Like everyone's opinion about BGC forums used to train AI models

If you would like to post, you'll need to register. Note that if you have a BCG store account, you'll need a new, separate account here (we keep the two sites separate for security purposes).

After reading your discussion and valid points mentioned I would like to add a comment that is not directly connected to legal framework of using forums data:
  • There is a long history of proscribed human activities of the past that are normal today
    • Autopsy as an example
  • No amount of legal effort stops the activity that is attractive to people
    • Drugs and substance abuse
  • It is normal that we learn directly or indirectly from what other people achieved through their trial and error cycle
    • It would take me years to learn what I got instantly here
  • AI is not a sinister dark force in a Jedi universe
    • In a same way as books are not considered to be illegal use of other humans experience
We should not be stopped by fear of the ideas being used elsewhere IMO. The way forward is to be ahead of the curve and to use results generated by AI from our discussions.
 
After reading your discussion and valid points mentioned I would like to add a comment that is not directly connected to legal framework of using forums data:
  • There is a long history of proscribed human activities of the past that are normal today
    • Autopsy as an example
  • No amount of legal effort stops the activity that is attractive to people
    • Drugs and substance abuse
  • It is normal that we learn directly or indirectly from what other people achieved through their trial and error cycle
    • It would take me years to learn what I got instantly here
  • AI is not a sinister dark force in a Jedi universe
    • In a same way as books are not considered to be illegal use of other humans experience
We should not be stopped by fear of the ideas being used elsewhere IMO. The way forward is to be ahead of the curve and to use results generated by AI from our discussions.
good points. like it or not, AI is here to stay. Who we use and to a limited extent contribute to it, is a personal choice. Without being a hermit and living under a rock, we can not avoid AI
 
Tbh the Nazi soldier images debacle shows that “machine learning” is woefully short of the standard you might expect as “being taught”. Be interesting how the AI companies solve that massive short cut.
 
I am just beginning to understand how AI works.

From what I learned so far AI involves consuming a large amount of material widely available everywhere. The computer then collates and analyzes the information to come up with "original" new material. With some things with fixed rules like games like chess or go, the computer can outdo human capability. They can also do something "original" that sounds like it was created by a human being.

In this context what we have discussed and created in this forum is just a small part of what AI would ingest and accumulate to go into a calculation. They are likely to ingest every forum on photography on the net, around the world.
There is no way this could be monetized to Steve's or our benefit.

I think what is being built on this site is a human community. I think it is highly unlikely that one of the contributors on this site is an AI agent as opposed to a real human being. Although what do I know.

Like many of us I don't know where all this is going. Will AI computers become smart enough to become a threat to human civilization? No less a genius than Stephen Hawking believed that AI is a threat to humanity.

One thing I think is human beings, and all life forms, are moved and developed through evolution to feature traits that promote survival and propagation. AI by contrast seems unrooted to any principles and not directed to any particular goal other than what its developers suggest. What is the moral foundation of its existence what are its purposes and goals? Where does it find a sense of morality or a code of conduct or behavior? Does it even care what it does and whether it survives? Why should it? It is too new to be influenced by evolution.

Human beings are uniquely adept at survival and dominance. But they are not good at balance, restraint and stewardship of their own environment. If we don't change we will not survive. A plague of locusts, after consuming everything, dies out.

Maybe AI is at root a tool of human beings and its behavior will be determined by the ways in which it is developed and directed.
 
Okay, my thoughts.

Ignoring the agreement we all signed when we joined the forum (or at least I think I agreed to something), here are some things I consider

- it is Steve's site
- the site without all our contributions is worthless
- there is a lot of trust we place in the site and most of all, are glad to share our knowledge freely (why else would we post it on a public forum).

