Google employees say AI researcher did not resign, despite what exec claims
cnbc.com
cnbc.com
And yet there are a bazillion articles debating what verb best describes that pair of events. I get that there is legal relevance, and let the lawyers argue that one. But the rest of this seems to be a court of public opinion thing, in which case it should have no relevance given that everyone is agreed on the facts. Why do we care so much about the one-word linguistics of it? How about you just don't use a single verb alone, and instead spend the literally two sentences it takes to explain that part?
By calling it a resignation, Google is making it sound as if it was an amicable choice she made, not their judgement call about her actions.
Words matter. Word choices always matter. Especailly single words which come with a lot of built-in context.
She did not get a chance to offer up her resignation, Google terminated their employment relationship before that opportunity arose.
And thus, it is completely reasonable to boil it down to:
Google fired Timnit Gebru.
Words matter. Google is lying when they claim that Timnit Gebru resigned. Google should absolutely be called out for this lie.
It's mostly the side that supports Google management that is focussed on it as the primary issue; the side backing Gebru is more focussed on the triggering event itself, the unusual (for Google AI, and apparently for the field in general, as far as internal reviews go) review process applied to the paper on which Gebru was lead author and which she was told to retract, with the resigned/fired issue as an additional indicator of Google management being deceitful, but not the primary issue.
Additionally, in this world where barely anybody reads past a headline, that discussion does have _some_ merit, even if every other part of this issue is _more_ important...
https://techcrunch.com/2013/03/21/a-dongle-joke-that-spirale...
Source? Last I heard the two men were hired elsewhere fairly quickly. Their names are also hardly ever associated with the story.
Don't make genitalia jokes unless you have balls of steel.
This should be handled internally and not aired in public across several media sites. Why should I care about the dirty laundry of FAANG and some random individual?
The media of course enables it, which I think is a sign of a growing, large scale, rational fragility across society (demand controls their publication, there is a large demand for this type of information)
I will have to search for them now but I saw people doxxing some of the current Googler's identities if they agreed with the resignation. This is extremely toxic.
Even the old media seems to love Twitter and it escapes FB level of scrutiny.
Facebook is now far less egregious & more of a soft environment to socialize with people you personally know. Bubbles/pockets of societal function and some large-influence which is lesser enabled than the prior.
Twitter is a platform that justifies large societal movement through "influential information".
(I don't have the best english to articulate here)
Now social media sites encourage people being public figures, which is an insanely bad advice for the vast majority of individuals.
No one forced her to do any of the following things:
1. Griping about the racist reviewers of her paper on Twitter.
2. Sending an email to a internal list criticizing Google's behavior and telling people to stop working on diversity.
3. Demanding the names of reviewers and threatening to resign if they weren't provided.
#2 or #3 would get you fired or cause your resignation to be accepted any place that I've worked.
#1 wouldn't get you fired but it's not going to help your career.
FAANG employs a tremendous number of technology workers, are key drivers the industry-wide technical agenda formally (standards bodies) and informally (open source project management, areas of active research), are often the first line of serious education for new workers, and the fallback for workers when riskier undertakings fail.
If anything far more of their internal management (project and people) processes should be scrutinized.
I have no idea what or how this individual matters at all or the tech that she is relevant in. Not everyone at FAANG are industry leaders, and there are several cogs just like at IBM. I fail to see why I should care for every individual starling who can't stand out other than the company they work for.
This random individual along with James Damore and that random FB grad are not on the same level.
Your arguments tell me more that FAANG should be broken apart. Public attention and scrutiny would achieve nothing in affecting Google's decision making here.
> I have no idea
Why are you commenting then?
It's rather telling that your entire perspective is to demonize and diminish Gebru, who you don't even bother to name in your criticisms.
Why are you trying to establish some "level" distinction between Damore and Gebru. These aren't even remotely comparable situations.
Public attention and scrutiny is why we're talking about this, and have been, for a week now. It's pretty head in the sand to think this doesn't have impact.
Fine, there are AI stars too, but the article could lead about what they have done and why they are important no? A discussion involving Jim Keller would lead to his contributions to x86, Apple A10 series, Tesla chips and Zen.
