I suppose consider yourself lucky that you haven't yet been inconvenienced by an issue completely within the airline's control and been offered $50 in funny money even though you had to pay out of pocket for your dingy hotel and transport in/out of the airport at 11pm/6am the next morning. And no, they won't reimburse that, because you could have slept on the terminal floor for free

I on the other hand will never understand anyone extending any grace to an airline. They are constantly testing just how poor of an experience they can deliver to their customers and stay in business.

For starters, they could have offered more than $100 to OP here

Perhaps they want the airlines to do a better job maintaining their fleet such that the airline is capable of meeting its obligations with respect to arriving on time without crashing.

Do proper maintenance so they don’t have engine issues, and carry insurance to compensate passengers in case they do?

Yeah I don't get what the airline did wrong there.

A decent reimbursement.

Not parents fault the airline flies around with defective hardware.

Usually this is regulated by law.

Uhm, properly compensate passengers that have concrete harm to point to instead of offering a laughable alibi payment? What I'll never understand is why some neoliberal schools of thought defend conglomerates like they're their children... This company has enough money to compensate for fuck-ups, at least it should have. Every insurance on the planet can do the simple math involved for this risk assesment.

In the United States legal system, when you take a plea because the Government was threaning you with a 20 year sentence the judge requires in court and under oath for you to state that you were not coerced/pressured in any way to take the plea (because pleas wouldn't be valid if the Government pressured you into them and into signing away your rights).

Before you push back at your work you have to consider possible loss of work and therefore housing/medical insurance/food.

The entire US system is designed around baked in blackmail of the individual into conforming. Not surprising at all to see the airline turn the no fly list into that pressure, it's the main technique we know to motivate people in the USA.

It is being scrutinized. The sale is overseen by the courts. Also, the media is scrutinizing. Also, PII has already been addressed by the court, from the article: "If you’ve flown Spirit and worry that Google will soon know about a testy conversation you had with the airline’s call center, you’re being told not to worry. The court filing says the data was deidentified before being put on sale and Google has promised to scrub any PII it finds in the trove."

Usually when you work a job you sign a little thing that says "yeah you own everything I produce for you, no matter how small".

Which is, of course, ridiculous, and follows the trend of absurdist contract law wrangling in corporations. Similar to non-competes and NDAs.

It makes sense to some degree, but the fact that semi-private conversations are included in that makes no sense. These have little to no business purpose.

Makes one appreciate living in place with sufficient constitutional protections against this sort of stuff. Even for work stuff selling this info wouldn't fly in some parts of the world.

Not for nothing, but you have probably already consented. Typically user agreements allow for this kind of sale if you’ve authorized use and processing but YMMV.

Since it’s work communications, consent was already given.

When you join a company, you typically sign an agreement that talks about how the company owns all your output. Thumbs upping a Teams message is work output and they own it.

Every email sent and received. Every keystroke. Etc etc etc.

If you don’t want your employer to log and sell it, start your own company. Or use a personal device. I do the latter.

The party owning this data (Spirit Airlines) is consenting to the sale. Employees and customers of Spirit consented when they started employment and did business with Spirit, respectively.

> How can such data be sold without the consent of all involved parties?

In the US, whoever owns the computer owns the data on it. Courts have routinely ruled that you have no say in what other people collect about you. The goal of bankruptcy courts is to minimize the losses of the creditors. And bankruptcy courts routinely rewrite contracts except where statute prevents it (like mortgages).

In the EU, you own the data about yourself. A lot of people utterly hate GDPR, but that's reason that you own the data about yourself.

The OP article is also wrong on multiple counts. "Customer behavior" data like call recordings and email addresses/activity is specifically not included in Google's purchase. See page 18 of the court document they link, the "Google's Data Purchase Request" column on the right lists what is and isn't included.

I'd be amused if a sub-sub-agent organically decided to do it anyway - even if just for a notable figure that an LLM can identify with its weights alone. What are the controls? Who's going to keep Google accountable? Hah.

I know Meta it's not Google, but it's worth remembering that these promises haven't had a great measure of success in the past:

> Facebook has been fined €110m (£94m) by the EU for providing misleading information about its 2014 takeover of WhatsApp. (...) When Facebook took over the WhatsApp messaging service in 2014, it told the commission it would not be able to match user accounts on both platforms, but went on to do exactly that.

https://www.theguardian.com/business/2017/may/18/facebook-fi...

