Meta Pauses Employee-Tracking Program Following Internal Data Leak (wired.com) 22
Meta has paused its Model Compatibility Initiative that tracked employee mouse movements, clicks, keystrokes, and screen content to train AI agents, after some of its collected data became accessible to more employees than intended. Meta says it has no evidence the information was improperly accessed and will not restart the program until it is confident in its safeguards. Wired reports: Meta rolled out the Model Compatibility Initiative (MCI) tool in April to US employees. The tool "collects computer inputs such as mouse movements, click locations and keystrokes, as well as screen content," according to workers who have been petitioning against it over privacy, security, and personal liberty concerns. When MCI launched, employees couldn't opt out, but that changed to a limited degree after workers protested. Meta executives have repeatedly defended the data-gathering project, saying it was necessary to train AI systems to operate computer software the way humans do and that employees were the best examples for the artificial intelligence to learn from.
On Monday, a Meta engineer issued an internal security notice stating that databases filled with information gathered by MCI had been exposed to anyone inside the company. A former employee actively involved in pushing back against MCI describes the lapse as "a mess" -- and one that employees had expected would occur. "When workers raised concerns, leadership doubled down and failed to acknowledge the risks workers raised about the safety and privacy of worker and customer data," the person says. "Leadership has clearly created an authoritarian environment where workers are no longer respected or heard."
But after critical comments poured into internal forums on Monday expressing frustration about the security issue, Meta shocked some of its staff by pausing MCI altogether, telling WIRED about the development several hours before announcing it to employees. A few workers told WIRED they were confused in the meantime because the tool was continuing to run on their laptops. Late on Monday, Stephane Kasriel, a Meta vice president overseeing AI research, announced the pause and told staff that the security issue had been discovered on June 18 and addressed within four hours. But the initial fix didn't stick and access to the data had to be further locked down. The issue made "some MCI-derived data" accessible to more people than intended, he wrote, without elaborating.
On Monday, a Meta engineer issued an internal security notice stating that databases filled with information gathered by MCI had been exposed to anyone inside the company. A former employee actively involved in pushing back against MCI describes the lapse as "a mess" -- and one that employees had expected would occur. "When workers raised concerns, leadership doubled down and failed to acknowledge the risks workers raised about the safety and privacy of worker and customer data," the person says. "Leadership has clearly created an authoritarian environment where workers are no longer respected or heard."
But after critical comments poured into internal forums on Monday expressing frustration about the security issue, Meta shocked some of its staff by pausing MCI altogether, telling WIRED about the development several hours before announcing it to employees. A few workers told WIRED they were confused in the meantime because the tool was continuing to run on their laptops. Late on Monday, Stephane Kasriel, a Meta vice president overseeing AI research, announced the pause and told staff that the security issue had been discovered on June 18 and addressed within four hours. But the initial fix didn't stick and access to the data had to be further locked down. The issue made "some MCI-derived data" accessible to more people than intended, he wrote, without elaborating.
Spotlight Accountability. (Score:3)
Meta has paused its Model Compatibility Initiative..after some of its collected data became accessible to more employees than intended.
When a candle is all that is needed to put a spotlight on your behavior to change it, it says a lot about your fucking behavior.
Those are supposed to be valued employees. Not suspects.
Re:Spotlight Accountability. (Score:4, Insightful)
Yeah... you're supposed to think you're a "valued employee"... the company couldn't give a *i*s less about you.
And, working for a company like that, you should be expecting nothing less than replacement, at some point.
There's no such thing as a 'valued employee'... just an expendable nobody.
Re: (Score:2)
I don't see the problem. The data is being used only to train AI agents. I'm sure it would never be used for any other purpose, intentionally or unintentionally...
And I'm also sure that none of the employees being tracked are at risk of losing their jobs to said AI agents.
Re: (Score:2)
Not that valuable for AI (agents are software using AI). It's just surveillance. They may still be using AI to detect patterns who is "unproductive".
Re: (Score:2)
Meta has paused its Model Compatibility Initiative..after some of its collected data became accessible to more employees than intended.
When a candle is all that is needed to put a spotlight on your behavior to change it, it says a lot about your fucking behavior.
Those are supposed to be valued employees. Not suspects.
Unfortunately, in the modern day, suspects is exactly what we all are. To our government. To our managers. To our police forces. To our politicians. We're all guilty of something, and they believe they deserve to know every detail of our lives until they find out what we are guilty of.
Maybe forty-plus years of continuous political one-upmanship on the rhetoric of fear has led to this, or maybe this is just the end result of our path anyway, but it's our reality.
Re: (Score:3)
Those are supposed to be valued employees. Not suspects.
It's only a matter of time, though.
If you want to know the best case for how this ends, read Player Piano by Kurt Vonnegut.
You can't look away from it (Score:2)
Meta executives have repeatedly defended the data-gathering project, saying it was necessary to train AI systems to operate computer software the way humans do and that employees were the best examples for the artificial intelligence to learn from.
Okay, this puts a chill up my spine.
Re: (Score:3)
I find it hysterical. The current massive problems with LLMs are not a result of too little training data. They are a result (in part) of low quality training data. And they are just gathering more of it.
Old joke, bad business practice (Score:5, Insightful)
Cop sees a guy at midnight on his hands and knees looking under a street light. Asks him "What are you doing?"
Guy says "Looking for my car keys."
Cops help him look for a while and then asks "Are you sure you dropped them here?"
Guy replies "No, I dropped them over there in the dark but there is no chance at all I could find them there, it is too dark."
When you track things, you think you make decisions/rules based on what you track. But you can't really track the good stuff like effectiveness, creativity, or intelligence. Mouse clicks, key strokes etc. are what they track, so they build their AI on that stuff.
It's no different than looking under the street light - you won't find what you want, just the stuff that is easy to find.
This kind of thing is an unnecessary invasion of privacy that will result in an AI copying the mistakes of humans, not their best behavior.
Re:Old joke, bad business practice (Score:5, Insightful)
You can detect who's taking a break, though. But tracking IT worker productivity by activity time is as useless as using meters of mouse movement. The productive part is not typing the actual code, but thinking about it. And this may involve you sitting in front of the machine not touching any part of it, or taking a walk in the park being "unproductive" while thinking about the solution.
You insensitive clod! (Score:2)
I tab between input fields. Not play with the stupid mouse!
Re: (Score:3)
You boss is keeping tabs on you.
Re: (Score:2)
Re: (Score:2)
A working place with a balcony or terrace is nice. Do a few steps, get some fresh air and follow the 30-30-30 rule (usually increase all of the numbers a bit).
Re: (Score:2)
The law has something to say about that in certain jurisdictions (especially workplaces) and has put limits on what your workplace can actually monitor and to what degree and who can view it... sure, the computer you use is your works, but does that mean they can capture your SIN and other info for your manager to review or some other shmuck when you're filling out data for your HR? or accessing a confidential internal reporting page to report your boss for stealing from the org? Or you're on the health i
Title Correction: (Score:1)
"Privacy Rapist Meta[stasize] Pauses Employee-Tracking Program Following Internal Privacy Data Rape"
There FTFY.
Re: (Score:1)
Sebby the rape fetishist!
Here you go [slashdot.org] you fragile & ignorant Meta[stasize]/Alphaslop employee.
Most weirdest fcuked up company in history (Score:3)
Sorry for any soul who has to work there. Creepy as creepy can get.
Not the last of the boggling paradoxes that these champions of capitalism and liberty end up running their companies like an authoritarian state.
Nope, nope, nope.
Glad I didn't (Score:2)