# The Open Letter to "Pause All AI Development" **URL:** \ **Category:** Great Debates\ **Tags:** ai\ **Created:** [March 30, 2023, 12:49am UTC](https://boards.straightdope.com/t/the-open-letter-to-pause-all-ai-development/981862 "2023-03-30T00:49:32Z")\ **Posts on this page:** 1\ **Showing post:** 329
**Author:** ![Darren\_Garrison](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/darren_garrison/32/92_2.png) [@Darren\_Garrison](https://boards.straightdope.com/u/Darren_Garrison)\ **Post date:** [May 10, 2023, 10:42am UTC](https://boards.straightdope.com/t/the-open-letter-to-pause-all-ai-development/981862/329 "2023-05-10T10:42:32Z")
> [@Dr.Strangelove](#): > > Yeah–it’s almost a blend of a digital and analog computer. You accept some noise in the results, but the switching elements are still digital. Not everything can be made resilient, but inference seems like a good fit, especially when you get to many billions of weights. Errors will just get averaged away to nothing. Like this technology: > [@The next page in the book of AI evolution is here, powered by GPT 3.5, and I am very, nay, extremely impressed](https://boards.straightdope.com/t/the-next-page-in-the-book-of-ai-evolution-is-here-powered-by-gpt-3-5-and-i-am-very-nay-extremely-impressed/975945/1135): > > I sometimes wonder whatever happened with these, which do have waves. Their imprecision isn’t well compatible with things like economic and engineering calculations where you want very precise numbers, but they seem like they would be a good fit for AI like Stable Diffusion and ChatGPT, which seem to be more loose with the precision. (Evidenced by the models being to suited to use 8-bit floating point rather than 64-bit.) --- _[View the full topic](https://boards.straightdope.com/t/the-open-letter-to-pause-all-ai-development/981862)._