But did they agree to not re-sell the data to someone else and let them re-identify users?

I would not be offended by the blood donation thing. They generalize based on administrative regions (Amazonas in this case) and not whether you visited a big developed city or not.

I had a similar blood donation issue for visiting a particular island in the Philippines, and could not donate for 4 months.

That was only 3 myths. Hardly a brazilian

I'm sitting here chuckling because you felt the need to post this.

Your points are valid. And there probably are plenty of Americans who needed the correction. But still.

The sloth was returned to his owner and I did tip him. That kid is probably still prowling the Amazon (as an adult now), looking for sucker tourists like me.

I love the irony of you checking them out for more information in response to a comment of them being worried about who reads their data. Nothing wrong with it, just make me chuckle

Suckers pay companies for expensive SSL monitoring products, smart people just post to HN.

Doh, thanks!

> the public’s overwhelming appetite for “FREE”

In the early pioneering days of the commercialized Internet, email addresses were inextricably linked to your ISP. You paid for an ISP connection and you got an email box, with MTA and MUA service to match. You were reluctant to switch or leave your ISP, because that also meant leaving behind your email address. Of course, a minority of nerds got around this with their own domains, etc.

However, it seems that Google, AOL, Yahoo!, Hotmail, and other players got into providing free email services and eventually grew into giants that supplanted every other MTA service. This was not an accident and it was not merely our appetite for “FREE” but it was a very calculated plan by the industry. Those early ISPs did not have a business model that admitted monetizing our private data; they seemed to have a more respectful attitude for keeping it private. Perhaps that was a result of being telecommunications-based companies, rather than advertising or entertainment.

If a service like email provides such endless treasure troves of personal data, including a social graph and glimpses into our private daily lives, why not provide it for free and monetize opportunistically on the data itself? The free email services killed the paid services, not by being better or cheaper, but by being bigger, centralized, and more persistent. The main reason I signed up for Yahoo! was because it would be an utterly stable presence. I saw my parents and others so hopelessly attached to an ISP-based email, but I couldn't end up like that.

After streaks of losing my home and non-payment of bills and moving around over decades, the most stable point of contact for me has been a "free" email address.

There are different no fly lists. Airlines maintain their own list of people they’ve banned from their planes.

There is "the" no-fly list maintained by the government, which is nominally for people who are too dangerous to be allowed on an airplane, yet not dangerous enough to charge criminally.

But each airline also maintains their own internal no-fly list for people they prefer to no longer have as customers, for whatever reason. You might end up on this for some abuse of the system that doesn't pose any sort of safety risk, so the government doesn't care, but the airline doesn't like. For example, excessively doing hidden-city ticketing (where you book a flight with a connection, then skip the second leg of the trip, because weird pricing rules make it cheaper than just booking a ticket to the connecting city) can get you banned from the airline, but the government would be completely uninterested.

unfortunately "de-identified" data is typically re-identified quite trivially. so i guess we just hope google keeps its promise, and is competent in its scrubbing.

> Google has promised to

This part is worrying.

See the last 3 sentences of GP's post

I have a bridge to sell you.

> I don't understand why a CS rep would get this invested to the point of wanting to cause you real harm

This is the kind of power tripping that easily corrupts people, especially those who don't have much power outside of work. The US national security apparatus is vast and powerful. Many people get giddy at the thought of inflicting punishment on those who "deserve" it. Act rude to a fast food worker, and you get spit in your food. Lots of people cheer the worker who spit in the food of a customer who is merely rude, impatient, or demanding. Or is guilty of being a cop, politician, rich, etc.

I can easily imagine the kind of CS rep who would delight in putting a customer who didn't just go along with things and spoke up for themselves... thats 'rude' and 'disrespectful' to some. Police are the most notorious for this kind of petty "you will respect my authority" but it is in every industry.

HN / hacker culture celebrates this in the 'Bastard Operator from Hell' [0], the sysop who will ruin your work and life with their IT wizardry, no matter if you're an intern or CEO, if they do not feel respected. Or if you interrupt their gaming with your support call.

