Twitter | Search | |
Kate Crawford
Now the dust has settled on Google's AI principles, it's time to ask about governance. How are they implemented? Who decides? There's no mention of process, or people, or how they'll evaluate if a tool is 'beneficial'. Are they... autonomous ethics?
We’re announcing seven principles to guide our work in AI.
Google Google @Google
Reply Retweet Like More
Rob McCargow Jun 8
Replying to @katecrawford
And any external/independent scrutiny?
Reply Retweet Like
Kate Crawford Jun 8
Replying to @RobMcCargow
Exactly. An independent board would be a start.
Reply Retweet Like
Tom Saunders Jun 8
Replying to @katecrawford
It would be good if Google could use some of its tech to make this policy the basis of a big, inclusive conversation - are these the right principles? What have we missed? How should we implement them? What problems do you think we should be working on? etc etc
Reply Retweet Like
Stephen Lynch Jun 8
Replying to @katecrawford
What's the most common engagement model for these things? The committee structure? Some folks have discussed the idea of a kind of ombudsman, at least as it pertains to customer-facing ethical codes.
Reply Retweet Like
Kate Crawford Jun 8
Replying to @katecrawford
Principles minus process, or verification, or internal appeal structure, or independent review = no real accountability
Reply Retweet Like
Brian Smith Jun 8
Replying to @katecrawford
It's astounding how many orgs "value transparency" except for themselves.
Reply Retweet Like
Chad Vavra Jun 8
At what point do we, as humans, say that AI falls within the definition of an organism that effects our biosphere? That is when I think they should have to submit to peer review, like every other mature science.
Reply Retweet Like
Chad Vavra Jun 8
FTR: I think we are long past that definition.
Reply Retweet Like
Atul Acharya Jun 8
Replying to @katecrawford
My concerns, exactly. Who governs, how, under what processes, verifications and reviews. It's good that there's public scrutiny for . We need more of it.
Reply Retweet Like
Matt Jackson Jun 8
Replying to @katecrawford
The language is also pretty soft throughout, particularly here: "We will give opportunity for notice and consent, encourage architectures with privacy safeguards, and provide appropriate transparency and control over the use of data."
Reply Retweet Like
Andrew Maier Jun 8
The constitution is just a document as well. It’s taken people who believe in it — and question it, too — to animate our own great experiment.
Reply Retweet Like
John C. Havens Jun 8
Also giving people access and control over their data. Just sayin.
Reply Retweet Like
Ryan Carrier Jun 8
We're going to make this happen! The timing and the societal temperature is right... even Bezos has agreed with the idea publically.
Reply Retweet Like
Ryan Carrier Jun 8
How about like this?
Reply Retweet Like
Dr D Jun 8
Replying to @katecrawford
Reads very reactionary, vague and self-congratulatory.
Reply Retweet Like