Trust :: Data FinTech Cybersecurity Summit
MIT Media Lab E14-633
Cambridge, MA 02139
Thursday, January 11, 2018
8:00 AM - 2:30 PM
Senior Executives from our sponsor companies are invited to attend our first Cybersecurity Summit to be held at the MIT Media Lab on Thursday, January 11th, 2018. The goal of this workshop is to present new research into mechanism for sharing data, where some of that data is money, and identity.
Expected outcomes of this summit, are learnings about the state-of-the-art in cybersecuty for fintech, and an action plan to experiment with some of these new frameworks for data sharing amongst organizations.
This event is best suited for senior executives in a CIO, CTO or product owner role with broad responsibility for formulating and executing IT strategy.
8:00 - 8:30 Trust :: Data Consortium Annual Report: Sandy Pentland
Increasingly digital identity is playing a crucial role in the digital economy, in providing access to various online services today and future services based on ledgers/blockchain technologies. This project explores the connection between data providers and identity providers, methods for compliance to various regulations (e.g. GDPR), new economic models for future data markets, and the role of identity-enabled smart-contracts, user-centric consent management and social networks.
The Tradecoin project seeks to address the current challenges with regards to the stability of certain types of digital currencies, the use of assets-backed registries, new forms of collective-based reserve currencies and the issues of privacy. Among others, new mechanisms should be developed that allows entities who share a common reserve currency to collectivity define and enforce monetary policies based on the shared digital currency. Of interest also is the relationship between social connections and the diversity of activities as measured by the usage of some forms of digital currencies.
The OPAL-ML project seeks to explore the use of machine learning techniques in a distributed manner to improve performance and accuracy of OPAL-based data sharing. In an architecture with distributed instances of OPAL data providers (data servers), one approach could be to train the algorithm separately at each data server instance. Each data server instance could hold slightly different training datasets. The model trained at each data server instance would then be serialized and made available to the remote Querier. The OPAL principles remain enforced, where the Querier does not see the raw data but obtains the benefit of distributed data stores performing independent training.
In order for a group of data providers (e.g. data owners) to exchange information using the OPAL paradigm, the group copied form into a Trust Network for data sharing (informally referred to as a "Circle of Trust"). Here, in addition to collectively authoring new algorithms for execution in their respective OPAL backends, the group must develop a new set legal rules and system-specific rules that clearly articulate the required combination of technical standards and systems, business processes and procedures, and legal rules that, taken together, establish a trustworthy system for information sharing based on the OPAL model.
12:00 - 1:30 Working Lunch (Breakout Sessions)
- Breakout Session A: E14-633
- Breakout Session B: E14-244
- Breakout Session C: E15-341
- Breakout Session D: E15-385
1:30 - 2:30 Practical Next Steps (Reconvene in E14-633)
For an invitation to this conference, please email email@example.com