I don't want a hand wavy discussion over ethics and AI, I want an explanation as to what they have contributed to the field that justifies why their opinion matters.
>Why are you trying to establish some "level" distinction between Damore and Gebru.
I am trying to establish that these random individuals don't matter to me. They are not known for their tech contributions but for the dirty laundry being exposed to the public.
These individuals from an external perspective seem to be known for being employed at Google and for some public controversy that cannot be influenced externally.
My point is why are these individuals in the news? Why should I care? They aren't industry leaders and from my perspective appear about as important as 'GE fires engine tester'.
So, for us that care about google's use of AI and ethical use of it, it's important to understand what happened.
> She tweeted Wednesday that Google had abruptly fired her after a disagreement about a research paper that scrutinized possible bias in large-language AI models, which the company asked her to retract.
The article should be focusing on the disagreement here.
Instead the article focuses on employment which I don't care about.
I have not heard anyone express this fear, certainly not in the AI community.
I have not seen any part of the Timnit Gebru incident that paints Google in a bad light or corroborates any type of inconsistency or unfair judgment.
Google has done plenty of that stuff in other situations, just not in any part of the handling of Timnit Gebru’s very inappropriate behavior.
If Google fires a janitor, does that mean they don’t care about a clean work environment? If they fire a VP of engineering, does that mean they don’t care about engineering?
No, it means that person didn’t work out. And considering how she is handling it by going nuclear with the press and the twitter mobbing, I for one am not the least bit surprised.
That said, I don't believe that decisions about application of AI can be made internally at Google. These are basically PR jobs that come with the product.
It's 1 employee out of 120k. There are probably other ethicists in AI at Google.
Also the story seems to have devolved more specifically to employment rather than talking about the specific ethical issues relating to Google's language models.
I mean google employees including most of her coworkers seem to think it's problematic.
None of that is useful here which from an external perspective looks like a employment dispute. She isn't exactly doing a Snowden here.
Even if the individuals internally view it as problematic, I am not gonna stop using Android, Google or Youtube. What are these individuals expecting to achieve from airing the dirty laundry in public?
Are they expecting people's opinions to have changed at all about Google?
The firing/resigning/petitions are irrelevant.
The wider community important information would be the specific ethical issue she is keen on raising.
When the triggering event involved a collaboration involving non-Google parties, that became pretty much impossible.
> Why should I care about the dirty laundry of FAANG and some random individual?
Maybe you shouldn't, but not everyone outside of Google is you. Other people in the field of AI ethics might want to know, because Google’s practices when it comes to employee research have a significant impact on how publications from Google authors are approached. And people who might consider working at Google, especially in the same field, have an interest in information about the working environment, positive or negative.
[1] https://twitter.com/timnitGebru/status/1334900391302098944
The exact phrasing she used is (probably - IANA employment lawyer) very important. If she said something like "Do X, or I'll have to reconsider if I want to continue working here", that doesn't sound like an ultimatum. But if she said "Do X, or I'm going to quit on December 5th", then that definitely sounds like an ultimatum you're begging the company to take you up on.
She had already sent an email asking other people to stop their work on diversity. It's at best naive to think that Google (or any employer) would give her a few more weeks to lower morale or cause trouble.
- I will look for work elsewhere
- I will work on a last date
- My last date is <X>
- This job isn't working out
- I don't want to work here anymore
- I offer my resignation
- I quit
Edit: if you say "I don't want to work here anymore." And you get your "resignation date moved up" you still get your unemployment benefits because you were fired.
I can't speak for California, but in my state, if you are fired and not laid off[1], you don't get unemployment benefits. I've anecdotally heard this is also the case in California. However, I've also heard that many tech companies state they will not dispute unemployment claims when fired (my company certainly says that), so people in these "privileged" companies often do collect unemployment even when fired.
[1] I hope people know the difference.
There might be specific criteria that needs to be met for each, so details matter. My guess is the reason my employer does not dispute unemployment claims is they've calculated that the cost to dispute them exceeds the amount they expect their unemployment insurance contributions to go up.
I am assuming that was the red line, I am not a manager but it would have been for me.
