In this episode, Michael Fischer, the founder of DB DAO, discusses the role of databases in Web3 and how they can be used improve scientific research and tokenize scientific articles.
Fischer also shares his own experience with medical conditions and his thoughts on the healthcare industry. Fischer has a PhD in computer science from Stanford University, where he studied natural language processing and AI.. He is co-author of the book "Regulating AI" and leads the DeSci NYC community.
Announcements
- Health Unchained is a media partner for the Blockchain in Healthcare Today Conference in New Orleans in September 2023 Blockchain in healthcare Today in New Orleans, LA September 2023 - https://conv2xsymposium.com/
- Attending Desci London event Jan 15-16 - https://www.desci.london/ - in-person hackathon on Jan 13 and 14 https://desci-london.devfolio.co/
Links and Resources
Topics Covered Include:
00:00:00 - Exploring Blockchain Technology: A Stanford Graduate's Journey
00:00:38 - Introduction to Health Unchained Episode 109 with Michael Fischer
00:06:14 - Alternative Peer Review System Using Tokens
00:08:46 - Building A Community Around DeSci
00:13:43 - Building An Open Database Team For Data Preservation
00:19:40 - Exploring Decentralized Database Governance And Data Ecosystems
00:20:26 - Creating A Mission For A Database: Aligning People Towards A Common Goal
00:21:48 - Creating A Governance Structure For Data Collection And Rejection
00:26:06 - Exploring Corporate Governance Structures In Multi-sig Protocols
00:32:51 - Interoperability Of Web Three With Web Two Applications For User Benefits
00:35:39 - Scouts Earn Rewards For Contributing Data To Database
00:38:53 - The Benefits And Risks Of AI-powered Content Curation
00:42:07 - Data Generation And Web Three Tooling
00:48:01 - Using Zero Knowledge Technology To Create A Web Three Database For Data Science Queries
00:49:35 - Reducing Reliance With ZK And DLT Protocols
00:50:38 - Health Unchained News Corner
00:51:46 - GPT-3: AI-enabled Chat Applications Reaching New Heights
00:54:54 - Effects Of Injury And Medical Conditions On Health Care Industry
00:58:49 - Structuring Data To Improve Patient Communities
News Corner
Google Research and DeepMind have created MedPaLM, an open-source large language model that can answer medical questions. It combines HealthSearchQA, a new dataset of medical questions, with six existing datasets covering professional medical exams, research, and consumer queries. The model was developed on PaLM (Pathways Learning Model), a 540 billion parameter large language model, and is intended for use by both medical professionals and non-professionals. The developers of MedPaLM are considering using blockchain technology to secure the data input and generated by the model. They hope that the tool will be successfully used in practice with real patients.
Health Unchained Links
