Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Zoom terms now allow training AI on user content with no opt out

explore.zoom.us

Post reply on HN
Print view View original

221–230 of 538 posts

Re: Zoom terms now allow training AI on user content with no opt out

#221

Post by Havoc »

duringmath wrote: post #54

Using user generated data to train an AI is no different than scanning it for spam or any other administrative function, and using public data to train your AI model is fair use and everyone should get over it already.

>Using user generated data to train an AI is no different than scanning it for spam

That's definitely not true.

Under some circumstances LLMs can spit out large chunks of the original content verbatim. Meaning this can actively leak the contents of a confidential discussion out into a completely different context, a risk that does not exist with spam scanning.

Re: Zoom terms now allow training AI on user content with no opt out

#222

Post by CreRecombinase »

version_five wrote: post #47

To be fair, there's no way Microsoft isn't doing something worse with data from Teams. And the only reason google "products" exist is to mine data to sell ads. I can see how we'd expect better from a company focused on video chat, but it's not like any of their big competitors are actually treating your data respectfully.

That’s absolutely not true about Microsoft. They’re very clear about how they use your data. If they’re mining data in secret against their own terms that’s felonious criminal behavior.

Re: Zoom terms now allow training AI on user content with no opt out

#223

Post by lewhoo »

jimmaswell wrote: post #200

Maybe this can be our magic bullet to save free content after we killed ads: train AI on user input/behavior (at least for a brief period of time until the EU throws another fit and California follows suit). Perfectly fair tradeoff to me - it should be fair use for AI to train on anything at all in my view, same as it's fair use for a human to look at something and learn abstract lessons from it.

Oh not the "like a human" again. But would you really allow any human on every zoom call ? I think this is a good example where this analogy breaks pretty obviously.

Re: Zoom terms now allow training AI on user content with no opt out

#224

Post by albert_e »

account-5 wrote: post #102
albert_e wrote: post #63

At the scale of Zoom and MS Teams ... you could theoretically train an AI model that can autonomously conduct all meetings businesses ever need -- all day every day -- without any human ever needing to attend. So much productivity claimed back! GenAI provides the agenda, GenAI bots log in with a AI avatar and spout hallucinations, bots agree to disagree and setup a followup meeting next week after resolving fake cale…

And this continues even after the human race disappears. There's a dystopian sci-fi novel here somewhere.

Yep! I realize this kind of sounds like Ray Bradbury's imagined dystopian future where a fully automated house continues to go about its programmed routine after all its inhabitants had died in a nuclear event that obliterated the rest of the city.

https://en.wikipedia.org/wiki/There_Will_Come_Soft_Rains_(sh...

Re: Zoom terms now allow training AI on user content with no opt out

#226

Post by LinuxBender »

deusexml wrote: post #218

I guess the equivalent terms for Google Meet are much more privacy friendly: > Control over your data > ... > Google does not store video, audio, or chat data unless a meeting participant initiates a recording during the Meet session. https://support.google.com/meet/answer/9852160 Though, I suppose this isn't exactly the same as a TOS.

Out of curiosity based on recent discussions and debates about AI and copyright, would it be considered storing if AI processes the information first and then stores the derivative works?

Re: Zoom terms now allow training AI on user content with no opt out

#228

Post by TheRealPomax »

jamesgill wrote: post #206

It seems to me that every ‘AI’ effort I see today utterly depends on circumventing or ignoring privacy.

I like the part where the reason the US keeps allowing it is that "if we don't, China will", as if China has access to the same flood of data that US FAANG companies and contenders like Zoom have access to every second of every day.

Sure, at the government level it has access to the same data as everyone else, but that firewall's still there, can't have an AI trained on data that might give a more worldly view on matters the party doesn't want citizens exposed to. A Chinese AI will be pretty useless for western audiences, so best they can do is make the hardware.

Which they already do.

Re: Zoom terms now allow training AI on user content with no opt out

#229

Post by icco »

galleywest200 wrote: post #177

One of my doctors uses Zoom for tele-health appointments... How is this not a violation of HIPAA?

I am not a lawyer, but https://explore.zoom.us/docs/doc/Zoom-hipaa.pdf reads like as long as they don't disclose your information, they aren't in violation.

Re: Zoom terms now allow training AI on user content with no opt out

#230

Post by TheRealPomax »

icco wrote: post #227
jamesgill wrote: post #206

It seems to me that every ‘AI’ effort I see today utterly depends on circumventing or ignoring privacy.

Definitely, but so does every advertising play. While AI companies are more egregious, this isn't anything new.

That seems an intentionally blind take on the matter. That's like saying that because you've seen a kid shoplift a pack of gum at a gas station, an organized crime outfit stealing entire ATMs from every gas station in a 200 mile radius isn't anything new.

Scale matters.

Post reply on HN