Thus if Steve's wants to let an AI company use this site to train an AI model, great. He will reap a finance reward. I would like to see Steve share that with the membership, in particular the active members. This could work by give active members, moderators, some small share of the profits. Or perhaps require the AI company to provide free access to its photographic (or perhaps all) AI tools either inpepituity or for X years.
I want nothing to do with AI period. It's caused too much trouble and it makes it easier for people to lie. In my area recently, a cargo ship hit a bridge. When photos of it first appeared even the local tv reporters didn't believe it and thought it was AI. Another example, when the storms came through the Carolinas people were posting images of the storms with huge waves. They were fake. A photographer actually got on social media with his true images and exposed the fakers. What is sad is that meteorologists actually use images of storms for their data. When someone fakes an image and posts it as real then it causes trouble and extra work for meteorologists to sift through to get accurate information. I also belong to a beginners photography group. I caught someone using AI to generate an outdoor image of her dog which was really taken inside her home. The big giveaways was an artifact in the background and the catchlights in the dog's eyes. When I called her out on it she admitted it. She there's to try to learn photography. She's not going to learn photography that way. I suggested to her to forget AI, learn the exposure triangle, learn how to be a better photographer and she would get more satisfaction. She agreed. This isn't just a problem with photography, there's music and the written word. Teachers are trying their best to keep students from using AI to do their research papers. It is once again a technology which has run amok and used for nefarious purposes.

Having said that, I would never use and I don't use AI for my photography. I don't have a problem with content aware fill and such as it's already using parts of my image that's already there. That's my image(s). I do have a problem with using pieces of someone else's work. That I am not going to do.

The stock company I contribute to sent an email out to all of us a couple of years ago asking us to consider submitting images for their "data sets" and I declined. Since I still have rights to my images I contribute, I'm holding them to their word that they will not use my images for AI. I refuse to do that even though it may add $ to my account. I'm not willing to have pieces of my images floating out there in AI land no matter how much money they offer me. I'm not going to possibly contribute to someone else's fakery.
 
The saying about pictures don't lie is totally out the window now. Can you imagine in a court case if someone who is innocent got a guilty sentence based on a AI generated image? So, the saying a picture is worth a thousand words can now say a picture is worth a thousand words of lies.

Yes, I do have strong feelings about it because I see friends reposting images which are fake (like the Carolina storms) and actually believing it's real. That's sad.
 
First of all i am not proposing, as far as I know Steve has not licensed this site nor has he been approached by any AI compnay. I thought about question after hearing that Reddit had an IPO and the value of the social media site was based upon the value that AI learnings that could be gleaned from the site.

Now what about this site? Great place to learn photography - granted that is narrow area for AI to learn but BCG forums could be an excellent want for AI to learn about photography and possible learn to evaluate images.

Like to your thoughts on this. Again I have no knowledge whether or not Steve has even considered licensing the site for AI learning, I have NOT dug into the licensing terms, I do not know if any AI company has approached Steve. So consider this a fun discussion, how would you feel if you learned, either by Steve announcing it or from another source, that BCG forums was being used to training AI models in photography, in particular wildlife photography.

Please do not get mad at Steve, he has not done anything as far as I know, and I would like to have other's share their thoughts as I have mixed feelings about this.

My thoughts will be on the next msg
Anything on the Internet has been scrapped to educate AI. That is a major issue with AI at this point because no one ask or was given permission to take a lot of the information or art.
 
The saying about pictures don't lie is totally out the window now. Can you imagine in a court case if someone who is innocent got a guilty sentence based on a AI generated image? So, the saying a picture is worth a thousand words can now say a picture is worth a thousand words of lies.
Pictures always had the ability to lie. The choice of lens, aperture, shutter speed, what’s in the frame and not at the least distort reality. Then people manipulating a scene whether baiting wildlife or moving a twig or leaf if you get technical. Then film choice color rendering or digital post processing alters things even more.

Manufactured evidence isn’t new either. All that has really changed is how easy and accessible it is now. One simple solution that would work for the time being is if every AI video and picture was required to have a digital watermark to indicate the origin. But that only keeps honest people honest.

But to round back to the original topic again, it’s going to happen that bots and data loaders will read the data for their purpose. My sense of “revenge” for that is to use that AI for things that benefit me.
 
