Unlawful processing of data is absolutely not “opt out” by default, and simply assuming that “legitimate interest” applies does not mean that you can allow on opt out.
Here's just a quick reminder that if you're a LinkedIn user based in an EU member state, and if you do not believe that “legitimate interest” applies, you can file a complaint with your national supervisory authority (SA)/data protection authority (DPA); see https://en.wikipedia.org/wiki/National_data_protection_autho...
I'm not in the EU but I am in Europe. It was on for me.
From my reading of the article, this is Linkedin (an American company) that has decided to collect data about Europeans - the EU is not involved. Linkedin has decided that making it opt-out is covered by "legitimate interests". So I guess the real issue here is that the rest of you just don't get the option - you are included anyway.
Making this feature opt-out is a clear violation of the GDPR. Linkedin claims they have a "legitimate interest" in collecting this data for AI training without consent, but this argument is laughable.
Even as written in the regs "legitimate interest" shouts "we are your preference not to be stalked by advertisers or provide us with free training material, but fuck you and your silly little preferences we want to anyway so here have another hoop to jump through", and it is stretched even further from there.
I noticed some companies ask for a LI profile, and sometimes this field is mandatory. So a few years ago I created an empty LI profile, just saying "consultant" started 20 years ago.
Strangely enough, it wasn't a problem at all. They'd still call me, interview me and make an offer. So I believe people overestimate the importance of LI, at least for some cases.
This is an important issue from me because it's the only so-called "social" site that seems somewhat enforced by the system we're living in, so it has been abused in various ways as people believe they have no choice but to share the details related to their education, past and present employment with the general public.
Facebook for me is much more enforced. A lot of mom&pop stores publish their opening hours on facebook only. Clubs communicate with their members via facebook, so you can't subscribe to the yearly bbq etc without it. Small local sales, events,.... All small things, but death by a thoudand papercuts.
I held out on facebook since last year, but have given in. My wife is my only friend, even if facebook is pushing a never ending desperate barrage of youmightknow mails.
I've definitely seen my coworkers discriminate against interview panel candidates who don't have fleshed out linkedins. They look "sketchy" is what they'll say, one guy routinely compares the dates on the CV against linkedin profile dates and if they don't match flag the person as honesty issues. Also they'll bring up if they don't have enough skill endorsements in core competencies for the roles. I had no skill endorsements so after hearing my coworkers talk about how they check for that I got a few! So aggravating but it happens and people definitely rule out candidates over their lack of visibility on linkedin.
"Should you choose to inscribe the lexical designation 'DOMINATE' within the commentary section, I shall subsequently furnish you with my proprietary methodology for accumulating substantial financial wealth in the millions."
"Comment the word DOMINATE and I'll share my secret method for making millions of dollars."
"yo if u type DOMINATE in comments ill tell u how 2 get mad rich like millions n stuff fr fr no cap"
"Congratulations on tiur furst step towards financial independence. Next,you may sign up on our online pig-butchering trading platform, and make sure you buy and hold our new Dominus (INUS) shitcoin. We also have drops daily, make sure you hodl, and it will GO TO THE MOON soon. If you invest USD15000, you'll be a millionaire by 2029"
Seriously, I cannot fathom any real benefit that LinkedIn would accrue from training on the pathologically upbeat pablum posted by its users and bots acting on behalf of its users.
I have at least one contact who focuses exclusively on pushing his libertarian views with a side of particularly conservative talking points. Honestly I'm curious to find out how the mix of that and what you described looks.
I know it's a rather privileged thing to say, but I just passed up a job opportunity because I don't want to get back to 30 paid vacation days. Once you get used to 47 (took a voluntary paycut of 5 percent for 12 extra days), where you get so many side projects done because you have the time, it's hard to go back to the normal 28-35. So I have no idea why the US rarely goes above 20 or whatever they have.
I dropped to a four-day week (with 20% pay drop so in the same ballpark as your 5% for 12 days off) for stress reasons “temporarily” almost exactly two years ago. The burnout problem is finally looking like it might permanently recede (family health matters are much more stable, some work matters have been addressed, I've sorted a few personal things out, some external problems have been fixed or just told to fuck off, etc.) but I think I'm going to stick with the four days week as long as I can.
