RE: World AI Governance Body

Prachant Bradwell,

Also for purposes of discussion:


4. What do you think about a consortium model, e.g. an international Artificial Intelligence Data Consortium, perhaps improving upon the Linguistic Data Consortium model (https://www.ldc.upenn.edu/about)?


Best regards,
Adam Sobieski


________________________________
From: Adam Sobieski <adamsobieski@hotmail.com>
Sent: Friday, July 5, 2019 6:36:24 PM
To: Bradwell (US), Prachant; semantic-web@w3.org
Subject: RE: World AI Governance Body


Prachant Bradwell,



Interesting ideas. Some points for discussion:



  1.  Why would or should scientists choose to defer to the voting results of an intranational or international supermajority of participants?
     *   Historical opinions to consider include those of James Madison.
  2.  Instead of voting-based systems, what about argumentation-based systems and other group reasoning and group decision-making systems (see: [1], [2])?
  3.  What do you think about democratizing and/or crowdsourcing content for portions of artificial intelligence textbooks and courses, e.g. portions discussing comparative ethical standards pertaining to artificial intelligence (see: [3])?





Best regards,

Adam Sobieski



[1] Klein, Mark. "Achieving collective intelligence via large-scale on-line argumentation." In Second International Conference on Internet and Web Applications and Services (ICIW'07), pp. 58-58. IEEE, 2007.

[2] Carrascosa, Iván Palomares. Large Group Decision Making: Creating Decision Support Approaches at Scale. Springer, 2018.

[3] Russell, Stuart J., and Peter Norvig. Artificial intelligence: a modern approach. Malaysia; Pearson Education Limited,, 2016.



________________________________
From: Bradwell (US), Prachant <prachant.bradwell@boeing.com>
Sent: Friday, July 5, 2019 4:30:30 PM
To: semantic-web@w3.org
Subject: World AI Governance Body

Hi all,

What if there was a world governance body (e.g. UN, World Economic Forum, etc.)  which enables voting for “high value data opportunities” that AI/linked data can be used to solve significant world problems.

The voting system would work similar to Reddit, in which designated voters can “upvote” or “downvote” a high value opportunity. This would be the mechanism for prioritization. The highest scores would receive the top priority.

Those opportunities which receive 0 or fewer would not be actionable until further review.

This could create work for nonprofits, private, and public entities through competition and/or collaboration, enabling quick development of solutions.

This could enable us to prioritize and attack key issues such as climate change with advanced technologies on a world stage.

Last, policy for this voting body would require a supermajority vote; which would in my opinion help enable truly global ethical decisions.

Thoughts?

Sent from my iPhone