J Rice

You can acquire significantly higher quality responses for complex tasks from ChatGPT by simply adding "Plan out the solution prior to generating <code/solution>" to the end of the prompt. The screenshot shows an in-depth solution to the problem using this technique. A portion of the

#ChatPGT has driven the White House to release a proposal regarding AI and the rights of humans. https://www.whitehouse.gov/ostp/ai-bill-of-rights/ I give each principle, its brief description, and some thoughts for each. The "AI Bill of Rights" has five principles: 1️⃣ Safe and Effective Systems:

A simple trick I discovered to help validate the information that ChatGPT or most LLMs write. As I have shown numerous times, ChatGPT can be confidently incorrect...and so confident it comes across as an authority. Therefore it is hard to distinguish between fact and its compelling fiction. Because of

You may not realize it, but every time you start a ChatGPT session it is given a prompt that you have never seen. There was a workaround to force ChatGPT to tell its pre-prompt, and that pre-prompt would tell ChatGPT who it was, what the training data cutoff was, and

Another serious gotcha using ChatGPT. Be careful when you do this! This can be another hallucination. I have seen several examples of people using the technique of attempting to tell ChatGPT to only source data from a specific site. This is a misunderstanding of how ChatGPT works. ChatGPT does not

It will be an interesting discussion. A 6-month pause has limited value, if any. This is a bell that can't be un-rung, a genie that isn't going back into the bottle...no matter the idiom. The point is still the same. It is out there. We

Possible upcoming defamation suit against ChatGPT? https://www.reuters.com/technology/australian-mayor-readies-worlds-first-defamation-lawsuit-over-chatgpt-content-2023-04-05/ The source of the information it was trained on is most likely liable for this, but how does that work when ChatGPT doesn't give the provenance of the information and rewrites the erroneous information in new

Have you ever wanted to make ChatGPT hallucinate? Well, now you can in one easy step. This is a continuation of my Prompt Engineering Series. And this one is a Gotcha Ever imagined dogs could eat cosmic rays to improve dental health outcomes? Create a fake URL with words in

ChatGPT has a Data Privacy problem. No surprise about that, but this article has some more details. https://www.wired.com/story/italy-ban-chatgpt-privacy-gdpr/ It does a good job of conveying Italy's issues with ChatGPT in a concise way: 🚫 OpenAI lacks age controls for those under age 13 🤥 It

Today I am going to cover some basics of Prompt Engineering. These are my best practices. A few have tried to establish patterns. There are no definitive best practices to date. This is what makes this a lot of fun...patterns and practices will drive consistent behaviors. I enjoy finding