In the last few weeks, ChatGPT and AI tools in general have become viral on the internet, with millions around the world embracing them into their daily lives. ChatGPT has been especially popular among users, given its extensive conversational capabilities. However, these AI tools require the user to input a ‘prompt' to get the desired result.
However, a new GPT-based program called Auto-GPT has been released on GitHub and it has the ability to autonomously come up with ‘self-prompts'. Auto-GPT, with its self-prompting capabilities, has led many people to wonder about how far AI technology has really come in a very short timeframe.
What Is Auto-GPT and How Does It Work?
Auto-GPT is an open-source Python programme that was released on GitHub by a developer who goes by the username ‘Significant Gravitas'. Built on the foundation of GPT-4, this open-source application enables AI to function autonomously without constant user inputs. Once you provide Auto-GPT with an overall objective, it will take step-by-step actions to accomplish that objective. This approach is based on the idea of "AI agents," which can access the internet and execute tasks on a computer independently without requiring constant guidance.
There is also a simple demonstration posted on GitHub where Auto-GPT has been given the goal of browsing the internet and coming up with completely original recipes for upcoming events or holidays. Chef-GPT, as this particular program was dubbed, started scouting the internet to come up with unique original recipes and then as per instructions saved the recipes on the user's computer. The truly impressive feat here is Auto-GPT's ability to search the internet on its own and perform tasks like saving files onto the computer. These are functions that are beyond the capabilities of chatbots like ChatGPT.
People across the world are also coming up with many different use cases for Auto-GPT on the internet. A Twitter user posted an example of ‘E-commerce-GPT', where he had given Auto-GPT the goal to autonomously develop and run an e-commerce business. Another programmer created ‘Robo-GPT', which was built to autonomously write, analyse, rewrite and save code. Others have also started to develop tools similar to Auto-GPT, such as the ‘Microsoft Jarvis', and ‘BabyAGI', which also have the ability to act on self-prompts.
How To Use Auto-GPT?
The process to use Auto-GPT can be somewhat complex for someone who is not very informed about coding and technology. In order to run Auto-GPT, users must have Python 3.8 or above, an OpenAI API key to access GPT, as well as a Pinecone API key. Fortunately, links for all of these tools are available on the Auto-GPT page on GitHub. Once you have these tools in place, click on ‘Code' on GitHub and download the Zip file.
To get started with Auto-GPT, the user needs to open a command-line programme like PowerShell. They will then need to clone the repository by typing "git clone https://github.com/Torantulino/Auto-GPT.git" into the command line.
After the repository has been cloned, the user must navigate to the project directory by typing "cd 'Auto-GPT'" into the command line. Once they are in the correct directory, they should install the required dependencies by typing "pip install -r requirements.txt" into the command line.
Finally, the user needs to rename the file called ".env.template" to ".env" and fill in their OpenAI API key. By following these steps, the user will be ready to use Auto-GPT. Once Auto-GPT is successfully installed, the user will need to name the bot and then give it a goal or set of goals to achieve.
Also Read: OpenAI Will Pay People To Report Vulnerabilities In ChatGPT
Essential Business Intelligence, Continuous LIVE TV, Sharp Market Insights, Practical Personal Finance Advice and Latest Stories — On NDTV Profit.