ChatGPT and DALL-E 3 AI-generated images are now watermarked, though OpenAI admits it's "not a silver bullet to address issues of provenance

[Windows Central]

OpenAI adds watermarks to ChatGPT and DALL-E 3, but this may not be enough to solve the issue of AI deepfakes. The change won't affect text or voice generated using AI.

ChatGPT and DALL-E 3 AI-generated images are now watermarked, though OpenAI admits it's "not a silver bullet to address issues of provenance : Read more

 

[Kevin Okemwa]

Watermarks on AI-generated images is a step in the right direction and might help identify deepfakes, but will it be enough to remedy this situation? Share your thoughts with me

 

[TheFerrango]

Probably not. Watermarks may be forced into large, consumer used LLMs and generators, but on-premise models and large commercial ones won't be verifiable. It's a step in the right direction, but it's nowhere near being a solution

 

[Kevin Okemwa]

Probably not. Watermarks may be forced into large, consumer used LLMs and generators, but on-premise models and large commercial ones won't be verifiable. It's a step in the right direction, but it's nowhere near being a solution

I think so too. More censorship and guardrails may eventually impact the quality of the service offered by these tools as well.

 

[TheFerrango]

I think so too. More censorship and guardrails may eventually impact the quality of the service offered by these tools as well.

Oh yeah, I didn't mention regulations and censorship because I have mixed feelings about them, and have limited power anyway