On This Day 6 years ago: the first general meeting of the Partnership on AI took place in Berlin. PAI funds studies and publishes guidelines on questions of AI ethics and safety.
It just published a set of guidelines for the safe deployment of foundation models:
ETHICS
-
Partnership on AI publishes foundation models safety guidelines
By
–
-
Lab-grown human brain cells transplanted into newborn rats
By
–
Scientists have transplanted lab-grown clumps of human brain cells into the brains of newborn rats, where they integrated with the rodents' own neural circuits. Sounds gross, but this can help us learn more about human neuropsychiatric disorders.
-
$10M AI Safety Fund and Frontier Model Forum Leadership
By
–
Announcing a $10M AI Safety fund, and the selection of the Executive Director for the Frontier Model Forum:
-
FTX Fraud: Industrialized Crime Network, Not Simple Mastermind
By
–
For institutional reasons the FTX fracas has been described as variously a mastermind or a small “inner circle.” The truth is a bit worse than that: it was an effectively industrialized operation, which purchased competence from the market, then deployed it to illicit ends.
-
Trustworthy AI: Building Transparent and Validated Models
By
–
SAS’ Reggie Townsend, VP Data Ethics Practice tells how businesses can capitalize on Trustworthy #AI in the right way. “You must appreciate how your models are built and optimized and have the ability to validate that over time.” Get the facts: http://
2.sas.com/6018u5ux0 @CIOonline -
AI Accessibility Ethics and Coordination Risks in Large Systems
By
–
(Because the shape of this is understandable by non-specialist, because they would be morally offended because could see themselves actually using the thing, because the larger graphs are administered by people who are much more capable of coordinating codes of silence, etc.)
-
Scientists Collected 570K ChatGPT Conversations for API Access
By
–
these mad scientists really just allowed strangers to use their chatGPT api key for free in return they collected 570k real conversations with chatGPT I’ve heard this data is super valuable, but was this really worth it?
-
Market Pricing and Access Substitutes in AI Systems
By
–
4) There exist close substitutes to the thing provided by the market at market prices, in effectively unlimited quantity, which do not need to prioritize by either PMC brainsweat or timestamps. They’re prioritized in the traditional capitalist fashion: ability to pay.
-
Abundance, Not Allocation: True Solution to Scarcity
By
–
3) There is no cure for scarcity but abundance. Everything else merely allocates scarcity. Any thought about the fairness of the allocation strategy which is not “Or we could simply choose to make more of the thing” will not achieve fairness.
-
Grant Selection Criteria and Racial Demographics in American Institutions
By
–
I’m left pondering some conflicting thoughts, which I will muse about out loud: 1) Ordering one’s beneficiaries by ability to present as American PMC while explicitly asking them about their racial background so that you can qualify for grants: as American as apple pie, in 2023.