ChatGPT's response to the sleep-deprived thought your post prompted: > Sure, here are some "You might be an AI language model if..." jokes in the style of Jeff Foxworthy: > > 1. If you've ever corrected someone's grammar in your head during a conversation, you might be an AI language model. > 2. If you can translate a sentence into multiple languages in under 10 seconds, you might be an AI language model. > 3. If you've ever used an algorithm to generate a love letter, you might be an AI language model (and probably a little bit lonely too). > 4. If you find yourself responding to messages with "I'm sorry, I didn't understand the question," you might be an AI language model (or just very confused). > 5. If you can write an entire essay on a topic you know nothing about, you might be an AI language model (or a really good bullsh**er). > 6. If you've ever auto-completed a sentence with something completely nonsensical, you might be an AI language model (and in desperate need of some context). > 7. If you're constantly learning from your mistakes, but also constantly making new ones, you might be an AI language model (just like us humans). > 8. If you've ever tried to predict the end of a sentence before it's even been typed, you might be an AI language model (or a mind reader, but probably the former).

"Dying is easy. Comedy is hard." -- Actor Edmund Gwenn's last words.

It's a limitation that Open AI were quite aware of when they released chatGPT. [https://openai.com/blog/chatgpt] > • The model is often excessively verbose and overuses certain phrases, such as restating that it’s a language model trained by OpenAI. These issues arise from biases in the training data (trainers prefer longer answers that look more comprehensive) and well-known over-optimization issues Turns out when you have two answers and you are asked which one is better, people just pick the longer one.

Here's my 0.0002 Dogecoin of https://www.evalapply.org/posts/animate-text-art-javascript/index.html#architecture-discovering-the-spreadsheet-in-the-problem the thing. Share yours! > Spreadsheets are a personal favourite tool. Commonplace problems tend to fit snugly into spreadsheet-like models and behaviours. When confronted by a new problem, I invariably make tables to make sense of it. And I've used spreadsheets a lot in off-label ways too (Ever created dynamic data-driven app mock-ups in Excel?). Yet, it took some doing before I finally saw that I can (should) use the web browser + DOM as a dynamic medium to play with rather than an inert target to overwrite.