I am not really even sure how meaningful an email telling your coworkers to stop working on their projects is. It really depends on the tone. When I was at Google, I used the internal social media extensively. I am sure at some point someone said something like "I'm really depressed about how my project is going" and I would reply with something like "You should join my team! I have a req!" That's technically telling someone to stop working on their project, but is a completely acceptable thing to do.
Ultimately the exact details matter, and we don't have them.
If you immediately fire someone for that, that may be illegal age discrimination.
Cheers.
What do you do?
If there’s no notice period (legal only in “at will” states, I believe) then it would seem reasonable to fire them immediately.
I think you'll find from a legal standpoint, it does.[1] At least I know that they teach in negotiations courses that any time you make a counteroffer, the original offer is legally null and void. So if I give you an employment offer that is good for 30 days, and you counter the following day with a higher salary, then I legally can say "No, and we will not hire you" (i.e. the 30 days clause became invalid once you gave me a counteroffer).
This isn't exactly the same, but we are at least taught to treat it similarly.
Also, she didn't merely say "I will look for work elsewhere". She said something to the effect of "Let's meet to decide my end date." There's a significant difference.
[1] Although being an at-will job, the legality is rather moot.
In fact, it specifically does not: https://cuiab.ca.gov/board/precedentdecisions/docs/pb39.pdf
A manager told it was effective immediately and mentioned her final paycheck.
What's the parallel between those facts and the dispute here? Just not seeing any applicable precedent.
> What's the parallel between those facts and the dispute here?
That it's, at best for Google, the exact same thing, except that Google gave Gebru, in writing, an explicit, non-resignation reason for the early separation, making it even less disputable that it was a firing and not merely accepting the (disputed) resignation? Paraphrased most favorably to the Google management narrative, Gebru said “do X or I will conduct discussions around setting a date on which I will resign” and Google management said “we accept your resignation and, further, based on an internal group e-mail you sent after this dispute started which we find unacceptable, we are ending your employment immediately.”
The only interesting implication is unemployment, but if it's a termination, Google might tell the unemployment office it was for insubordination.
Otherwise, this is an all-in bet, and you need to be fully committed to leaving before doing it.
If I tell Perl to
do_the_thing() || die();
I've seen people walked out the day they give notice, but I assume they weren't technically fired, and paying out their two weeks is the right thing to do.
Sending an email to your boss saying "I'll resign unless X" is a serious ultimatum. You need to be prepared to not have a job when at the end of it, and you were the one who initiated the process. I'm not too concerned with if she technically should have been given a chance to take it back. Don't send an email threatening to resign, lose your job, then say "actually, I was fired."
> and without respecting her stated plan to work on an end date.
You don't really want to let an employee already sending inappropriate emails to a wide audience have another week to send more.
I've been in a professionally self-destructive place before and I didn't realize it until later. 100% speculation here, but I wouldn't be shocked if she had a death wish and was hoping for some version of this outcome. Could be frustration with work, covid depression, or something else.
The debate about her being fired or resigning is a debate about who to believe about what led up to it. People who think she resigned see her saying she was fired as evidence of a victim complex and publicity seeking. People who think she was fired see Google saying she resigned as evidence of unfair treatment and dishonesty.
> I’m always amazed at how people can continue to do thing after thing like this and then turn around and ask me for some sort of extra DEI work or input. This happened to me last year. I was in the middle of a potential lawsuit for which Kat Herller and I hired feminist lawyers who threatened to sue Google (which is when they backed off--before that Google lawyers were prepared to throw us under the bus and our leaders were following as instructed) and the next day I get some random “impact award.” Pure gaslighting.
This excerpt is from the email that according to her triggered the firing. This person threatened her employer with a lawsuit in the past and now people are outraged that she got "abrubtly fired". Arguably, the way Google treated her was even too lenient: many companies would have fired her at the time her original threat was made.
It is part of Google's culture to not blindly agree with leadership on every issue and you're encouraged to voice differing opinions, but I guess some take this too far in the other direction and assume that every kind of negative feedback is appreciated and "protected".
> you know these last 2 weeks were supposed to be my vacations and we’re currently on a road trip.
Probably best not to send an email threatening to resign, then go on vacation and expect things to be normal.