Not sure why people perceive AI as being more of a threat than previous methods of unauthorized use of images. IMO anyone who is concerned that they might be materially impacted by such activity shouldn't be posting on public spaces in the first place. Simple as that.
AI brings with it a general threat to image authenticity on a scale and quality not seen before, as a result any image you present in the future, to anyone, may be received differently, perhaps with less trust that it actually represents anything real that has actually occurred. That general perception will happen whether or not any individual has posted material anywhere in the past. Quite soon people may start to believe anything of note must just be AI generated, I already see it in groups where someone may not be familiar with a particular technique and falsely calls an image AI.
 
AI brings with it a general threat to image authenticity on a scale and quality not seen before, as a result any image you present in the future, to anyone, may be received differently, perhaps with less trust that it actually represents anything real that has actually occurred. That general perception will happen whether or not any individual has posted material anywhere in the past. Quite soon people may start to believe anything of note must just be AI generated, I already see it in groups where someone may not be familiar with a particular technique and falsely calls an image AI.
Oh. I get it. So there are indeed potential real damages to those of us not making a living from photography. I'm not sure if my ego would survive someone accusing me of posting AI generated/altered imagery. :confused:
 
I can honestly say that I would not like the site intentionally sharing with an AI program and it would limit my participation. Having said that, as Steve says, it is a public site and we can’t keep dishonest people from doing what they want with the data. This is a valuable forum and Steve is doing his best to keep it that way. Thanks Steve!
 
First of all i am not proposing, as far as I know Steve has not licensed this site nor has he been approached by any AI compnay. I thought about question after hearing that Reddit had an IPO and the value of the social media site was based upon the value that AI learnings that could be gleaned from the site.

Now what about this site? Great place to learn photography - granted that is narrow area for AI to learn but BCG forums could be an excellent want for AI to learn about photography and possible learn to evaluate images.

Like to your thoughts on this. Again I have no knowledge whether or not Steve has even considered licensing the site for AI learning, I have NOT dug into the licensing terms, I do not know if any AI company has approached Steve. So consider this a fun discussion, how would you feel if you learned, either by Steve announcing it or from another source, that BCG forums was being used to training AI models in photography, in particular wildlife photography.

Please do not get mad at Steve, he has not done anything as far as I know, and I would like to have other's share their thoughts as I have mixed feelings about this.

My thoughts will be on the next msg
My opinion.

The subject of AI is dynamic, complex as well as in cases somewhat still green fruit on the vine............the benefits or consequences will over time be clearer, Steve will adapt i feel fairly and quickly if needed.

Its Steve's site, i believe he is ethical reasonable and intelligent, above all he has no BS about him.

If someone doesn't like things or agree they have the option to move on, quietly.

I completely rely on his judgement in making appropriate calls if and when needed to address any matters arising be they commercial or ethical or AI.

As to making money, i never deny anyone from that right, sadly with the internet you need to adapt and go with the flow as things change if you don't adapt quickly you will be left behind, so if you want to survive or grow you need to GO with change decisively and quickly, if the call is AI that is the new influencer so be it, if its not AI it will be something else, what ever it is you need to act and adapt quickly or pass it by, this is also my personal experience.

Of course the way money is made does matter to most people but not to all people, i feel Steve is more than capable of knowing whats right or wrong or what his needs are.

Bottom line, its Steve's call to do what ever he thinks is appropriate, its his site, its the audience that changes, if we don't like something we can say something or move on quietly.

If you posted images on BCF then as long as you know before posting that AI may have the right to use part or all of the image is still your option to say yes or no before you post, if its no don't post it on the BCF or any site, simple. If commercial matters are applicable then it be so, this is the green fruit i am referring to that needs time to be understood or evolve clearly.

Any image being used by AI we know how to handle that. Policing it is whats required and hard or is it even possible to do so reliably, the laws are still thin.

I recall possibly ? Sony developing something to offer the protecting of images or files from AI ? in camera, not certain.

AGAIN, for BCF its Steve call,
if he doesn't make money and adapt he wont be here like many others, simple.
happy days.

Only an opinion
 
Last edited:
Back
Top