Barriers to Building Medical Foundation Models

Join Pondering Durian and José Macedo as they dive deep into the future of AI-powered healthcare with Tanishq Abraham, founder and CEO of Sophont AI. At just 21, Tanishq has already graduated high school at 10, college at 14, earned a PhD in biomedical engineering at 19, and served as research director at Stability AI. Now he's building open-source foundation models to revolutionize healthcare through multimodal AI systems that can integrate diverse patient data for better diagnosis and treatment.

Sophont AI: https://sophontai.com

Tanishq Abraham: https://www.tanishq.ai/blog

🎯 Key Highlights

▸ Tanishq's accelerated academic journey and unique daily routine

▸ Why healthcare needs multimodal foundation models vs. specialized AI tools

▸ The "parable of the elephant" - integrating all patient data holistically

▸ Open source vs. proprietary models in medicine: trust and transparency

▸ Building a remote, Discord-based AI research company at 21

▸ Healthcare in 2035: continuous monitoring and proactive care

▸ Compute constraints in medical AI vs. general AI development

▸ US-China AI race: concerns about America falling behind in open source

▸ From astronomy to longevity: other fields Tanishq would explore

▸ Why medical AI can have faster patient impact than drug development

💡 Subscribe for more crypto & AI insights! 🔔

🧠 Follow the Alpha

▸ Tanishq's Twitter: @iScienceLuvr

▸ Sophont's Twitter: @SophontAI

🔗 Connect with Delphi

🌐 Portal: https://delphidigital.io/

🧠 Intelligence: https://www.delphiintelligence.io

🐦 Twitter: https://x.com/delphi_intel

🎧 Listen on

Spotify:

Apple Podcasts:

Youtube: https://www.youtube.com/channel/UC9Yy99ZlQIX9-PdG_xHj43Q

Timestamps

00:00 – Intro: Tanishq Abraham, Sophont AI

01:15 – The Tanishq Abraham production function

03:00 – Managing a startup team via Discord

05:00 – Accelerated education: college at 6 years old

09:15 – Finding your tribe and making friends across ages

13:00 – From astronomy to biomedical engineering

17:30 – Healthcare manifesto: why we need multimodal AI

21:00 – The elephant parable: integrating patient data

23:15 – Open source advantages in medical AI

25:00 – Academic vs. industry challenges in medical AI

28:15 – Healthcare experience in 2035-2040

32:15 – Compute vs. data constraints in medical AI

35:15 – Competition landscape and positioning

38:30 – Longevity predictions: 150-200 years?

41:30 – US-China AI race concerns

49:30 – Other research areas of interest

54:15 – Why medical AI offers faster patient impact

Disclaimer

This podcast is strictly informational and educational and is not investment advice or a solicitation to buy or sell any tokens or securities or to make any financial decisions. Do not trade or invest in any project, tokens, or securities based upon this podcast episode. The host and members at Delphi Ventures may personally own tokens or art that are mentioned on the podcast. Our current show features paid sponsorships which may be featured at the start, middle, and/or the end of the episode. These sponsorships are for informational purposes only and are not a solicitation to use any product, service or token.