The first principles it writes are that AI is just a tool, AI is powerful, AI is not conscious. The tactic they suggest is instructing the model to work out its own solution before rushing to a conclusion. They also point to the example of using code execution to perform more accurate calculations or call external API's. And this kind of gets back to how they're now open sourcing evaluation models so that people can get more involved in helping them improve chat CPT over time.
OpenAI has published a new guide with 6 recommendations for better ChatGPT prompting. The recommendations come in response to numerous complaints on Twitter that ChatGPT has gotten worse over the last few weeks.
The AI Breakdown helps you understand the most important news and discussions in AI.
Subscribe to The AI Breakdown newsletter: https://theaibreakdown.beehiiv.com/subscribe
Subscribe to The AI Breakdown on YouTube: https://www.youtube.com/@TheAIBreakdown
Join the community: bit.ly/aibreakdown
Learn more: http://breakdown.network/