Time | Agenda Item | Lead | Notes |
5 min | | Chairs | Antitrust Policy Notice: Attendees are reminded to adhere to the meeting agenda and not participate in activities prohibited under antitrust and competition laws. ToIP Policy: Only members of ToIP who have signed the necessary agreements are permitted to participate in this activity beyond an observer role. ToIP TSWG IPR Policy: see TF wiki page. AI & Metaverse Technology Task Force
|
5 mins | Introduction of new members Any general announcements, news, that could be of interest to the TF Last week's action items
| All |
|
40 mins | A proposal to initiate a deliverable on Authentic AI : request for comments and interests | @Wenjing Chu | This will be the second iteration of draft proposal for soliciting feedbacks/interests. If there's sufficient interest to pursue, we will come back to have a deeper discussion on Dec 21 (which will be the last meeting of TF for this year). We discussed the document and made updates to the document. See document. We reviewed the draft @Wenjing Chu created and all the comments we had received on the google doc. The proposal is to produce a timely report on this hot topic but stay in the authentic AI / trustworthy AI area which intersects ToIP's tool kit (e.g. VIDs, TSP, ACDC, ... ). @Daniel Bachenheimer questioned why it's only "authenticity" - @Wenjing Chu explained that he is borrowing the concept from C2PA where authentic content meant not only authenticity of the author/creator but also integrity of the image (not being tampered with). One of the first comments in the google doc is also defining terms clearly as part of the work. One of @Steven Milstein 's suggestions is to have a 'case study' to focus application use cases. @Wenjing Chu proposed e.g. on RAG style AI applications. We had a discussion and agreed we should also generalize RAG into dataset management to avoid closely tying to a solution that may become dated or narrow quickly. We had a discussion around using C2PA for content authenticity (through provenance metadata) and use VID + TSP for communication of such content therefore combing two techniques for a complete solution. We had a discussion on parties in AI ecosystem e.g. consent, agreement, license, leveraging VID + TSP + ACDC etc. and apply such a solution to the case study. We also discussed, even though we should start with a case study, if we get more attention and contribution, we could apply the same solution scheme to other use cases. We will recap and resume the discussions at the Jan 18 meeting in the new year.
|
10 mins | Happy holidays! | ALL | |
0 mins | | Chairs | We will cancel Jan 4 and restart on Jan 18, 2024. Keep the same time slot (and discuss on Jan 18 if there're needs to adjust)
|
|
|
|
|