All the large tech companies train their employees how to request advice from counsel and use privileged and confidential email communication. I've never worked for Google but worked for two of the other MAGMAs and it was pretty standard. Nothing I saw about the training or common usage seemed suspicious; I think it's in the eye of the beholder.
After several big companies got their internal email used against them in lawsuits, esp. antitrust lawsuits, they focused heavily on training employees how and when to discuss sensitive topics related to competition, e.g. "don't say exaggerated things like "we're going to kill the competition", etc. And the trainings emphasize being extremely accurate and precise, and not speculating as to legality unless you actually are a corporate counsel. They teach you to take such questions to corporate counsel, e.g. "We're thinking about doing X, what do you think?" on P&C email threads.
I'm sure that some people could overuse this, but again, I didn't see anything suspicious, and I remember seeing one guy get called down for addressing email to corporate counsel as P&C and not actually asking a question.
I'm sure in every org there are weasels who would try to misuse this, but I'd have to see more evidence before I would believe that this was cultural at Google.
>You have to assume the authorities, and others, are listening.
Given that Google is a publicly traded company and thus has to follow federal securities laws and keep much business communication on the books anyway I don't know why we're framing this as 'the government may be listening'.
I don't have any doubt that none of what Google does is illegal. Same goes for 99% of the employers I had.
But illegality is not the only reason things end up on NYT.
For example, my former boss was clearly using his positions in various nationwide institutions to make sure things were discussed the way he wanted to. Depending on the institutions (for example being in an audit one as a "consultant" and another one being audited as a "chief-whatever-officer" is not illegal but it's downright suspicious) I never said anything about it but the newspapers caught up...
But there is nothing suspicious or sensational about that advice! It's a short, concise way of warning employees of really any business that electronic communication never goes away and to be mindful of how they communicate.
"You have to assume the authorities, and others are listening" doesnt this seem sort of outrageous to an outsider, since Google pioneered the practice of searching every user's emails? like, they dish it out but also hide their own ?
Secondly the extreme and secretive legal agreement not to talk about your work, even years after leaving the company.. it seems Google got away with a lot, by virtue of "superpowers" that is objectionable for good reason, for the business community in general.
It makes sense to ask your lawyer legal questions, but I think this complaint is more about CCing lawyers on sensitive communications so you can write "Attorney Client Privileged" despite not actually wanting to discuss the matter with your attorneys.
I'm almost surprised they didn't just put a lawyer on every mailing list and put the magic words in a mandatory footer.
IANAL, but the trainings I took emphasized that regardless of the P&C label, it was NOT P&C if you weren't asking the lawyer a question, and any part of the discussion that happened before someone engaged counsel was not P&C. We also were instructed not to add people to such threads, but rather request that the lawyer do it.
I am also not a lawyer, but I definitely saw a lot of a behavior where one would CC lawyer cat, and start the email with "please advise" on emails that contained sensitive information but no legal questions.
It bothered me a little, but I also wanted to be able to discuss sensitive numbers in emails and also be productive, so I kinda ignored it.
No one does this and the training specifically says it won't work. You have to address the communication directly to the counsel with a specific question. And if there are too many people in the CC, that's also evidence that it wasn't really a question to counsel and it would be deprivileged.
Of course the training can't say that it will work, but it obviously does. It means that in the initial discovery, these emails won't be sent to the other party, claiming privilege. Then, if somehow the other party finds out that they exist, they need to litigate this particular case to decide if such and such email was indeed privileged or not.
Now, if a company had an official policy of misusing their legal team in this way, they could probably be assured for that in itself, the lawyers who accepted it personally penalized etc. So, of course the company trainings will tell you very specifically not to do this, and perhaps lawyers will even reply from time to time to emails that are trying to abuse this system. But it seems, if the government is right, that in this case this was all a rise trying to establish plausible deniability for all involved of a widespread practice.
Information being withheld during discovery due to privilege still has to be disclosed to the other party as being withheld. It is called a privilege log and is exactly for the purposes of being able to litigate the propriety of said withholding/applicability of whatever privilege.
Yeah, I'm not an expert, but my understanding is that's clearly not privileged. It's not surprising the government is trying to get access to them, but I also don't think it's evidence of Google trying to hide anything. There is training, but a lot of people ignore it and treat Legal haphazardly.
