> Think what would happen if there was no watermark?
Everything would be worse and lies would be even easier to propagate. I understand the point in building this—understanding if it is possible—but what’s then the point in releasing it with instructions?
It’s like inventing a new kind of undetectable poison and saying “Could you imagine what would happen if people put this is in each other’s drinks?! Anyway, here’s the recipe”.
> Important Disclaimer
> THIS SOFTWARE IS PROVIDED FOR EDUCATIONAL AND RESEARCH PURPOSES ONLY. Use this tool responsibly and ethically.
This means absolutely nothing. Absolutely no one, exactly zero people in the history of mankind who have unethical uses for this tool will care about the message.
What even is the responsible and ethical use for your tool? Is there any?
Consider that you can’t push ethics down the line, they have to start with you. Should you have shared this? Is that ethical? Again, I understand the building of it, especially to demonstrate to OpenAI their efforts to prevent deliberate misuse (assuming that’s indeed the goal of the watermark) are laughably inadequate, but should you be distributing it to every random petty phisher and scammer on the internet?
> Think what would happen if there was no watermark?
Everything would be worse and lies would be even easier to propagate. I understand the point in building this—understanding if it is possible—but what’s then the point in releasing it with instructions?
It’s like inventing a new kind of undetectable poison and saying “Could you imagine what would happen if people put this is in each other’s drinks?! Anyway, here’s the recipe”.
> Important Disclaimer
> THIS SOFTWARE IS PROVIDED FOR EDUCATIONAL AND RESEARCH PURPOSES ONLY. Use this tool responsibly and ethically.
This means absolutely nothing. Absolutely no one, exactly zero people in the history of mankind who have unethical uses for this tool will care about the message.
What even is the responsible and ethical use for your tool? Is there any?
Consider that you can’t push ethics down the line, they have to start with you. Should you have shared this? Is that ethical? Again, I understand the building of it, especially to demonstrate to OpenAI their efforts to prevent deliberate misuse (assuming that’s indeed the goal of the watermark) are laughably inadequate, but should you be distributing it to every random petty phisher and scammer on the internet?