Tensions in Google's ethical AI group increase as it sends demands to CEO • Hacker News Board

Re: Tensions in Google's ethical AI group increase as it sends demands to CEO

#71

Post by atsjie » Thu, Dec 17, 2020, 12:40 PM UTC

fatbird wrote: post #40

tomcam wrote: post #24

I have contempt for a lot of the things that Google does, but I have a sincere question about all of this. Why should they employ anyone who criticizes them that way in public? I would say the same about Peter Norvig or Guido van Rossum. If you ran the company, or the department, or owned your own firm where employees aired your dirty laundry like this, why should you be expected to keep them on?

Because the academic standard for research, and the original motivation for tenure, allows for researchers to criticize anyone, including their own institution, as long as it's done and presented as the proper product of research. In other words, if a researcher is doing their research properly, it shouldn't matter if they're attacking their own institution or even colleagues, because it passes scientific muster and…

"Timnit Gebru: Google and big tech are 'institutionally racist'"

https://www.bbc.com/news/technology-55281862

It's good to be critical and reflect on the company you're working for. There are bad cultures and attitudes in any big company, not just tech.

But if you're going to sprout your opinions publically about your (ex)-employer like that, sorry, but I wouldn't want her in my team. She sounds like too much of a headache to deal with. Her work must be truly extraordinary to warrent having that headache.

Re: Tensions in Google's ethical AI group increase as it sends demands to CEO

#72

Post by notsureaboutpg » Thu, Dec 17, 2020, 12:43 PM UTC

thenightcrawler wrote: post #28

tomcam wrote: post #24

I have contempt for a lot of the things that Google does, but I have a sincere question about all of this. Why should they employ anyone who criticizes them that way in public? I would say the same about Peter Norvig or Guido van Rossum. If you ran the company, or the department, or owned your own firm where employees aired your dirty laundry like this, why should you be expected to keep them on?

Was it public before she got fired?

She also mentioned she had sued Google before while working there

Re: Tensions in Google's ethical AI group increase as it sends demands to CEO

#73

Post by notsureaboutpg » Thu, Dec 17, 2020, 12:50 PM UTC

justforfunhere wrote: post #67

What kind of day to day work someone in an Ethical AI group would do?

They write papers to propose the idea of "AI Cards" which are descriptions of AI models, the situations they are useful in, and some stats about their performance when used on subjects who are from minority groups and/or women.

A real paper she was listed as an author on.

Re: Tensions in Google's ethical AI group increase as it sends demands to CEO

#74

Post by baryphonic » Thu, Dec 17, 2020, 1:12 PM UTC

_y5hn wrote: post #52

baryphonic wrote: post #31

Earlier quoted context omitted.

Truly astounding when the rallying cry to "end white supremacy" and "give representation to underprivileged/underrepresented groups" involves light-skinned Anglophone people dictating to darker-skinned Spanish-speaking people how the Spanish language should work, and in a way that betrays total ignorance of actually communicating in Spanish.

See the new series, The Long Story, it's hillarious and sad.

Wow, that's unbelievable! Much of the dialogue in the screenshots is extreme cringe. Do wokes ever interact with actual human beings, not to mention high school kids? And I still don't know how we're supposed to pronounce "Mx" like in the developer's handle. "Mix?" "Mux?" (Would a male be "Demux" in this scheme?)

The saddest part are the glowing reviews in Vice Motherboard and The Verge.

Re: Tensions in Google's ethical AI group increase as it sends demands to CEO

#75

Post by je42 » Thu, Dec 17, 2020, 1:50 PM UTC

tomcam wrote: post #24

I have contempt for a lot of the things that Google does, but I have a sincere question about all of this. Why should they employ anyone who criticizes them that way in public? I would say the same about Peter Norvig or Guido van Rossum. If you ran the company, or the department, or owned your own firm where employees aired your dirty laundry like this, why should you be expected to keep them on?

If they want to fire her because of her public twitter message, they should be very explicit about it.

However, the paper trail that has been published publicly does not suggest she has been fired because of these messages.