The article claims that even the CEO of Google did this when sending an email to the CEO of YouTube. These are people who should know better, and if true, then it’s not exactly the innocuous case of the inexperienced individual accidentally marking a communication as attorney-client privileged.
I don't think "attorneys didn't respond" is enough evidence to declare "that's clearly not privileged".
An attorney missing an email is irrelevant to whether or not an email is privileged. The more important factors are probably 1. Was the message actually to the attorney, or were they just included for the appearance of privileged and 2. Was a response actually expected from the attorney?
Well, if you copied the attorney on every email it would cease to hold ambiguity on if the communication was actually private or not. If you just copy every communication with an attorney no judge would believe it.
If you keep it down to some reasonable percentage of communications, you're granted benefit of the doubt in the ambiguous case (innocent until proven guilty and all).
It's also not legally correct. Just putting the words "privileged and confidential" on a document and copying an in-house lawyer doesn't render it subject to privilege.
That being said, the practice is incredibly widespread. While it borders on unethical, I'm yet to see anyone actually penalized for slapping P&C on every e-mail willy-nilly. At worst, courts just rule that privilege doesn't apply and that a document must be produced.
Even that is rare, for obvious reasons (if opposing litigants don't even know a document exists, how can they argue that the "P&C" label wasn't legitimate?).
Withholding a document due to privilege has requirements, namely production of a privilege log. It should pretty much never be the case that an opposing party is unaware of a document’s existence due to privilege.
> All the large tech companies train their employees how to request advice from counsel and use privileged and confidential email communication
What is described here is not training on “how to request advice from counsel and use privileged and confidential email communication”, it is direction to pretextually pantomime requesting advice of counsel for all discussion touching on certain topics in an effort to provide a pretextual fig leaf for deliberately withholding nonprivileged information expected to be relevant to future litigation with a premeditated false privilege claim.
They are both very different kinds of restructures.
Alphabet is a pure holding company. It doesn't have many employees directly working for it (other than top executives, some finance people and other minor overhead). And different Alphabet companies operate fully independently with next to zero sharing of employees, resources, money or tech. When people in these threads say Google, they specifically mean the company which makes up most of Alphabet and works on search, ads etc., not Calico or DeepMind.
Meta on the other hand is a rebrand of Facebook (the company). There is no holding structure. If you were a Facebook employee, you now work for Meta. Employees at Instagram, Whatsapp, Oculus, shared infra teams, and other minor projects all work for Meta as well. Facebook now refers to just the app, nothing else. The different brands are simply products that the company works on, not legal entities in themselves.
> Meta on the other hand is a rebrand of Facebook (the company). There is no holding structure. If you were a Facebook employee, you now work for Meta. Employees at Instagram, Whatsapp, Oculus, shared infra teams, and other minor projects all work for Meta as well.
It doesn't make much difference to your line of reasoning, but at least WhatsApp operates as wholy owned subsidiaries, so there is some amount of holding company; note that the terms of service of Instagram are between users and Meta, but tos for WhatsApp is between users and WhatsApp LLC or WhatsApp Ireland Limited depending on the country code of the registered number. Note: I worked for WhatsApp until 2019; I don't know if WA Ireland is a subsidiary of WA LLC or Meta. There's some other overseas WhatsApp legal entities as well, although I don't have a list.
Because the whole point of the backronym is and always has always been to sound cool. Microsoft was excluded from FAANG for no other reason than it couldn't fit and still make the acronym cool. It's the same reason the acronym has not been FAAAN; FAANG is obviously cooler. As soon as Facebook became Meta, that opened up all kinds of new acronyms and now Microsoft is in and Netflix is out, because MAGMA is the new hotness.
> Out of curiosity, why do some people change "F" to "M" for Facebook to Meta, but leave the "G" there for Google?
Google is still a major tech company, though it is a subsidiary of Alphabet, while Facebook is now a product (not a subsidiary firm) of Meta, as I understand it.
The brief accurately describes Google training ("if you're discussing anything that might possibly be a sensitive issue, make sure to add an attorney to the thread and seek their advice"), as did the parent comment. I've never worked at any other big company but the whole thing did feel like a standard corporate practice.