Upton Sinclair said "It is difficult to get a man to understand something when his salary depends upon his not understanding it."
Seems like there should be a corollary for companies and difficult employees. Like public complaints that jump the chain of command from replaceable employees usually result in employee being replaced.
If this was an issue of the researcher unprofessionally airing her frustrations on a listserv with direct reports that's one thing. It's another thing to say being upset at all was an overreaction because the treatment was totally normal and that her frustration was a resignation, when in fact, none of that is true in any good faith interpretation.
Searching in HN under "Gebru" (the employee's last name) pulls up a bunch of past threads.
1) If you ask me, this is a story about an influential technology company, one whose approach to business has an outsized influence on the entire industry, has unethically fired an employee while making mis-statements to cover it up.
2) Others will say: this is about a ethicist being unfairly treated with different standards because of her gender and race, a microcosm of what it is like to be a minority race and/or female in the tech industry.
3) Still others will say: This is about an employee who disrupted the workplace. She rudely interacted with her colleagues and prominent names outside of the industry (Y. Lecun). She prevented collaboration by making spurious claims of sexism and racism.
4) Another line of thought that is a variant of #1: This is about an employee who did not follow company rules and processes. She was fired as a result. The statements privately and publicly about "resignation" were an attempt to not publicly disparage her
Did I miss any others? With a nod to the Rashomon Principle, your interpretation of these events depends on your perspective.
Sounds like a perfect fit for Google's new Chief Legal Officer. Or maybe SVP Social!
This whole thing is a confusing mess.
It's terrifying that these are the kind of views people in high places of major tech companies have. And that we are entrusting our tools of societal communication to those who see themselves as fighting in a culture war.
* Saying a high profile project is botched, mismanaged, and doomed to fail. (Anyone remember google+?)
* Mocking and criticizing the SVP who led the project.
* Rejoicing in the departure of the said SVP.
* Talking about one of the founders having an affair. (Well, to be fair, the news broke outside first.)
* Memes disagreeing with company decisions, including one that literally said "What the fuck are you doing" for firing someone.
* And, last but not least, complaining that the quality of microkitchen has gone downhill since $(the year you joined).
* Picked a public Twitter fight with Yang LeCun
* Her peers are so afraid of critiquing her paper, that they demand anonymity through HR
* Tells co-workers to stop working on documents
One of the things that has been emphasized in tech for the past several years is not to allow justifying retaining toxic people on account of their talent. In addition, your life experiences don’t justify being toxic. For example, even if a person’s Dad was an abusive alcoholic, that would not justify them being abusive to co-workers.
My point wasn't even about the truthiness of the claims of either side, but rather that if party A claims X, and then parties B and C just repeat the claims without being able to add anything - that's not news. That's just a person having allies who believe their claims.
If Microsoft chimed in and said Timnit actually resigned, no one would write an article on that (besides for how bizarre of a statement that would be by another company), because no one expects Microsoft to be apprised of what happens between Google and its employees.
Full text:
Just wanted to say that I was stunned by what just happened to Timnit Gebru, who was in my team until a few days ago (I'm sure you've all read what happened already). I have always been and will remain a strong supporter of her scientific work to make sure that AI becomes a positive force for the minoritized, as well as her generosity and tireless actions to lift the voices of the silenced ones. She taught me a lot and still is. I stand by you, Timnit. I also stand by the rest of my team who, like me, was stunned and is trying to make sense of all this. In particular the Ethical AI team, but also the broader Brain Research team.
Also I read the highlights of the paper linked here and it seems fine and reasonable. These are reasonable questions to ask. Sure we can mitigate the cost of training these models with green energy or finding new optimization./compression techniques etc... but that doesn't mean that with ever evolving data volumes we will ever fully mitigate that. Bitcoin for all its merit consumes a lot of energy. Is that more than USD or something else?
Another thing to consider is that AI right now has a rich get richer winner take all dynamic. We have titans battling it out for NLP supremacy. Deep mind just revolutionized protein folding which has direct implications into biomedical research and technology. Small academic labs are at an extreme disadvantage unless they collaborate with big tech. Interestingly physics research evolved in a similar way with large particle colliders and telescopes started up. So you can make a direct analogy to that, that the basic research can only be done with million dollar investments. That's an ethical question. GPT-3 trained on misinformation would make pizza gate look tame in comparison. So there are huge social implications.
