Can’t type anything on iOS :(
All of the twitter links are broken for me
All very well, the question is, how will you know? And if you can’t reliably differentiate between ai or human contributions how could this be enforced?
Surely the right approach for this is to use canvas, so you only pay the performance penalty once on startup
> ...using ChatGPT... I noticed that the AI struggled to create characters with evil alignments. > Gemini seemed to struggle even more with creating characters that could engage in questionable or evil actions, even…
Surprising to see these models stumbling on what at first glance seems like a simple task, it would be interesting to see how the non-vision models fare if you convert the problems to ascii art
Can’t type anything on iOS :(
All of the twitter links are broken for me
All very well, the question is, how will you know? And if you can’t reliably differentiate between ai or human contributions how could this be enforced?
Surely the right approach for this is to use canvas, so you only pay the performance penalty once on startup
> ...using ChatGPT... I noticed that the AI struggled to create characters with evil alignments. > Gemini seemed to struggle even more with creating characters that could engage in questionable or evil actions, even…
Surprising to see these models stumbling on what at first glance seems like a simple task, it would be interesting to see how the non-vision models fare if you convert the problems to ascii art