Linux Shell written based on OpenAI GPT3 (film, 5 minutes)
On River's Educational Channel, a new video was released where the author shares his excitement about receiving the OpenAI Private Beta API key. He had been waiting for this package for months and is now planning to use this AI power to make command line tasks in Linux easier. His idea revolves around creating an AI-powered shell that would accept user commands and figure out the necessary commands to execute. He predicts that this could save him several minutes from his workflow each week, which could be a significant advantage for AI in its quest to revolutionize human-computer interaction in operating systems.
The author notes that OpenAI provides a simple Python library to access its API, which ultimately boils down to making basic HTTP requests. In the video, he puts this process to the test with various examples that showcase how helpful AI can be in daily tasks. For instance, a person who often uses tools they are not very familiar with might struggle to remember how to run them correctly, and OpenAI can guide the user toward the appropriate commands.
Moreover, the video demonstrates how OpenAI can fill in missing details in queries. The author illustrates this with an example of formatting disks, where necessary information is automatically filled in. Viewers will surely appreciate this functionality as a useful aid. Additionally, the AI can generate commands that involve multiple steps—important for more complex tasks. The author acknowledges that due to the stochastic nature of OpenAI’s models, users might receive different results for the same queries, adding an element of excitement to the process.
Unfortunately, the author is unable to share his project widely due to limitations on using the OpenAI API and charges for each request. Currently, running one example costs about one cent. Nonetheless, he encourages viewers to leave suggestions in the comments for commands they would like him to try. He recommended subscribing to his channel to see how he handles their requests in future videos. In terms of statistics, the video's numbers are impressive: at the time of writing this article, it has garnered 38,564 views and 1,844 likes, indicating strong interest in the topic.
River's Educational Channel appears to be a source of engaging content related to technology and programming innovations. Through the exploration of OpenAI’s capabilities, the author is opening new perspectives for those looking to enhance their productivity when working with Linux. This is surely not the end of his journey with AI, and upcoming videos are likely to bring even more inspiration to the tech community.
Toggle timeline summary
-
Introduction and excitement about a new delivery.
-
Revealing the OpenAI Private Beta API key.
-
Discussing the first project idea: an AI-powered command line shell.
-
Explaining how the AI shell will simplify command execution.
-
Anticipating time savings in the typical workflow.
-
Mentioning OpenAI's Python library for API access.
-
Discussing challenges with unfamiliar tools in command line.
-
Highlighting AI assistance for online tasks.
-
The AI's ability to fill in unspecified details.
-
Demonstrating handling of multi-step commands.
-
Commenting on the AI's tendency to hallucinate results.
-
Humorous observation about unexpected results.
-
Explaining the stochastic nature of OpenAI's models.
-
Expressing a desire to share the tool online but facing limitations.
-
Noting the cost of using the AI model per command.
-
Inviting viewers to suggest commands for future videos.
-
Encouraging subscriptions for upcoming content.
Transcription
Hello, everyone, and check out what I just got in the mail today. I've been waiting for this for months. Enclosed is your OpenAI Private Beta API key, blah blah blah. And here's the API key. Gotta make sure I blur it out. So I've been thinking about what to make now that I have this incredible power, and the first idea I had was making the command line easier to use. So I got to work. Basically, the idea is that instead of constantly googling for how to accomplish various tasks on the command line in Linux, I want an AI-powered shell. I'll tell the shell what I want to do, and the shell will figure out the actual command to run. Over the course of a year, this could save entire minutes from my typical workflow and could be a huge win for AIs in their quest to obsolete humans. As you can see, there's not a whole lot of code that needs to be written. OpenAI provides a Python library to access their API, but it's really just doing simple HTTP requests underneath the hood. Now, let's try it out on some basic stuff. Often, I need to use a tool that I'm not super familiar with, so it's hard to remember how to run it, especially with converting or working with specific file formats. OpenAI can help out with that. And, it can help us with stuff online, too. Even better, if we don't specify some details, it'll fill it in for us. Note how we didn't need to specify which file system type or even which hard drive to format. It's super helpful. It can also figure out commands that require multiple steps. Well, A for effort, I guess. OpenAI's model also has an incredible ability to hallucinate URLs. You can just ask it for something, and it'll imagine where you might be able to download it. Did OpenAI just Rickroll me? The best part of all this is that OpenAI's models are stochastic, which means that there's some randomness in the result. You might get different commands for the same request, and I find that that really keeps things fresh and exciting while I'm working. I wish I could put this online for everybody to try out, but I would need approval from OpenAI to let more than five people use it, and there's a charge for each request. The examples that I download cost about one cent each to run with their most powerful model. But if you have a command that you want me to try running, let me know in the comments below. I want to make a follow-up video with those requests, so subscribe if you want to see it when it comes out. Thanks for watching.