In the long run, companies are just going to get more and more aggressive with their retention policies, so all this will be irrelevant.
They sure like their email at Google. At most other organizations, sensitive email chains would be open to discovery but would just end by saying "call me" :)
when I started at GOOG, email lived forever. Then at some point, supposedly document discovery was costing n*2 dollars for n accounts and this was some sort of justification for creating a 1-year and 5-year expunge (maybe I have the times off). I ended up losing a fair amount of my long-term correspondence (scientific stuff) because I didn't want to vault things indefinitely.
What amazes me is that the leadership repeatedly puts these sorts of incriminating things in email after being repeatedly told not to.
Operational security is hard and people have a lot on their plate and in their minds, so forget and slip up etc. My mental model is that you are never seeing the totality of communications, but instead imagine people are having communications on several channels (let's say in person, via a phone call, over Whatsapp/SMS/DMs and over email). A conversation in one channel will spill over into another channel. For instance in person or via a phone call will often lead to a "one more thing" moment or followup and the person will follow up over email or dm instead of in-person/voice.
I think of myself as someone who is relatively skeptical of authority, but even so it's hard to disabuse myself of the notion that company leadership is there due to meritocratic reasons.
Just because they're VPs in the biggest company in the world doesn't mean they're actually smarter or more capable than a random person selected from the population at large. Some of them might be, but SOO many people are in their spots due to the accident of their birth and other random processes that happened afterwards.
I wouldn't say any of the SVPs I met at Google during my era (probably 6-7 SVPs) that I thought any of them were dumb. And several were some of the smartest people I've met. But that's the problem with Google: INT 18, WIS 3.
High intellect, low wisdom. The shorthand here is common in games where your character has certain capabilities/stats: intellect (int), strength (str), defence (def), etc.
It's a D&D reference. Starting stats for (human) characters typically range from 3-18, corresponding to the range of 3 six sided dice. "INT 18" means "maximum normal human Intelligence." "WIS 3" means "lowest possible human Wisdom stat," where "Wisdom" roughly translates to "common sense."
> Some of them might be, but SOO many people are in their spots due to the accident of their birth and other random processes that happened afterwards.
Disagree. If you are not smart, you wouldn't last long in those positions.
Maybe getting sued is not the end of the world. Like OK, you talk about it, you argue, and then the judge says this, the business says they'll appeal so that in the news it looks like the plaintiff is a bitch, but they never appeal because they know they'll get gutted clean as a fish appealing their stupid case. Well then they pay.
So, I am a man and if a man becomes a husband, then in the wedding, when signing the marriage contract, that gives the wife the right to sue. Well both have the right to sue. But it doesn't matter, getting married means getting sued, having children means getting sued, hiring employees means getting sued, contracts means getting sued, using IP means getting sued, and you know what? Good, because otherwise men would leave their wives stranded, children would have nothing, promises would be all broken and IP would get stolen. Lawsuits are inescapable, a business I worked for would literally cheat every single employee, partner, and customer, without exception if they could, and the reason they can't is the lawsuit. That's it. You have to pay.
Lawsuits equal paychecks. Complaining, getting a mediator, writing a polite but firm note to the employer, none of that adds up to dogshit. Well when I sued the business, in my conversations with my lawyer, she kept clearing up for me that there's no criminal consequences for suing, it's not a crime. Over and over, it's not a crime to sue, it's not wrong to sue. Funny, a relative I knew who knew the laws perfectly was like "can't we get so-and-so to talk things out, without doing something as violent as going to the Worker's Comp board?" Well you know what suing replaced right? Something violent, in fact. Before lawyers who talked there were champions who fought, and before champions who fought people would just challenge each other and duke it out, to the death sometimes. So collecting on someone meant threatening to beat him up. So it was violent, but so what? Stealing is violent. Everything you have can be stolen, and then you have to (in those days) agree to become a slave (check the thesaurus for what variation on slave you'll be called) all because your possessions were stolen. And get beaten all the time, live on terrible food and hunger, really cold, not just the coercion of slavery but the overlooked poverty of slavery. Or serfdom, in thrall, et cetera. Or going hungry until your reptile brain coerces you into stealing something yourself, and you get caught and sent to jail, coercion and violence. Stealing is violent. Creating the expectation of payment, in particular by A/B testing what makes people think they'll get paid but which your lawyer can argue you can sneak out of, that is violent.