[0] https://www.alnet.org/bofh/Bastard.html

I don't understand why a CS rep would get this invested to the point of wanting to cause you real harm.

Two possibilities come to mind... 1 - The CS rep has been instructed to do this. Scary, but corporate leaders can be assholes and wield lots of power within their orgs, so doesn't seem completely unlikely to me.

2 - The CS was just a dick.

Frankly, given the behavior of various SuperMegaCorps over the past few decades, I'm going with #1.

That's because they're seemingly perfunctory. What's worse than no law is a bad one that doesn't do anything but make you feel like something is actually being done.

That’s a bit unfair. They also catered to broke people who weren’t crazy or argumentative. On a planeload of 100 people, there must have been 60 of them at least who were just broke. Or on some routes, Spirit was the only carrier with a direct flight so they were just normal people who wanted to get to a certain place efficiently (Hi, that’s me, I flew them for this reason). Keep in mind Spirit had an excellent safety record, too, so airline choice in this case was primarily a question of having luxuries or not.

Not trying to be snarky, and perhaps it wasn't well stated, but the last paragraph I said I'm concerned about identification via my writing style. If they have my emails, they would have my writing style. It doesn't have to be tied to PII there, they can cross reference it with my blog. I'm speculating because I read that you can identify people by a few sentences of their writing.

"Deidentification" seems really murky and imprecise at best.

You have to trust that this really "deidentifies". Time and time again it was shown, that the measures taken were not enough to anonymize.

E.g. the parent wrote that he fears, he could be identified by his writing style, which is totally plausible. How would you "deidentify" this?

Even before LLMs there were multiple papers written about ways to to reidentify people with ML and other statistical analysis. It is probably now even more trivial especially if you are Google.

No such animal.

I have a bridge for sale, hardly seen use, pay me ${money} and you can collect it in New York City. Interested?

Your comment reminded me of a funny interaction I had a week or so ago.

I got a call that started with the usual automated message, "This call is being recorded." After the person joined, I pushed the record button on my iPhone, "This call is being recorded."

They were surprised and asked why I'm recording.

I said, "You're recording, so I'm recording too."

The rep insisted that the company doesn't like this but that they will continue with the call anyway.

Anyway, I wish people took this stuff much more seriously. It always seems to boil down to, "I don't have anything to hide" type of conversations and I've never managed to convince anyone that privacy as a concept isn't about having something to hide.

Some aspects of privacy policies don't survive bankruptcy, I'd wager usage consent does not either. IANAL.

I don't think there's any legal weight to the purpose of a recording unless you entered into a contract that has a clause to that effect. They have to tell you that the call is being recorded because of wiretapping laws. They stick "for training purposes" on there just to soften the language and reassure you that they have a good, non-nefarious reason to record. It doesn't actually limit what they're allowed to do with the recording.

You’re years too late.

> I made the counter argument that watermarks would find their ways into everything and eventually be bound to attestation.

Yup.

Elsewhere in another front page thread today: "oh but apps blocking screenshots because of 'sensitive content' can be bypassed by taking a photo of your screen with another phone".

Any tech-savvy person with two brain cells reading this and that: "gee, I wonder if the same magic imperceptible watermark that survives multiple rounds of cropping and printing and scanning, that's used to tag AI-generated content, could also be used to tag sensitive data, or ads, or which app is rendering it on screen, and then the camera app could refuse photographing it...".

I don't know why people don't see that AI watermarks are DRM, and DRM is universal, and there are many clients...

I think I'm starting to feel like I'd prefer living through WWIII as opposed to whatever it is we are living through now.

Finally we know how they assign people to either that A Ark or the B Ark.

LLMs aren't a database. They're an attempt at brute-forcing an artificial mind. The who and what aren't really interesting there, it'll forget most of such details anyway. What matters is the patterns visible in the text at various scales. How people write. Why they write. To whom they write, in response to what. How does e-mails about mistakes correlate with PDFs they're referring to. How people work with ticketing systems - like how, actually, a ticket plays out. The jargon, the acronyms, the vibes, the causal links. It's all in there, and it's another slice through the set of things humans do, to be combined with other slices already in the training data, and enriching the whole.

(Something something we will add your distinctiveness to our own, you will be assimilated, ...)

