Former Governor Michael Dukakis wrote in his letter calling for contributions to the AI World Society (AIWS) Summit, “The real world applications of AI will bring revolutionary changes and will have profound effects on the future of humanity. The changes will bring challenges to societal norms and economic models that we have relied on for decades. And we would be wise to prepare for all that will mean…” But, “our national governments have been slow to act. And international bodies such as the United Nations have yet to effectively address the problem.”
The AIWS Summit is filling in this void, serving as a place where the brightest minds on the planet can work together, to find the innovative solutions that will help us build a brighter future. This week, we are pleased to present a talk by MIT Professor Alex ‘Sandy’ Pentland for the AIWS Summit.
Professor Pentland directs the Connection Science and Human Dynamics labs at MIT. He is one of the most-cited scientists in the world, and Forbes recently declared him one of the “7 most powerful data scientists in the world” along with Google founders and the Chief Technical Officer of the United States. He co-led the World Economic Forum discussion in Davos that led to the EU privacy regulation GDPR, and was central in forging the transparency and accountability mechanisms in the UN’s Sustainable Development Goals. He has received numerous awards and prizes such as the McKinsey Award from Harvard Business Review, the 40th Anniversary of the Internet from DARPA, and the Brandeis Award for work in privacy.
In his talk for AIWS Summit, Professor Pentland introduced a key project of his research group about techniques and open-source software for helping countries and companies deal with AI in a way that is effective, efficient, but also ethical. In today’s world where data is everywhere, most in the hands of private companies, he raised the question as to how this data can be used by governments, social and civic systems such that it is trustworthy and unbiased and that people understand what is happening.
He talked about his method of Open Algorithms as a way to address this question. He advocates the idea of leaving data where it is collected and have open algorithms answer inquiries about the data, instead of transferring all the data into one single pool. The latter is vulnerable to concerns about security, ownership, and privacy.
In his proposed framework, there should be a decentralized federation of different players and interests that agree to answer certain questions for certain functions such that everyone can audit. We can keep track of what questions are being asked about what data, and the people who collected or own the data can monitor the entire process. Decisions made by a country can be audited, or questions about fairness or bias can be answered, because we now have a record of what was done with the data and who did it.
Several countries, including Estonia, Israel, and Australia, have adopted this framework and conducted pilot projects to explore how to get better insights about the country from the public-private data partnership and come up with better policies to serve their people.
The full video of Professor Penland’s talk can be seen here.