At one point, these guys, the CEO and CFO were talking about a debt the CEO had to the CFO, like how bad could it be you never talk about it, and the CFO said "Well I don't complain about it every day! There are people in this company with rent to pay and people to feed!" Later he had arguments with banks, all kinds of people, the minute the call ended he said "how did I get into this shit!?"
And the guys were in 20% interest debt, they were "young and dumb" for accepting it and "getting raped" like the CFO said. Then they refinanced, and they got the Banco Security "prize" (it's actually called a prize) which meant that bank took on 80% of the risk of the loan (that's what jacks up the interest, the risk), a loan with that bank, the prize is a loan, so they paid I don't know, 6%. Over inflation. And for a business like that that meant the CEO had to use his assets as collateral. The CFO said they needed a miracle.
They would have been in great shape, actually, I created something for them that was worth, conservatively, a million dollars. It was to find optimal pricing to escape competition, so sometimes you would undercut your competitors, the obvious thing to do, but other times, mark up into a price segment where the good could be alone. And do it conforming to theorems, like really bulletproof. I talked to a Stanford professor, and he said that algo would be worth huge sums of money. And I have tons more stuff, most of it much better. Actually nowadays I've had to run extensive experiments because the algos are disgusting, way off the charts but it kept making no sense they could be that good, kept looking for the gotcha and kept thinking I found it and the gotchas were like a mirage, receding as I approached.
All I wanted was $1000. But he actually had to pay. Even if I sold it for $1 million now, which I can't be bothered with because of the my backlog of algos, of which my portfolio online is a small fraction, the only thing I would want would be to pay 99.9% of it to send that $1000 back in time.
Just a judge. So the main purpose of suing them was to make them lose money for cheating me, I didn't have high hopes. You gotta always make people who harm you pay a price, some price, the only thing is it can't be zero. Any time I am defeated I make the opponent's victory a pyrrhic victory. So in that sense I got something valuable out of suing them. Thing is, worker's comp lawyers that defend the employers generally price themselves saying, "if I don't win your case, it'll cost you nothing. You only pay me if I clear your responsibility." That sounds gross but it's good, it means they lost money unconditionally the minute I filed suit.
If they lost the suit, that means they would have paid, and I would have given them the algorithm. Just because they had no sense of honor doesn't mean I didn't either.
I used to be in the email biz, and spoke to attorneys whose retention preferences were 7 days and others who would like to keep email until the heat death of the universe.
I decided it made more sense to move my scientific correspondence to personal gmail (I've not deleted any emails since I signed up in ~2005 or so) and decided not to give a fuck about my google email. if the company lost something useful because it aged it my email, that's google's problem, not mine.
Slack deletes posts older than a few weeks on their free account, and flogs longer term access as the principle feature of their paid account. But I assume some customers feel that Slack is asking them to pay for extended legal liability.
Just FYI, I don't believe Slack deletes them. They just hide them. If I go from a free tier to paid, I believe my message history becomes available again.
>sensitive email chains would be open to discovery but would just end by saying "call me" :)
Probably less common in tech than finance, but recording of 100% of calls for key people is quite common.
In regularly in training sessions about random stuff that end up being A/V recorded since one random dude in the training has above requirement legally.
> "Google has explicitly and repeatedly instructed its employees to shield important business communications from discovery by using false requests for legal advice"
Considering the DoJ is expecting to use these documents for their own case against Google, I don't see how they can argue that it was improper for the employees to get lawyers involved.
You can't just CC legal on pretty much everything to subvert regulatory action. If legal's openly ignoring these CCs, I don't see why attorney/client privilege should necessarily apply.
It doesn't apply. It's not even a clever workaround. It has to be a communication that requests or involves legal advice. If you just forward a public email to your lawyer or CC the attorney on normal business communication it is not privileged.
If you forward such an email to the attorney and ask something like "thoughts? what are the risks here?" that ensuing email chain is privileged. If emails are CC'd to a lawyer to evade discovery and the other side finds out, that's a big problem, even bigger problem if the other side is the feds.