Re: Tensions in Google's ethical AI group increase as it sends demands to CEO

#76

Post by bzb6 » Thu, Dec 17, 2020, 2:03 PM UTC

SpicyLemonZest wrote: post #46

thenightcrawler wrote: post #28

Earlier quoted context omitted.

Was it public before she got fired?

It was, yes. Public statements she made before getting fired include: > A Black woman can't survive around here doing any type of research let alone research thats designed to make those in the majority uncomfortable. And guess who can apparently unilaterally order you to do things without even thinking you're someone they should converse with. ( https://twitter.com/timnitGebru/status/1331757629996109824 ) > [The VPs…

A black woman? How is she black?

Re: Tensions in Google's ethical AI group increase as it sends demands to CEO

#77

Post by pietrovismara » Thu, Dec 17, 2020, 2:05 PM UTC

shoo wrote: post #30

One thing I've wondered: what is the purpose of Google's ethical AI group? What utility does it provide to the corporation? I suspect the primary motivation for funding ethical AI work is _public relations_ (or less kindly, corporate propaganda) - to help mitigate damage to Google's brands and reputation by public concern surrounding artificial intelligence. It is well understood that it can be difficult to get someo…

It's just marketing/pr. It's supposed to give them a clean enough image while they keep doing whatever they want under the hood. It won't certainly alter the status quo or influence any decision that could result in reduced profit for shareholders.

Re: Tensions in Google's ethical AI group increase as it sends demands to CEO

#78

Post by croissants » Thu, Dec 17, 2020, 2:09 PM UTC

jmeister wrote: post #48

They should hire Pedro Domingos to clean house and rebuild the group. This talk by him on AI ethics has a lot of maturity and insight on the topic: https://m.youtube.com/watch?v=yIRL4xtmXE4

Please no, Pedro Domingos has spent the past week picking fights on ML Twitter, at this point a random grad student probably scores higher on "maturity".

Re: Tensions in Google's ethical AI group increase as it sends demands to CEO

#79

Post by tootie » Thu, Dec 17, 2020, 2:10 PM UTC

atsjie wrote: post #71

fatbird wrote: post #40

Earlier quoted context omitted.

Because the academic standard for research, and the original motivation for tenure, allows for researchers to criticize anyone, including their own institution, as long as it's done and presented as the proper product of research. In other words, if a researcher is doing their research properly, it shouldn't matter if they're attacking their own institution or even colleagues, because it passes scientific muster and…

"Timnit Gebru: Google and big tech are 'institutionally racist'" https://www.bbc.com/news/technology-55281862 It's good to be critical and reflect on the company you're working for. There are bad cultures and attitudes in any big company, not just tech. But if you're going to sprout your opinions publically about your (ex)-employer like that, sorry, but I wouldn't want her in my team. She sounds like too much of a he…

I wouldn't want to run a racist company. And I wouldn't want to create phony pretense for firing someone if there was ostensibly valid justification they could have used.

Re: Tensions in Google's ethical AI group increase as it sends demands to CEO

#80

Post by tootie » Thu, Dec 17, 2020, 2:13 PM UTC

fatbird wrote: post #40

tomcam wrote: post #24

I have contempt for a lot of the things that Google does, but I have a sincere question about all of this. Why should they employ anyone who criticizes them that way in public? I would say the same about Peter Norvig or Guido van Rossum. If you ran the company, or the department, or owned your own firm where employees aired your dirty laundry like this, why should you be expected to keep them on?

Because the academic standard for research, and the original motivation for tenure, allows for researchers to criticize anyone, including their own institution, as long as it's done and presented as the proper product of research. In other words, if a researcher is doing their research properly, it shouldn't matter if they're attacking their own institution or even colleagues, because it passes scientific muster and…

This is it exactly. She's applying a standard of academic rigor to Google and they can either accept it or not. If they want to filter what kind of research they publish, then they have to admit to themselves (and their employees) that they baking pro-Google bias into their work and it need not be taken too seriously by academics. Maybe that kind of bias is irrelevant when you're publishing algorithm or mathematical work where bias is less inherent, but if you have an AI Ethics team you should be prepared for some controversy. And they clearly weren't and panicked.