ChatGPT has a new competitor and this one comes with some serious super powers. OpenAI’s ChatGPT is known for generating human-like texts but the key is to feed in the right prompt. Auto-GPT, the new entrant in the generative language space, frees user from the task of typing prompts.
Developed by Significant Gravitas, Auto-GPT is a new version of ChatGPT that uses the GPT-4 API to complete tasks on its own, without needing you to give it specific prompts. Instead, you just set a goal for it, and it generates its own prompts and completes the task for you. Basically, the technology frees you from the task of thinking of all the prompts yourself. Auto-GPT is an open-source app that can be used for a variety of tasks, such as generating text, answering questions, and more.
ALSO READ l Infosys founder Narayana Murthy on why ChatGPT can’t beat the human mind
How to set up Auto GPT
According to the Auto-GPT GitHub, Auto-GPT chatbot program requires an environment and the ChatGPT API to work. To set up Auto-GPT, you need to first set up your environment using VSCode + devcontainer, Docker or Python 3.10 or later. You can use a Windows, Mac or Linux PC for this. Next, you need to set up the ChatGPT API by going to OpenAI’s site and creating a pay-as-you-go account. Once you have created an account, you can get an API key that you will use to connect your Auto-GPT program to the GPT-4 large language model. With the environment and the ChatGPT API set up, you can start using Auto-GPT to create your own chatbot.
ALSO READ l German authors, performers call for tougher ChatGPT rules amid copyright concerns
Just like other AI systems, Auto-GPT also comes with its own set of limitations. The GitHub page notes that it isn’t a polished application and may not be suitable for complex and real-world business cases. Most important of all, the program is quite expensive to run so it is advisable to set up and monitor your API key limits to avoid unnecessary charges.