So what if you added an email header that said "P&C, any risks here?" to every email and had a lawyer who set up an auto reply that said "everything looking legal here"?
Then you better pray to whatever deity you hold dear that the judge presiding over this case, and the attorney litigating against you are both absolute idiots, who won't see what you are trying to do.
Generally speaking, judges don't like it when you try to hack the law.
Could we answer this by removing email from the question?
If I go to a lawyer and ask them a question in person, seeking legal advice, even if they don't response and just show me the door, is what I am saying protected? I would think yes.
If I ensure my lawyer is in every single discussion I ever have with anyone else, just to say all my discussions are protected, what would happen? I assume that wouldn't work and there is already a standard on how to handle this for in person meetings.
Shouldn't that same standard be applied to discussions happening through email?
Attorney Client Privilege is abused significantly at tech companies. Remember that the privilege is for a very narrow allowance of cases where the communication is directly between the client and the lawyer with the client requesting legal advice.
Marking an entire group chat with 3000 people [A/C priv] does NOT make all content in the group magically invisible to subpeona.
When I worked at (pre-IBM) Red Hat, I was explicitly told not to do this, as it would undermine legitimate claims of A/C privilege. I also had one of my machines effectively impounded ("you can add data but don't delete any") for over a year because it might have contained information relevant to ongoing litigation. I wasn't even told what litigation that was, though it was almost certainly patent related, and AFAIK nobody ever looked for any information on that machine. I guess some companies just kind of take their responsibilities more seriously than others.
Why on earth would we afford any sort of attorney-client privilege to a business entity? Maybe it's time to have another talk about what exactly the public is getting from this limited liability thing.
A lot of the law depends on narrow specificity. I know that as an individual I want "all the badness" uncovered, but the same rules around discovery apply to corporate entities as individuals. If you want to change this, be wary of the (law of) unintended consequences. A charge should be narrow and specific. If you get a wide warrant, then sure: all the words should be open to view. But if the problem is financial then the IPR mails are not relevant, its the finance mails you need to be shown. Hiding finance mails in IPR mails would be a crime probably. The warrant to finance probably includes these, by implication. Once you are required to present, its fair to incur a cost outcome of "how narrowly can we respond and not be in the wrong" Because we don't punish people for as-yet unknown crimes, in this process.
Attorney-Client privilege goes to asking your Attorney about how narrow the responses can be, and still be legal.
Sort-of, the underlying question is "why did we allow corporate entities to become legal 'persons' in the first place" ?
I am not a lawyer. Find a lawyer to tear this line of reasoning apart, it is probably a low-bar goal.
>I know that as an individual I want "all the badness" uncovered, but the same rules around discovery apply to corporate entities as individuals.
The point is: why? Corporate entities and individuals follow the same rules because we (read: societies, governments, parliaments) decided that would be the case. But is that the best outcome? Should we revisit that decision?
Because that was the most straightforward legal mechanism for achieving all kinds of desirable things like forming any sort of contract at all with a company as a whole rather than one of its employees or holding a company liable for damages resulting from actions of its employees.
sort of, in my mind, like why public limited companies had to emerge, in order to drive some outcomes. unlimited risk investment wasn't attractive even with upsides.
as a child, I understood "corporation" to mean "town council" and entirely wanted the products, which tended to public utility functions. I didn't actually understand corporate in any other sense for years.
Because they are literally a client of the lawyers. We can argue about corporate personhood, but the definition of “attorney client privilege” is very clear. It’s not “attorney human privilege.”
That makes sense to protect the rights of individuals, but why should those same rights be extended to businesses?
Maybe the world would be a better place if companies (and government departments) couldn't protect communications with legal counsel?
I haven't put in enough thought to have a coherent position, but the argument for the current arrangements seem to boil down to "because that is how it always was". Maybe corporate personhood brought along the rights of human personhood and maybe that is a good thing, but it seems to have just happened by default. Maybe we should debate if attorney client privilege is a privilege of clients, or of humans?
The point of attorney client privilege is to encourage the client to get the advice of counsel, who can hopefully guide the client to staying within the law.
That policy goal applies equally well for both individuals and corporations.