(Hell, the fact that it's all from one org would make it a great dataset to have in the open for sociological studies. I bet that today, aided by LLMs to sift through it, you could use it to map how information flows through a large org - how incident on the floor travels through time and layers of management until it reaches the C-suite, what of it survives, how it gets reacted to, how the reactions flow down...)

> How does this have value? Is any and every sentence in an e-mail considered 'fact' and thus to be fed into the AI?

I have a hunch what this is for. AI companies want to make bigger inroads into nontechnical work settings. But LLM progress outside of fields where verifiable rewards for RL post-training can be synthetically generated (coding, math) has been pretty flat. Buying years of operational data from a company like an airline could be used to reconstruct long-horizon task trajectories in areas like customer service or marketing.

All the tech companies are trying to find data the others don’t have access to. That is one way they try to edge out the competition.

A major selling point of AI chat bots is for customer service automation. A clean dataset like this is a huge find. I'm not sure what you mean by "facts" or "gold nuggets". Training data doesn't need to be factual.

> Is any and every sentence in an e-mail considered 'fact' and thus to be fed into the AI?

If it comes with the context, yes. More data the better. Someone considered it served some purpose at some point. Thus it contains, no matter how tiny, a sliver of information.

It will have a different writing style from the average blog post. Maybe they just want to train an AI that sounds less like an AI. Or one that speaks in vapid management style.

This data shows exactly how a huge company of thousands of employees works and coordinates.

The perfect data to train an agent swarm on how to run a company.

Maybe it's innefficient and inane, but it's how you start.

The first LLMs, GPT-1, 2, were trained on complete garbage, the average document from the common crawl is random non-sense, yet they worked, and now we can use LLMs to filter the data for the next training run.

Truth is even weirder. plenty of growth is possible but modern liberal democracy requires that all progress must be contingent on the production of enormous amounts of text that nobody would ever read.

3 years ago, the Wall Street journal covered a company trying to use AI to generate documents required for the approval of new nuclear reactor reactor designs, which sounds dangerous, until you get to the point where they'd cite 2 million pages as necessary for a typical application. [1] I don't need to explain why no individual or institution could read that, much less examine it in detail. I remember a rather funny question I found in a comment to that story - "How many pages of those 2 million could contain pornographic images before anyone notices?"

It's obvious why companies are so desperate for training data - a text generator of sufficient quality is more conductive to the growth of the nuclear industry than any scientific breakthrough in nuclear physics. (And of course, if you want to prevent the development of a nuclear reactor by your competitors, being able to produce millions of pages of high-quality objections will do the trick.)

And it's not just nuclear power. When it comes to stuff like building rail lines, apartments or power plants (both conventional and renewable), you'd find that the main bottleneck is the necessity to produce documents. And of course, many documents can be subject to judicial review - a process that consumes even more text.

[1] https://www.wsj.com/tech/ai/microsoft-targets-nuclear-to-pow...

Any kind of fuel for giving active investors that FOMO tingle which then forces the steamroll of index funds to blindly follow.

I guess the appropriation "any sufficiently advanced stock market is indistinguishable from entertainment" doesn't quite stop at equating the trade floor with a casino. At some point, entertainment also becomes the modus operandi of corporations.

You wouldn't want to hard-wire the deidentification company's name into the contract between Google and Spirit. Otherwise, if the deident-company happens to go bankrupt or otherwise be unable to do the work then you'd need to re-do the Google-Spirit contract, which would be a massive pain. And you don't want to make "we can sign this with Spirit now" be dependent on "we have first signed the deal with the deident-company". So I think it's reasonable that the contract says "one or more third parties acceptable to or designated by Buyer" rather than being specific here.

It'll be a little startup from San Francisco called "El Goog".

> Ah, but Google promised to remove PII they found in this deidentified dataset, so worry not.

… and even if someone can prove that they didn't, the only consequences will be a teeny-tiny slap on the wrists.

This kind of stuff needs to come with promises to pay the P in the PII big bucks if the I is indeed I.

I basically agree, but I'd also say: Every Gmail user has already accepted such a promise as sufficient.

Ah, trust me bro :)

Sure it can be scraped, but it's so much easier if you just get a single database with everything in it.