The extra money would be very useful to me right now, but I'm surviving without it and the three-day weekend is really working for me. I'm even finding time to relearn a second language on case the UK follows the US too far down the full-fash route and I feel the need to get out of dodge [I refuse to be the sort of Brit who rocks up everywhere expecting everyone to speak English!].
In Denmark there's 4 weeks for the mother (right after giving birth) and then 24 weeks for the mother and their partner each, i.e., a total of 52 weeks. It's also possible to transfer weeks between the parents.
These days LinkedIn is a cesspool of AI-generated feel-good “inspirational” content anyway… I don’t see anything good coming out of training models from its user data.
"I am so grateful for the opportunity to help you with a recipe for guacamole today. It is a humbling, educational experience and I am looking forward to collaborating on with you. Please be aware that I am still open to an interesting occupation in my field and I welcome any referrals."
Wish I was like you. I have NOT touched them since MS acquisition. And now I don't have access to the email linked with the account anymore. Guess they are just gonna train the *it out of my data. Really really regret signing up for that service.
> The company says it will rely on “legitimate interests” as its legal basis and will offer an opt-out so members can refuse use of their data for training
"Legitimate interest" is a very specific term in context of GDPR. Not a lawyer, but have been looking into it previously, and I doubt "we want to feed data to our AI so we can make more money" passes the Legitimate Interest Assesment (LIA) test.
That looks like it would be easy to argue that it passes (claiming "makes the platform better for everyone", "not achievable without using the data", "the data is data that the people share voluntarily on the platform and isn't sensitive", "they're customers, we e-mailed them and they could opt out if they cared", "we expect this to have no impact on the individuals" (until the AI starts regurgitating sensitive details, but that's an "oops" for later), and "we are offering an opt-out even though we wouldn't have to" (claimed despite the lawyer strongly urging an opt-out, otherwise they wouldn't have even offered that).
They could argue whatever they like -- whether that'd be defensible if a probe is launched (and LinkedIn / Microsoft is big enough target for this) is another matter.
GDPR allows processing based either on consent (which doesn't need to be "written" but does need to be explicit and informed) or legitimate interest (or some other reasons that tend to be irrelevant for this kind of thing).
Legitimate interest does NOT require consent, is murky, and thus often gets used to justify things that should not exist under GDPR but the most likely consequence is that the company gets to do it for 3+ years before being told "no, you can't do that anymore"...
The GDPR is about personal data though. And content your produce is not by nature personal data "in abstract".
That content could contain personal data (such as when including it in your post), but that's an exception rather than a norm. And if we'd be following exceptions, even crawling websites could be illegal under the GDPR.
What do you expect from a company that bypasses security models on mobile devices to intercept emails? Of course it's opt-out... I'm guessing US users don't get any option.
The part that really gets me is that opting out doesn't affect models already trained on my data. It kinda feels like closing the barn door after the horse has already bolted.
I am flabbergasted that Microsoft thinks they need just a more data. Surely we are way past diminishing returns on what some phony, inspirational work memes can deliver.
Quality of LinkedIn feeds is the lowest of all social network sites I occasionally view. Instagram at least has attractive people and puppies.
I don't know if it's AI slop or people genuinely baring their souls and revealing corpo-BS pseudo-deep thoughts, with a drop of grifters pushing their BS "Uber but for nailcare on Blockchain" whitepapers, either way I don't see any value you could extract from it.
What is it with American big tech and blatant disregard for EU privacy laws.
“Legitimate interest” is being used here to skirt around the need for consent (I.e. explicit opt-in) specifically because they know users will not give consent (otherwise they could easily ask them). That is an immediate red flag when it comes to “balancing” the rights and needs of the user vs the data processors “legitimate interest”.
I would like to think this will go towards the upper bounds (up to 4% of global revenue), as they are doing so wilfully and to a large number of users. This is exactly the showcase the EU needs to make an example of how not to abuse the legitimate interest exceptions.
Its a really stupid time to test the EU over something that cant have much net value to LinkedIn - a 1bn EUR fine (approx 4% of LinkedIns revenue) is well within the current internal "cap" the EU have reached with issuance of fines.
But who knows. The EU is much more patient than I, and prefers boiling frogs to scalding them.
I'm not sure how they are going to make money from this. LinkedIn has to be one of the most worthless dataset in existence for training an AI, half of it already looks generated by a LLM itself and the rest is low value content.
