The person is Jeremy Hsu.
Artifical intelligence could make better funding decisions than senators.
Walter Bibikow is a photo editor.
An artificial intelligence system has learned how to make the most popular policy for redistributing public money in online games.
"Many of the problems that humans face are not simply technological, but require us to coordinate in society and in our economies for the greater good." Artificial intelligence needs to learn about human values.
The DeepMind team trained its artificial intelligence to learn from thousands of people and computer simulations. In the game, players start with a certain amount of money and must decide how much to contribute to help grow a pool of public funds, eventually receiving a share of the pot in return. The players voted on their favorite policies.
The policy developed by the artificial intelligence tried to reduce wealth disparity between players by redistributing public money according to how much they contributed. It discouraged free-riders from giving back anything unless they gave at least half of their starting funds.
This policy won more votes from human players than either anegalitarian approach of redistributing funds equally regardless of how much each person contributed, or alibertarian approach of handing out funds according to the proportion each person's contribution makes up of
Christopher Summerfield at DeepMind says that it was surprising that the artificial intelligence learned a policy that reflected a mixture of views.
The liberal egalitarian policy, which redistributed money according to the proportion of starting funds each player contributed, was popular because it didn't discourage free-riders.
DeepMind researchers warn that their work doesn't represent a recipe for government. They don't plan to make policy-making powered by artificial intelligence.
Annette Zimmermann at the University of York, UK says that the proposal isn't necessarily unique compared with what others have suggested. It's not a good idea to focus on a narrow idea of democracy as a way to find the most popular policies.
"Democracy isn't just about winning, it's about creating processes during which citizens can encounter each other and deliberate with each other as equals"
The needs of people in minority groups are not taken into account by the majority. According to Risse, that isn't a big concern for political scientists. Modern democracies face a bigger problem of the many being left out of the political process by the small minority of the economic elite.
Risse says the DeepMind research is fascinating in how it delivered a version of liberalism. He finds that a rather satisfactory result since he is in the liberal-egalitarian camp.
Nature Human Behavior was published in the journal.
There are more on this topic.
All Rights Reserved © 2024
