[Ongoing] Entries for next post

[jphall663]

~People like software and it's kind what this list was known for, so:~

~ xplique~ ~ ExplainaBoard~

~People are insane about GAI, so:~ ~* Google Privacy & Terms, Generative AI Prohibited Use Policy~

~Other interests:~ ~ Nvidia MLPerf~ ~ NewsGuard AI Tracking Center~

~thoughts @datherton09 ?~

[datherton09]

• Could do more concerted searches for software re: responsible AI or put out a call for people who might be interested in sharing their work/GitHub repositories with us in our next post.
• Could do a concerted search for other similar use/prohibited use policies for GAI among other companies and organizations (actually seems like a pretty good next task, I'm thinking).
• Possibly in the neighborhood of the AI incident trackers we could have GAI misinformation/disinformation tracking communities, studies, etc.?

[jphall663]

~- prohibited and acceptable use policies~ ~- JohnSnowLabs / langtest~

[jphall663]

~The Foundation Model Transparency Index~

[jphall663]

~Beyond Memorization: Violating Privacy via Inference with Large Language Models~

[jphall663]

Posted 11/14/23

[jphall663]

For next post:

• CSET incident taxonomy

[jphall663]

• Trust LLM
• AI regulation map (good picture)
• Scikit-explain
• Awesome ML Guidelines
• Vulnerabilities tutorial
• https://github.com/i-gallegos/Fair-LLM-Benchmark
• model-cards-and-datasheets
• https://github.com/IMDA-BTG/LLM-Evals-Catalogue
• New sections: infographics/cheatsheets & LLM red-teaming
• Trustible AI definitions

[jphall663]

https://safetyprompts.com/ https://github.com/lilacai/lilac LLM red-teaming resources

[datherton09]

https://docs.google.com/spreadsheets/u/1/d/e/2PACX-1vQObeTxvXtOs--zd98qG2xBHHuTTJOyNISBJPthZFr3at2LCrs3rcv73d4of1A78JV2eLuxECFXJY43/pubhtml

[jphall663]

https://osf.io/fvqg3/

[jphall663]

Start here for Apr Post

• weightwatcher
• insanely complicated, hopelessly inadequate
• "reasoning gap" paper
• "illusions" paper
• dialect bias paper

[jphall663]

CDT - socio-technical approaches EU AI act info graphic Angwin NYT piece