AI Impacts blog

Share this post

Event: Multipolar AI workshop with Robin Hanson

blog.aiimpacts.org

Discover more from AI Impacts blog

A blog from AI Impacts about the future of artificial intelligence
Continue reading
Sign in

Event: Multipolar AI workshop with Robin Hanson

Katja Grace
Jan 14, 2015
Share this post

Event: Multipolar AI workshop with Robin Hanson

blog.aiimpacts.org
Share

By Katja Grace, 14 January 2015

On Monday 26 January we will be holding a discussion on promising research projects relating to 'multipolar' AI scenarios. That is, future scenarios where society persists in containing a large number of similarly influential agents, rather than a single winner who takes all. The event will be run in collaboration with Robin Hanson, a leading researcher on the social consequences of whole brain emulation.

The goal of the meeting will be to identify promising concrete research projects.

We will consider projects under various headings, for example:

Causal origins and probability of multipolar scenarios

  • Collect past records of lumpiness of AI success

  • Survey military, business or academic projects which were particularly analogous to successful emulation or AI projects, to learn about the situations in which emulations or AI might appear.

  • Survey AI experts on the likelihood of AI emerging in the military, business or academia, and on the likely size of a successful AI project.

  • ...

Consequences of multipolar scenarios

  • Hanson’s project of detailing a default whole brain emulation scenario

  • How does the lumpiness of economic outcomes vary as a function of the lumpiness of origins?

  • Are there initial social institutions which might substantially influence longer term outcomes?

  • ...

Applicability of broader AI safety insights to multipolar outcomes

  • How useful are capability control methods, such as boxing, stunting, incentives, or tripwires in a multi-polar scenario?

  • How useful are motivation selection methods, such as direct specification, domesticity, indirect normatively, augmentation in a multipolar scenario?

  • Would selective pressures strongly favor the existence of goal-directed agents, in a multipolar scenario where a variety of AI designs are feasible?

  • ...

Details in brief

Time:

2pm until 2-3 hours later

There will be an evening social event at 7pm in the same location, which workshop attendees are welcome to stay or return for. Some participants will go to dinner at a nearby restaurant in between.

Date: Monday 26 January 2015

Location: Private Berkeley residence. Detail available upon RSVP.

RSVP: to katja.s.grace@gmail.com.

This is helpful, but not required (if you can deduce the location).

Other things: Tea, coffee and snacks provided.

Share this post

Event: Multipolar AI workshop with Robin Hanson

blog.aiimpacts.org
Share
Comments
Top
New
Community

No posts

Ready for more?

© 2023 AI Impacts
Privacy ∙ Terms ∙ Collection notice
Start WritingGet the app
Substack is the home for great writing