This is our first attempt in over two years to gather several shorter presentations into one MeetUp. We have three speakers:

Louie Helm: Self-Improving AI & AI Risks (30 min + 15 min discussion)

We note that: (1) intelligence (and artifical intelligence by extension) can be understood as "efficient cross-domain optimization"; (2) powerful cross-domain optimizers for X will often create still more powerful optimizers for X without human intervention, in a cycle of self-improvement; and (3) very powerful optimizing systems would cause human extinction unless their goals are specifically designed to be optimally fulfilled by a world that includes humans. We conclude that although there are large gaps in our models, and although self-improving AIs are probably decades or more in the future, such scenarios deserve serious near-term research.

Louie Helm is a Visiting Fellow at the Singularity Institute for Artificial Intelligence and a published researcher in mathematics and quantum computing. He is founder of the Seventeen or Bust distributed computing project and an internet entrepreneur who previously held positions in the supercomputing and defense industries. He holds an M.S. in Software Engineering from the University of Texas.
========================
Randall Reetz: Optimizing a fitness filter for complexity handling. (20 min + 10 min discussion)

(No abstract available)
========================
Kevin Keck: Grids, Clouds, and Collective Intelligence. (20 min + 15 min discussion)

(No abstract available)
========================

Beyond these presentations I will show a short video at noon sharp. We'll use any remaining time for the usual discussion about any and all AI related topics.