We've been making AI more human, but what does AI think about making you more human?
After training over 30,000 CEOs, explorer/scientist/bestselling author, Don Schmincke, decides to conduct an unusual experiment:
How would Artificial Intelligence train us to improve ourselves?
Don says, "I've written books and done over 2,000 speeches on leadership development, so I didn't need to take on another literary project.
My intention was to experiment with how AI would respond to a human request for 'personal development'."
As an experiment on uncovering self-help insights, Don used a chatbot born on November 2022., called ChatGPT (Chat Generative Pre-Trained Transformer). Built on top of OpenAI's GPT-3 family of large language models, it was fine-tuned with transfer learning and deep learning architectures. It then fed on what humans had published on the internet including Wikipedia, Twitter, Reddit, books, articles, webtexts, and any other data scraped from all corners of the web; about 300 billion words.
The chatbot accessed everything published, answered all of Don's questions, and wrote the book . . . all in less than 2 hours.
The results were surprising.