Even real datasets collected from society and social data will have implicit and explicit biases in them. There are huge ethical concerns with data, data collection and with the interpretations AI models will make.
Your credit score, your social credit score, a company that refuses to hire you because a chat bot has labeled you subversive.
The TSA/DHS blanket banned people from flying and gave no feedback or transparency into why. It took members of Congress to get some people redressed.
Google needs a massive clear-out. People aren't doing their jobs anymore there, they've become so obsessed with fictional narratives about non-existent racism and sexism.
I bet Jeffrey Dahmer didn’t have a history of complaining about cannibalism either, but that doesn’t necessarily make him the good guy.
1. Did in fact go through and pass the internal approval process that he implied it didn't.
2. Was then subjected to a second, intransparent "review" that is not customary practice at Google.
3. The results of that "review" were presented to the author as a decree (which also is not customary practice).
4. And when the author subsequently protested that she would not work under such rules (as well she should) that was taken as a pretext to let her go — without even involving her direct manager.
Under those circumstances, Dean's response is misleading and evasive, a practice often characterized as "gaslighting" or, in the term that the grandparent comment ridiculed, a "micro-aggression".
It's not about whether you're a "white dude", it's what rules you set for "white dudes" vs. "non-white non-dudes".
Did Jeff Dean think Timnit was a white dude and so let her publish, but when he found out she was a brown woman, he said he cannot stand for this?
These 4 points were all taken from the blog post I referenced, which I found well documented and thus credible:
https://googlewalkout.medium.com/setting-the-record-straight...
> Why would they publish multiple of her papers on basically the same exact subject, but then decide they need to stop her at all costs for this paper?
You tell me. If they were truly on "basically the same exact subject", there can't really be a scientific explanation for some to be publishable and others not, can there?
Or, your theories are flawed in some manner and don't actually capture the truth. I'm asking you how previous behavior would fit into the theories that you yourself believe.
It's obvious that people like Gebru terrorise their colleagues by claiming anyone who doesn't join their racist jihad against white men are terrible and need to be fired. When Pichai canned Damore he set Google up for a never ending strem of events like this one: he basically told the extremist activist wing of the workforce that they were in charge, not him. Very few tech CEOs in America are more deserving of being replaced than Pichai but given the voting structure of Google stock it's hard to see how that can happen.
As a reminder, this isn't the first problematic incident involving Timnit Gebru either. She previously went on an unhinged Twitter attack against Yann LeCunn, one of the pioneers of deep learning. He won a Turing award for his work on this topic and is one of the foremost experts on it, and understands its various nuances very well. When she was challenged, Timnit's line of attack was leveraging a word soup of progressive identity politics, demanding Yann LeCunn apologize to her, claiming harm was done to the community, etc. This highlighted her aggressive tactics (weaponization of her Twitter following and leveraging cancel culture) and it is clear she must be a very difficult employee to manage or work with.
Coming back to the article linked here: whenever there is some incident (a firing, a walkout, a protest) at one of these big tech companies with a large activist employee population, articles get written that purport to present the opinion of the employee base at large. But usually, these articles only represent the loud minority that is willing to speak up and faces no repercussions. The employees that agree with Timnit Gebru leaving the company (supporting the company's stance) don't have psychological safety. Timnit wields the power of a big following of social justice activists on Twitter, and could shame/destroy people who speak up against her. The small number of employees voicing their support for her, even with total psychological safety, is a tiny percentage of Google's employee base. Those numbers paint a false picture that she has broad support when she doesn't. Perspectives from other Google employees are only available on anonymous forums like Blind (https://www.teamblind.com/post/We-stand-with-Jeff-Dean-and-Y...) and Reddit (https://www.reddit.com/r/MachineLearning/comments/k77sxz/d_t...).
Lastly, this has all been discussed in depth previously in a few different discussions on HN. The one with the most comments is https://news.ycombinator.com/item?id=25292386.