Even if it was free I wouldn't include it in AI training.
You can opt out here: https://www.linkedin.com/mypreferences/m/settings/data-for-a...
Thanks. This should have been _opt in_, and not opt out, regardless of any LinkedIn newsletter emails that automatically go to the garbage bin.
Surprised the EU did not make "opt in" the mandatory default for something like this.
Unlawful processing of data is absolutely not “opt out” by default, and simply assuming that “legitimate interest” applies does not mean that you can allow on opt out.
Here's just a quick reminder that if you're a LinkedIn user based in an EU member state, and if you do not believe that “legitimate interest” applies, you can file a complaint with your national supervisory authority (SA)/data protection authority (DPA); see https://en.wikipedia.org/wiki/National_data_protection_autho...
Thanks, complaint sent.
Seconded.
I'm not in the EU but I am in Europe. It was on for me.
From my reading of the article, this is Linkedin (an American company) that has decided to collect data about Europeans - the EU is not involved. Linkedin has decided that making it opt-out is covered by "legitimate interests". So I guess the real issue here is that the rest of you just don't get the option - you are included anyway.
Making this feature opt-out is a clear violation of the GDPR. Linkedin claims they have a "legitimate interest" in collecting this data for AI training without consent, but this argument is laughable.
"Legitimate interest" is abused on an absurd scale
Even as written in the regs "legitimate interest" shouts "we are your preference not to be stalked by advertisers or provide us with free training material, but fuck you and your silly little preferences we want to anyway so here have another hoop to jump through", and it is stretched even further from there.
I wish it did, but I am not surprised it does not. The EU remains a fundamentally pro-business organization at its core.
Legislation for the AI slop will take few years probably
Why would they make it opt in? Nobody would sign up for it in that case!
I am not in the EU and it was disabled by default
Thanks. I hate to need to use linkedIn to find a job. This was the cherry on top of the cake.
I noticed some companies ask for a LI profile, and sometimes this field is mandatory. So a few years ago I created an empty LI profile, just saying "consultant" started 20 years ago.
Strangely enough, it wasn't a problem at all. They'd still call me, interview me and make an offer. So I believe people overestimate the importance of LI, at least for some cases.
This is an important issue from me because it's the only so-called "social" site that seems somewhat enforced by the system we're living in, so it has been abused in various ways as people believe they have no choice but to share the details related to their education, past and present employment with the general public.
Facebook for me is much more enforced. A lot of mom&pop stores publish their opening hours on facebook only. Clubs communicate with their members via facebook, so you can't subscribe to the yearly bbq etc without it. Small local sales, events,.... All small things, but death by a thoudand papercuts.
I held out on facebook since last year, but have given in. My wife is my only friend, even if facebook is pushing a never ending desperate barrage of youmightknow mails.
When I applied for my last job[0], they asked for a LI profile - I explained I didn't have one and why (data privacy concerns) on the application.
0. now my current job!
I've definitely seen my coworkers discriminate against interview panel candidates who don't have fleshed out linkedins. They look "sketchy" is what they'll say, one guy routinely compares the dates on the CV against linkedin profile dates and if they don't match flag the person as honesty issues. Also they'll bring up if they don't have enough skill endorsements in core competencies for the roles. I had no skill endorsements so after hearing my coworkers talk about how they check for that I got a few! So aggravating but it happens and people definitely rule out candidates over their lack of visibility on linkedin.
Except it was never a cake to begin with.
And it's not a cherry. It's undigested sweetcorn.
Thank you.
Training models on so much AI-generated content will certainly lead to diminishing quality.
As far as I can tell, LinkedIn's content is already 99% LLM generated posts.
The resulting models might be a terrible hybrid distillation of GPT5 and Claude with a strong preference for hustle culture & banal parables.
Evolution of LinkedIn posts over time...
"Should you choose to inscribe the lexical designation 'DOMINATE' within the commentary section, I shall subsequently furnish you with my proprietary methodology for accumulating substantial financial wealth in the millions."
"Comment the word DOMINATE and I'll share my secret method for making millions of dollars."
