A free online course called Modern-Day Oracles or Bullshit Machines? is trying to put structure around the public's rapidly changing relationship with ChatGPT-style systems. The source packet describes it as a humanities course built to help people learn how to work and thrive in an AI world, with short lessons that explain what large language models are, how they work and when they are worth using.
The course's pitch is straightforward. It does not treat AI as magic, and it does not treat it as pure hype. Instead, it argues that the tools are genuinely useful because they make computing accessible in ordinary language and can save people time in everyday tasks. That is a big part of why the technology has spread so quickly: people can ask questions, draft text, summarize material and interact with software without needing to know how to program.
But the course also says the same systems will add noise on a scale society has not seen before. The source page warns that AI systems will saturate the information environment with bullshit. That line is deliberately blunt, but it captures the dual view at the heart of the project. The point is not merely to celebrate progress. It is to teach users how to recognize when a model is helpful and when it is likely to steer them wrong.
The packet says the lessons are intentionally short, running about five to ten minutes each. That design choice matters because it makes the course easy to use as a quick reference rather than a formal semester-long commitment. The structure implies a practical objective: help people make decisions in the middle of real work, real research and real communication, not just in a classroom discussion. The course therefore functions as a basic literacy guide for a tool that is already embedded in daily routines.
The tone of the page is also worth noting. It does not call for panic, but it does call for skepticism. It frames AI as a major step forward in access to computing while also insisting that users should see through hype and know when a chatbot response deserves trust and when it needs verification. In that sense, the course is trying to solve one of the central problems of the ChatGPT era: not whether people will use these systems, but whether they will understand the limits of the answers they receive.
For readers who are already using AI or are being told they should, the course's appeal is that it offers a simple promise. It aims to teach not only what these systems can do, but how to avoid being misled by them. That is a practical skill set, and the source packet presents it as one that can be learned in small pieces without much overhead.
