598

OpenAI is reportedly going all-in as a for-profit company(mashable.com)

posted
*
by
Avatar
RmDebArc_5@sh.itjust.works
in
Avatar
technology@lemmy.world
121 comments
hidereport

Surprised pikachu face

Sort:
HotTopControversialNewOld
Avatar
aaaaace@lemmy.blahaj.zone
3 points

Face is looking increasingly compromised and distorted every photo.

permalink
report
reply
Avatar
glitchdx@lemmy.world
17 points

… They weren’t before?

permalink
report
reply
Avatar
sugar_in_your_tea@sh.itjust.works
10 points

Yeah, they weren’t as synergized. Now they’re coordinating with key stakeholders to maximize the efficiency of their aggressive roadmap. Or something, I kinda suck at business jargon.

permalink
report
parent
reply
Avatar
exanime@lemmy.world
55 points
*

… To the surprise of <checks notes> absolutely nobody

Actually I have a question and I admit knowing nothing of the legal framework here but…

Isn’t it absolutely ridiculous that a not-for-profit entity can exists solely for the purpose of developing a closed-source piece of software, demand to train it for free off copyrighted material, just to switch to a for-profit entity??

Sound 100% like tax avoidance. Like me registering a charity so I can throw a mega concert/party privately, secure preferencial treatment on supplies, get discounts on artists or even free performance and then switch to for profit as I start selling tickets

permalink
report
reply
Avatar
jacksilver@lemmy.world
13 points

Originally all their work was supposed to be published and shared with the world, hence the “open” in OpenAI. However somewhere along the way they made a for-profit break off of the original company and started pulling everything in that direction.

permalink
report
parent
reply
Avatar
socialmedia@lemmy.world
31 points

Just want to point out that it absolutely is possible to train an AI that will keep track of its sources for inspiration and can attribute those when it makes a response.

Meaning creators could be compensated for their parts of AI generated stuff, if anyone wanted to.

permalink
report
reply
Avatar
mm_maybe@sh.itjust.works
1 point

I think that there are some people working on this, and a few groups that have claimed to do it, but I’m not aware of any that actually meet the description you gave. Can you cite a paper or give a link of some sort?

permalink
report
parent
reply
Avatar
Trantarius@lemmy.dbzer0.com
5 points

Other than citing the entire training data set, how would this be possible?

permalink
report
parent
reply
Avatar
UnderpantsWeevil@lemmy.world
1 point

The entire training set isn’t used in each permutation. Your keywords are building the samples based on metadata tags tied back to the original images.

If you ask for “Iron Man in a cowboy hat”, the toolset will reach for some catalog of Iron Man images and some catalog of cowboy hat images and some catalog of person-in-cowboy-hat images, when looking for a basis of comparison as it renders the image.

These would be the images attributed to the output.

permalink
report
parent
reply
Avatar
Trantarius@lemmy.dbzer0.com
2 points

Do you have a source for this? This sounds like fine-tuning a model, which doesn’t prevent data from the original training set from influencing the output. The method you described would only work if the AI is trained from scratch on only images of iron man and cowboy hats. And I don’t think that’s how any of these models work.

permalink
report
parent
reply
Avatar
blorp@feddit.uk
9 points

Doesn’t Phind do this already? I haven’t used it much but I remember it showing its sources for answers of code-related stuff

permalink
report
parent
reply
Avatar
bluewing@lemm.ee
5 points

I use Phind solving computer problems. It does cite the sources it uses. At least for distro and general Linux issues. So far, it’s been a very good resource when I’ve needed it.

permalink
report
parent
reply
Avatar
Etterra@lemmy.world
55 points
*

OpenAI: It’s not fair to charge us to use copywriten works.

Also OpenAI: Also you have to pay us for using them.

permalink
report
reply
Avatar
buttfarts@lemy.lol
1 point

That’s why all human creative works done online need to be bean related. To fuck up the data stream and make it unintelligible for AIs and marketing algorithms.

permalink
report
parent
reply
Avatar
bitjunkie@lemmy.world
4 points

copyrighted*

permalink
report
parent
reply

Technology

!technology@lemmy.world

Create post

This is a most excellent place for technology news and articles.

Our Rules

  1. Follow the lemmy.world rules.
  2. Only tech related content.
  3. Be excellent to each another!
  4. Mod approved content bots can post up to 10 articles per day.
  5. Threads asking for personal tech support may be deleted.
  6. Politics threads may be removed.
  7. No memes allowed as posts, OK to post as comments.
  8. Only approved bots from the list below, to ask if your bot can be added please contact us.
  9. Check for duplicates before posting, duplicates may be removed

Approved Bots

Community stats

  • 17K

    Monthly active users

  • 12K

    Posts

  • 542K

    Comments

Community moderators