"yo if u type DOMINATE in comments ill tell u how 2 get mad rich like millions n stuff fr fr no cap"
Idiocracy movie as a template? :)
DOMINATE
"Congratulations on tiur furst step towards financial independence. Next,you may sign up on our online pig-butchering trading platform, and make sure you buy and hold our new Dominus (INUS) shitcoin. We also have drops daily, make sure you hodl, and it will GO TO THE MOON soon. If you invest USD15000, you'll be a millionaire by 2029"
Seriously, I cannot fathom any real benefit that LinkedIn would accrue from training on the pathologically upbeat pablum posted by its users and bots acting on behalf of its users.
I have at least one contact who focuses exclusively on pushing his libertarian views with a side of particularly conservative talking points. Honestly I'm curious to find out how the mix of that and what you described looks.
Frankly, at this point, I’m here for an AI feedback loop leading into model collapse.
Let it burn.
So, soon we will see AI generated job descriptions for American companies offering 6-month paternity leave and 30 days/year paid vacations?
I know it's a rather privileged thing to say, but I just passed up a job opportunity because I don't want to get back to 30 paid vacation days. Once you get used to 47 (took a voluntary paycut of 5 percent for 12 extra days), where you get so many side projects done because you have the time, it's hard to go back to the normal 28-35. So I have no idea why the US rarely goes above 20 or whatever they have.
I dropped to a four-day week (with 20% pay drop so in the same ballpark as your 5% for 12 days off) for stress reasons “temporarily” almost exactly two years ago. The burnout problem is finally looking like it might permanently recede (family health matters are much more stable, some work matters have been addressed, I've sorted a few personal things out, some external problems have been fixed or just told to fuck off, etc.) but I think I'm going to stick with the four days week as long as I can.
The extra money would be very useful to me right now, but I'm surviving without it and the three-day weekend is really working for me. I'm even finding time to relearn a second language on case the UK follows the US too far down the full-fash route and I feel the need to get out of dodge [I refuse to be the sort of Brit who rocks up everywhere expecting everyone to speak English!].
Off topic, but less than a decade ago paternity leave in the Netherlands was just two days, and it's still nowhere near six months, unfortunately.
In Denmark there's 4 weeks for the mother (right after giving birth) and then 24 weeks for the mother and their partner each, i.e., a total of 52 weeks. It's also possible to transfer weeks between the parents.
The companies want the FREEDOM to mistreat employees FREELY.
But the moment when recruit refuses offered wage, they will start crying that nobody wants to work anymore.
Perhaps this will fix the work-life balance for American people in the long term.
I've said this elsewhere: The EU should crack down on this. Opt-out is not acceptable behavior.
These days LinkedIn is a cesspool of AI-generated feel-good “inspirational” content anyway… I don’t see anything good coming out of training models from its user data.
"I am so grateful for the opportunity to help you with a recipe for guacamole today. It is a humbling, educational experience and I am looking forward to collaborating on with you. Please be aware that I am still open to an interesting occupation in my field and I welcome any referrals."
I wish them the best of luck with that.
Short of 4chan it's hard to think of a worse dataset. Do you really want a model trained on corporate thought leader gibberish?
yes, so you can produce more corporate thought leader gibberish.
Can barely contain my excitement
This is an incredibly dangerous development - when Ai learns the grind and hustle it is going to accelerate the timeline to Judgement day.
Only the people with the right grindset that do a 4am ice face bath will survive.
Glad I never had an account on LinkedIn
Wish I was like you. I have NOT touched them since MS acquisition. And now I don't have access to the email linked with the account anymore. Guess they are just gonna train the *it out of my data. Really really regret signing up for that service.
I can get you an invite if you want
> The company says it will rely on “legitimate interests” as its legal basis and will offer an opt-out so members can refuse use of their data for training
"Legitimate interest" is a very specific term in context of GDPR. Not a lawyer, but have been looking into it previously, and I doubt "we want to feed data to our AI so we can make more money" passes the Legitimate Interest Assesment (LIA) test.
Here's an example of a test that must pass (sorry, docx, but way better than a random explainer): https://ico.org.uk/media2/for-organisations/forms/2258435/gd...
That looks like it would be easy to argue that it passes (claiming "makes the platform better for everyone", "not achievable without using the data", "the data is data that the people share voluntarily on the platform and isn't sensitive", "they're customers, we e-mailed them and they could opt out if they cared", "we expect this to have no impact on the individuals" (until the AI starts regurgitating sensitive details, but that's an "oops" for later), and "we are offering an opt-out even though we wouldn't have to" (claimed despite the lawyer strongly urging an opt-out, otherwise they wouldn't have even offered that).
