• Login
  • Register

Work for a Member company and need a Member Portal account? Register here with your company email address.

Event

MIT Decentralized AI Summit

Chris Pease  

Friday
October 11, 2024
9:00am — 3:00pm ET

The Internet of AIs with Edge Compute, Siloed Data, Agents and Data Markets

In today's AI landscape, data, computation, and governance are centralized. Decentralized AI offers a promising alternative leveraging intelligent, network-based approaches to improve collaboration, incentives, and innovation at the edges. It offers a promising alternative enabling collaboration between distrusted, disincentivized, and disinterested entities.

AI has made remarkable strides, showcasing its vast ability to tackle large-scale challenges. These advancements open new possibilities for areas where data and knowledge are distributed across different organizations and regions such as healthcare, climate science, and supply chain management. We highlight five essential technical opportunities: (1) enhancing privacy, (2) ensuring verifiability, (3) aligning incentives, (4) enabling orchestration, and (5) improving crowdUX. By addressing these opportunities, diverse entities can collaborate to solve global issues while still achieving their local goals. This vision of Decentralized AI encourages further dialogue and research toward a decentralized, inclusive, and resilient AI future.

MIT Decentralized AI Research and Venture Hub 

Call for research projects, nonprofits and ventures to participate in the Decentralized AI  Summit Showcase. Present on the main stage and showcase your projects during the networking session. We also have limited opportunities for you to participate in panel discussions. Please submit completed forms by October 9th, 2024.

Program: 

Project Demo and Showcase:

Previous  Event

Decentralized AI Roundtable  on July 29, 2024 featured experts discussing global orchestration, user-owned models, multi-agent systems, on-device applications, and peer-to-peer protocols. The speakers explored technical challenges and innovative solutions, highlighting decentralized AI's potential for enhancing privacy, scalability, and efficiency, while interactive Q&A sessions delved into practical and theoretical implications, setting the stage for future research and implementation.

Organizing Committee

Engage

Reach out to us at dec-ai@media.mit.edu

Notice to MIT Faculty, Students, Staff or Affiliates 

To the extent you are presenting research findings that have not already been made public, we encourage you to work with your advisor/principal investigator and also relevant technology licensing office (for MIT researchers and students, that is MIT’s TLO to ensure you have proper supervisor approvals to discuss your research publicly, and also that the material presented will not compromise future patent filings

More Events