HackerTrans
TopNewTrendsCommentsPastAskShowJobs

_fbpp

no profile record

comments

_fbpp
·há 3 anos·discuss
All this under the auspices of "ethics", which as a reminder, is just an arbitrary set of rules which someone is trying to pass off as having a divine origin.

They are arbitrary only in the sense that, sans a religion's God laying down ethics from on high, all ethics are arbitrary human creation.

But we care about ethics all the same. Engineers care about ethics because if they do not, they kill people. Civil engineers care because faulty work kills people. Electrical engineers care because faulty work kills people. Mechanical engineers care because faulty work kills people.

And as software engineers, we should care because our careless work kills people.

And if we do not care, the government will force our hand. And they will not listen to pleas for them to be reasonable.

Software engineers abused people's personal data, the EU's GDPR has outlawed using all but the absolute minimum personal data. Meta has been told in court by the CJEU that advertising is not an acceptable use of personal data, even if needed to pay the bills. Ad-tech is a doomed industry.

So it's your choice. Start caring about ethics. Or the government locks our field in so much regulatory gridlock that you will wish they just outlawed it entirely.
_fbpp
·há 3 anos·discuss
The fun part is that the GDPR already does. The answer is you're not allowed to use personal data for AI. (And "personal data" here covers things like all public social media posts)

Facebook recently got told by the CJEU that, no, they can't use people's posts to target advertisements. Even if those ads are what's paying for the platform. That you can't claim such processing as "part of the contract" unless it is absolutely necessary in the same way the post office needs an address to send a parcel.

If Facebook can't even do that, there is no way LLMs will be allowed. (And remember. The GDPR does not care if your system doesn't distribute personal data. Any kind of processing at all falls under the GDPR's requirements)

OpenAI is already being chased by the EU's privacy agencies. Right now they're in the process of asking pointed questions, things will heat up after that.
_fbpp
·há 3 anos·discuss
The entire fair use claim is derived not from any legal basis, but rather, that "it has to be fair use" because it would be legally catastrophic for OpenAI et al if it weren't true.

If you look at the core argument in favour of fair use, it's that "LLMs do not copy the training data", yet this is obviously false.

For Github copilot and ChatGPT examples of it reciting large sections of training data are well known. Plenty can be found on HN. It doesn't generate a new valid windows serial key on the fly, it's memorized them.

If one wants to be cynical, it's not hard to see OpenAI/etc patching in filters to remove copyrighted content from the output precisely because it's legally catastrophic for their "fair use" claim to have the model spit out copyrighted content. As this is both copyright infringement by itself, and evidence that no matter how the internals of these models work, they store some of the training data anyway.