All-In PodcastDaniel Ek: Life After Spotify, Broken Healthcare Incentives, Catching Disease Early & AI's Potential
All-In Podcast
Introducing Daniel Ek 0:00
Daniel Ek joins the show as the co-founder of Spotify, a company he ran for two decades before becoming executive chairman on January 1st of this year. Spotify now has over 700 million active users and more than 300 million premium subscribers.
Spotify's Difficult Origin Story 2:00
Ek started Spotify in 2006 at age 23, before the iPhone, before YouTube, and while Facebook was still college-only. At the time, music piracy through services like Napster and Kazaa was rampant, and the RIAA was suing individual consumers for illegal downloads. Ek and his co-founder Martin kept brainstorming until they landed on the idea of making all the world's music available at your fingertips. Sweden turned out to be an ideal testing ground because its fast broadband had already devastated the local music market, cutting industry revenue by roughly 80 percent, and there was no legal alternative like iTunes available. Ek and his partner used their own money from prior startups to guarantee record labels their bonuses and budgets for a year if the idea failed, which convinced labels to sign on after years of negotiation. Spotify launched in late 2008, became a hit in Sweden and then the UK, and reached the United States in 2011, five years after the initial idea.
From Spotify to Healthcare 9:02
Ek says he had been thinking about healthcare since at least 2012 or 2013, based on an old interview his team recently showed him. He didn't have meaningful money to invest until Spotify went public in 2018, and once he tried angel investing he realized he preferred building companies to sitting on boards watching others struggle. That preference led him to found Neko, a healthcare startup that began in 2018, spent five years developing its product in Sweden, then expanded to the UK, and is now launching in the United States, following what he calls the same playbook as Spotify.
Why Healthcare Is Broken 11:01
Ek points out that the United States spends 18 percent of its GDP on healthcare, making it the largest line item in the national budget, with heart disease alone costing hundreds of billions of dollars, more than the revenue of many Fortune 10 companies. After researching drugs, cancers, and other causes, he concluded that chronic disease is the biggest driver of these costs, and that conditions like cardiology issues and skin cancer are especially over-represented. The key insight is that chronic diseases are largely preventable and cheap to treat if caught early, but the healthcare system lacks the data needed to catch them early, even though everyone agrees it should shift from reactive to preventative care.
Neko's Data-Driven Approach 13:32
Ek and his co-founder Yalmer believe the solution is gathering much more health data, across more types of measurement and over long stretches of time, so predictive analysis becomes possible. He credits the rise of smartphone sensors and modern machine learning, now commonly called AI, with making this kind of large-scale data collection and prediction feasible in a way it wasn't before. He compares it directly to Spotify, where more data led to better song recommendations, arguing the same principle now applies to predicting health outcomes. Neko itself costs 499 dollars and is a vertically integrated healthcare experience, with its own facilities, nurses, doctors, diagnostic equipment, and software, all built to create what Ek calls the most valuable hour you can invest in your health.
A Full Hour Health Visit 16:02
The Neko visit starts with drawing blood to measure 53 blood markers, then moves through a skin imaging rig that captures over 6,000 high resolution images across your body, indexing every mole, lesion, and rash. The visit also checks your heart, blood circulation, and grip strength, all markers with strong scientific backing. The hour ends with uninterrupted time with a clinician who walks you through every result, including blood work, and discusses what you might want to improve about your health. This experience launched in New York, with more US locations planned in coming months.
Eight Years Of Outcomes Data 17:32
Daniel Ek notes Neko has operated since 2018, completed more than 100,000 scans, and published data every year, including a third year survey. Around 1 percent of members are found to have a serious undiagnosed medical condition. The most encouraging finding across three years of study is that members with the worst health status improve the most, since seeing a clear picture of their health alongside a clinician's explanation motivates lifestyle change, including cases of people quitting smoking. He points to dermatology as an example: the average Neko member has 950 moles, far too many for a normal doctor's visit to check. Neko uses an AI system to flag risk factors, which a clinician reviews, and expert dermatologists review further if something is found, with the added benefit of tracking moles longitudinally year over year, something no doctor could remember unaided.
Cost, Economics, And Broken Incentives 22:31
The visit costs $499, comparable to Function Health and Superpower, which only do blood work, and Ek argues an annual visit should become as routine as a dentist appointment, recommending it be done yearly. Because Neko is vertically integrated and built everything itself, unit economics are positive and its clinics already operate profitably. Ek remains involved with Spotify as executive chair and runs Prima Materia, a company built with Shaquille O'Neal, alongside Neko. Asked about the broader US healthcare system, he says incentives are built for a past era focused on infectious disease and acute symptom treatment rather than prevention, and that insurance being tied to short-term employment discourages investment in long-payback preventative care. His approach is to lower costs so the return on investment doesn't require millions of speculative dollars, and to generate more longitudinal, multi-country data, comparing findings across populations such as Stockholm and the UK.
Neko Scans and Early Health Data 32:00
Daniel Ek explains that Neko, the health scanning venture, has gathered around 100,000 scans so far, still too small a base to draw firm population level conclusions, though he expects that to change as the dataset grows. The company has already run four clinical trials with two more underway and four planned, partnering with the research community around these data sets. He points out something surprising to people in tech: the healthcare system actually holds far less data than most industries, and he believes that growing this data by ten or a hundred times will let researchers draw genuinely useful conclusions, not just for Neko but for healthcare broadly.
Wearables and Expanding Diagnostics 33:30
Freedberg raises the idea of combining wearable data from devices like Oura, Whoop, Fitbit, and Apple Watch with bloodwork, noting that nobody has really merged the two yet. Ek confirms Neko already imports Apple Health data, and that many wearable makers allow syncing to Apple Health, which lets clinicians build a fuller 360 degree picture of a person's health. He stresses that Neko launched with one diagnostic offering but is already on its second generation of infrastructure, and the plan is to keep adding more diagnostics over time at the same price point, working toward the mission of delivering the most valuable hour someone can spend on their health.
AI, Openness, and Compute as a Safeguard 36:00
Ek reflects on how public opinion swings through overconfident phases, comparing it to how everyone briefly became an armchair virologist during COVID or a military expert during the Ukraine war, and says he has no strong stance on whether AI development should be slowed. He argues every powerful technology carries both extreme upside and extreme downside, and that the industry undersells its positive uses, pointing to Spotify's potential to soundtrack a person's life and Neko's mole tracking as genuine benefits of AI. On open versus closed models, he notes technology has always oscillated between the two, citing Windows versus Linux and iOS versus Android, and Spotify itself blends frontier models with its own fine-tuned ones. He also floats an underdiscussed idea: the sheer scale of compute used, such as 100,000 GPUs versus a home PC, could itself serve as a practical safeguard against misuse, a point the conversation partners agree deserves more attention than it currently gets.
The Startle Origin Story 48:01
Before Spotify, Daniel Ek briefly helped rebuild a struggling Finnish site called Paper Doll Heaven, backed by Index Ventures, which became Startle. Its pages took four minutes to load, so he brought in a new technical team, cut load time to under a second, traffic exploded, and Sequoia invested before he moved on to build Spotify.
Neko Health Locations 50:01
Neko Health's first location is at 300 Lafayette Street in New York, with a waitlist at nikohealth.com. The company plans to expand within New York and launch soon in Miami and Washington DC, aiming to reach more of the US over the next twelve to twenty four months.
AI-generated summary. It can be wrong or incomplete - check anything that matters against the original.
