New generative AI tool AutoGPT can self-prompt to develop and manage tasks

The launch of the artificial intelligence chatbot ChatGPT by OpenAI has paved way for several spin-offs and applications. One of them is Auto-GPT, an experimental open-source application that can run autonomously, based on basic and minimal prompts as input. AutoGPT is a Python application developed by Toran Bruce Richards.

AutoGPT can access the internet and has short- and long-term memory management. It uses GPT-4 for text generation and uses GPT-3.5 for file storage and summarisation, as per its GitHub. It requires little human interaction and is able to ‘self-prompt’ through the ‘added tasks’ functionality. It means that a human user can tell the application only the end goal and it will self-produce every prompt required to complete the task. AutoGPT can perform tasks like code debugging, email writing, and even work on innovative business ideas.

“Auto-GPT is an experimental open-source application showcasing the capabilities of the GPT-4 language model. This program, driven by GPT-4, chains together LLM "thoughts", to autonomously achieve whatever goal you set. As one of the first examples of GPT-4 running fully autonomously, Auto-GPT pushes the boundaries of what is possible with AI,” the GitHub description said.

Soon after its launch in March, the application went ‘viral’. On GitHub, AutoGPT has 88,000 stars and has been forked more than 13,400 times. 

The co-creator of Sophia the Robot, Ben Groetzel was recently quoted by ZDNet saying that jobs that do not require creativity and innovation could be replaced by ChatGPT and similar tools.