I remember that game. Always wanted to play it as a kit, but nobody I knew had it. Still, haven’t done so up to today…

Maybe this is how a modern take of the game would look like?

Retro Version:

The question I pose is not “how you implement a user-facing product”, but rather the “how can it be tolerable to offer a product when the output deliverable has tell-tale artifacts, and when the image provided to a user is watermarked, potentially with your organization and your personal ID verification traceable?”

And then, there is imagery content OpenAI is allowing, model trained on what is behind a unprompted wardrobe malfunction, for example. For a social media impression casting a negative light on a company, where they may retroactively cite “policy” for their bad output, or is the kind of dissent where a goverment may demand to know who to lock up for 100 years with manufactured conspiratorial allegations, do you want to be in the crossfire (or see if OpenAI will really reimburse your copyright legal bills like was said just once in 2023)? OpenAI are not transparent about identifying embedded metadata, that they are including (or never including and will never include).

When I get the chance, I’ll make a proper intro to Paca, always helping with AI around the town for sustainable everything.