I had a law professor say in his opinion the reason there is no mandatory reporter exception for attorney client privilege like there is for doctors and teachers is simply the fact that attorneys write the laws and are protecting each other.
Maybe the multi-billion dollar ones, yes. But the idea of a corporation, at its heart, is to collectively work towards a goal. Crony Capitalism just warped that goal to be about the betterment of the ultrarich. If FAANG or whoever is violating the law, that’s because of the company, not attorney-client privilege. The solution isn’t to kneecap every company by removing their rights, but to enforce the laws on the books against the anti-competitive corporations, and maybe write some new consumer protection ones.
The idea that a corporation is just a collective groups of folks is simply wrong.
Take this situation, nobody is going to poll Google employees and ask if they would prefer if Google waived attorney client privilege in this manner and act accordingly.
It's not just a group of folks making decisions, it's a structure for a small group of folks to manage the asset and labor of others with minimal oversight.
The reason we have laws is not because of how words are defined in a dictionary.
Instead, there are other underlying reasons for why laws exist, and it is not because of some silly argument such as "Well, this is what the the definition of a word is, therefore we are stuck and can't do anything about it!".
And other people are questioning what these other underlying reasons are.
Employees are not clients. Executives are not clients. The company is the client. Employees and executives are there to be sacrificed for the company in issues like this. The go-to defense just about always involves throwing the minions to the wolves to protect the shareholders.
I commented years ago regarding the single-sided nature of hackernews, in which you stepped in to present your existence on this site as a thwart to my argument. To which, you are as transparent now as you've ever been.
> You’re going to need to back that up with a court case
Back what up? That government entities have juridical personhood? Sure, see U.S. v. … well, anyone, actually: being able to be a party to lawsuit (except being the defendant in an in rem suit like civil forfeiture) requires legal personhood.
> especially when the constitution explicitly mentions government for example the 1st amendment
How does the reference to Congress in the first Amendment have any bearing on anything being discussed?
That's why I recommend Fastmail, they have "auto-purge":
> Automatically delete messages after they have been in the folder for a certain amount of time, from 1 day to 1 year. Pinned messages are never deleted.
Funny how US government constantly shields itself from transparency, but demands it, going as far as to have banks report any transactions over $600 to the IRS. Rules for me, but not for thee.
I'm not sure what corporations you're regularly interacting with, but the government is one of the most open bureaucracies I have to deal with. In principle, virtually anything that isn't national security or personal information can be FOIA'd for a nominal fee. If you don't like the treatment, there's usually a variety of processes to appeal or shortcut the bureaucracy, ranging from paperwork to senators.
Compare to Google, where the only way to get a rationale seems to be complaining about it on HN and hoping the PM responds and the only semi-effective feedback mechanism is massive public outcry.
That’s not true at all. A significant amount of deep journalistic reporting is based on information provided by the government in response to FOIA requests.
FOIA can be good but agencies routinely hold up information for years beyond the point of the info still having any use to the original person requesting it
Pedantic but the change was that any P2P service has to give you a form to fill out if you send or receive more than $600 in a calendar year. Since 1970 with the Bank Secrecy Act, banks have had to report ACH/cash/wire deposits over $10,000 to the IRS and are at their own liberty to report anything lower.
> false requests for legal advice," DOJ attorneys wrote in the brief, asking U.S. District Court Judge Amit Mehta to sanction Google and compel the disclosure of more documents.
The US Government is 44% of GDP - with big plans to grow. Google is not the problem here.
After several big companies got their internal email used against them in lawsuits, esp. antitrust lawsuits, they focused heavily on training employees how and when to discuss sensitive topics related to competition, e.g. "don't say exaggerated things like "we're going to kill the competition", etc. And the trainings emphasize being extremely accurate and precise, and not speculating as to legality unless you actually are a corporate counsel. They teach you to take such questions to corporate counsel, e.g. "We're thinking about doing X, what do you think?" on P&C email threads.
I'm sure that some people could overuse this, but again, I didn't see anything suspicious, and I remember seeing one guy get called down for addressing email to corporate counsel as P&C and not actually asking a question.
I'm sure in every org there are weasels who would try to misuse this, but I'd have to see more evidence before I would believe that this was cultural at Google.