In interview show, where we, the All In podcast, dedicate an hour to some of the great thinkers, the creators of our time, and today will be no exception. With us is Daniel Ek. You know him, the co-founder of Spotify, which he founded 20 years ago , and he led it for those two decades to the extraordinary state of affairs today. Over 700 million active [0:32] users, over 300 million premium subscribers, but on January 1st of this year, he changed and became the executive chairman with me and David Friedberg today. Daniel Eck, how are you, sir? Welcome to the program. Well, thank you very much for inviting me.…
Full transcript available for MurmurCast members
Sign Up to AccessMore from All-In Podcast
Jake Paul & The Chainsmokers: Turning Fame into Funds, Jake Enters Politics? & Venture Bubble Signs
A wide-ranging discussion featuring Jake Paul on entrepreneurship, boxing, and venture capital, followed by The Chainsmokers discussing their evolution from music producers to serious venture investors. Both segments explore how entertainment figures leverage fame and attention into sustainable business empires.
Jason Calacanis: AI Is About to “Blow People’s Minds”
Jason Calacanis argues that AI products like Grok and Muse are reaching a tipping point where average consumers can derive real, tangible benefits for everyday tasks. He predicts that by year-end, ordinary people will adopt these tools as free personal assistants, marking a significant moment in AI democratization that could benefit lower-income Americans.
Chamath on Obama’s AI Warning: Follow the Money
Chamath argues that AI development prioritizes commercial interests over societal needs, and that massive wealth concentration in left-leaning tech companies will be funneled through philanthropic structures to disproportionately influence Democratic political power rather than benefit the broader economy.
David Friedberg: Obama Doesn’t Understand AI and He’s Making Prosperity Political
David Friedberg criticizes Obama for dismissing agentic AI while claiming it's unnecessary for solving major problems like cancer, arguing this reflects poor understanding of AI technology and represents partisan politics that unnecessarily opposes transformative prosperity-enabling technology.
Jensen Huang: America Needs Open Source to Win AI... Because Open Source Lets Every American Win
Jensen Huang argues that open-source AI models are critical to American competitiveness and innovation, enabling diverse companies and entrepreneurs to build on shared technology regardless of origin. He contends that winning in AI requires broad participation across all industries and sectors, not just dominance by a few tech companies.