They could argue whatever they like -- whether that'd be defensible if a probe is launched (and LinkedIn / Microsoft is big enough target for this) is another matter.
GDPR doesn't allow "they knew and they could have opted out if they cared". You need explicit written consent.
GDPR allows processing based either on consent (which doesn't need to be "written" but does need to be explicit and informed) or legitimate interest (or some other reasons that tend to be irrelevant for this kind of thing).
Legitimate interest does NOT require consent, is murky, and thus often gets used to justify things that should not exist under GDPR but the most likely consequence is that the company gets to do it for 3+ years before being told "no, you can't do that anymore"...
but they will absolutely want to sell it to 3rd parties, else what's the point ?
The GDPR is about personal data though. And content your produce is not by nature personal data "in abstract".
That content could contain personal data (such as when including it in your post), but that's an exception rather than a norm. And if we'd be following exceptions, even crawling websites could be illegal under the GDPR.
What do you expect from a company that bypasses security models on mobile devices to intercept emails? Of course it's opt-out... I'm guessing US users don't get any option.
The part that really gets me is that opting out doesn't affect models already trained on my data. It kinda feels like closing the barn door after the horse has already bolted.
Why the hell does LinkedIn need to train AI models?
i'm excited to find out what AI will learn about b2b sales from all the cringe posts.
I am flabbergasted that Microsoft thinks they need just a more data. Surely we are way past diminishing returns on what some phony, inspirational work memes can deliver.
Why take the legal risk?
Quality of LinkedIn feeds is the lowest of all social network sites I occasionally view. Instagram at least has attractive people and puppies.
I don't know if it's AI slop or people genuinely baring their souls and revealing corpo-BS pseudo-deep thoughts, with a drop of grifters pushing their BS "Uber but for nailcare on Blockchain" whitepapers, either way I don't see any value you could extract from it.
"Uber but for nailcare on Blockchain". I'm laughing but I'm not sure if I should be crying. Thanks for the laugh!
What is it with American big tech and blatant disregard for EU privacy laws.
“Legitimate interest” is being used here to skirt around the need for consent (I.e. explicit opt-in) specifically because they know users will not give consent (otherwise they could easily ask them). That is an immediate red flag when it comes to “balancing” the rights and needs of the user vs the data processors “legitimate interest”.
i.e. big fine incoming for LinkedIn.
But will the "big fine" be more than the money LinkedIn will make by doing this?
I would like to think this will go towards the upper bounds (up to 4% of global revenue), as they are doing so wilfully and to a large number of users. This is exactly the showcase the EU needs to make an example of how not to abuse the legitimate interest exceptions.
Its a really stupid time to test the EU over something that cant have much net value to LinkedIn - a 1bn EUR fine (approx 4% of LinkedIns revenue) is well within the current internal "cap" the EU have reached with issuance of fines.
But who knows. The EU is much more patient than I, and prefers boiling frogs to scalding them.
I'm not sure how they are going to make money from this. LinkedIn has to be one of the most worthless dataset in existence for training an AI, half of it already looks generated by a LLM itself and the rest is low value content.
Even if it was free I wouldn't include it in AI training.
Maybe it will finally start recommending better cheese. My feed is so boring.
good insight
Commenting for reach
thank you so much guys
Are you open for exciting new opportunities in your area?
I wonder what this will do for enterprise sales posting
There's a film about this, Multiplicity.
LinkedIn is becoming the other Facebook...
I don't think they have ever pretended not to be a platform that gathers data about users.
link to opt out: https://www.linkedin.com/mypreferences/d/settings/data-for-a...
Great! Now AI will be optimized for humble-bragging about thought leadership and congratulating connections on work anniversaries they forgot about.
It is infuriating that this thing is opt-out instead of opt-in, irregardless of advanced notice to configure it.
Seeing that LinkedIn these days is 90% AI-slop, wouldn't this mean that the AI is being trained by AI?
is this what they mean by eat your own dog food
nah, dogs eat far better than this ...
Because of course they do lol
Stop AI-ifying your platforms