Earlier quoted context omitted.
Body of work aside, she absolutely is an aggressive-ranging-into-toxic Twitter personality and a very strident proponent of a divisive and dogmatic strain of diversity advocacy.
Her spat against Yann LeCun felt very wrong to me. That is not how fellow academics/researchers should treat each other. Maybe Yann LeCun was wrong, the way she conveyed the message disregarding any knowledge he might have on this was definitely not civil. She could have chosen a different way to address the complaints here, not by getting overtly confrontational. Maybe the review group got it wrong, maybe they were…
Google staff rally behind fired AI researcher
111–120 of 326 posts
Re: Google staff rally behind fired AI researcher
#112Google (and Google Brain in particular) benefited immensely from Gebru's work, and her being affiliated with them. This can be seen in how often they (including Jeff Dean) pointed to her work to show they were taking questions of fairness and bias in AI algorithms seriously. The flip side is that you have to respect the researcher, who frankly has more credibility than you do as an organization in this space. Even to…
> it is incredibly anodyne I don't get why anyone is surprised that Google doesn't want a paper saying ‘we should spend less money on ML’ published under their name, when the whole company is ML driven and pushing for growth there. What other large company would allow that sort of thing?
Re: Google staff rally behind fired AI researcher
#113I'm tired of FAANG "rallies" where the numerator is such a small percentage of the denominator. I bet one could get "hundreds" of Google employees to rally behind a flat earth and many more for the abolition of mayonnaise.
I don't work for Google but the abolition of mayonnaise is a cause I can get behind.
Re: Google staff rally behind fired AI researcher
#114I'm tired of FAANG "rallies" where the numerator is such a small percentage of the denominator. I bet one could get "hundreds" of Google employees to rally behind a flat earth and many more for the abolition of mayonnaise.
The numerator is such a small percentage because so many fear losing their jobs for speaking out. In that context, this number is quite significant. Even if you disagree with their stance, you have to admit that it takes courage to speak up in these kinds of situations.
Re: Google staff rally behind fired AI researcher
#115Earlier quoted context omitted.
> divisive and dogmatic strain of diversity advocacy. Reading the HN comments section sometimes feels akin to doing archival research on letters to the editor during the peak of the civil rights movement. It's astonishing how consistent the tropes are.
Relating GP to a '60s era segregation apologist for offering a dissenting opinion is proving their point.
I think it is very illuminating to read letters from that period, because they aren't just vitriolic racists like schools often teach. These are people trying to come off as "very reasonable", offering a civil, moderate path forward for the South, etc., etc. It's fascinating to see that rhetoric being deployed to defend what is an obvious evil and compare to who is using that rhetoric now.
Re: Google staff rally behind fired AI researcher
#116Earlier quoted context omitted.
That's exactly what a lot of companies do when someone gives 2 weeks notice. Send them home right away and pay them for the 2 weeks so you can keep them from filing unemployment and have them on the hook for any exit interview / handoff questions etc. They always want you to sign "something" as part of the process, too.
Yes, I'm aware of how FB&Google work. > when someone gives 2 weeks notice. But she didn't give 2 weeks notice. This seems to be the key detail.
If you give an employer an ultimatum and they don't want to give in to your demands then is that giving notice or quitting on the spot or something else? Should her managers/google legal have waited for her to tell them when exactly she was going to leave?
Re: Google staff rally behind fired AI researcher
#117Earlier quoted context omitted.
The numerator is such a small percentage because so many fear losing their jobs for speaking out. In that context, this number is quite significant. Even if you disagree with their stance, you have to admit that it takes courage to speak up in these kinds of situations.
This argument can be made about any unpopular subject. Americans really want full-blown Communism, but are too afraid of speaking out for fear of retaliation from jack-booted capitalist thugs. Prove me wrong.
Re: Google staff rally behind fired AI researcher
#118After reading everything that has come out (e.g. her email) and her history of highly toxic Twitter behavior, I honestly can't blame Google for accepting her resignation (she was not fired).
Firing her much sooner than the date she requested seems really foolish on Google's side. Arguably, one of the primary reasons a profit-oriented company would hire someone for this kind of job is for the good PR. If they offer to resign why would you not simply accept the terms of resignation, why would you rush it to happen faster.
Even if she is acting in bad faith Google could have saved themselves a big headache by at least giving her a month or something, even two weeks, to exit the company.
Re: Google staff rally behind fired AI researcher
#119Earlier quoted context omitted.
Heck, even kneeling during the national anthem is “too much”...
Because it "disrespects the veterans" in some people's eyes. Astonishing, isn't it, that the same people who found issue with Kaepernick's protest probably have never heard the name "Nate Boyer"[1] before-who is both a veteran of the armed forces, and a veteran of the NFL. It's never good enough. [1] https://ftw.usatoday.com/2018/05/nfl-49ers-colin-kaepernick-...
Re: Google staff rally behind fired AI researcher
#120She was working on a known-to-all controversial research project, which was asked to be retracted via an anonymous feedback doc delivered via HR! That is a very screwed up thing to do. All reactions from her from that to me are totally justified.