The appearance of ChatGPT last November has marked a before and after in the technological world, giving more visibility to generative AI models, which are showing how far they are capable of reaching in each new version that their developer companies go launching, to the point that important personalities are already asking to stop so many advances.

But it is no less surprising that there are people in the world capable of bringing the latest technology to devices that are already worthy of museum exhibitions. In this sense, Yeo Kheng Meng, an enthusiast from the world of retro computers, surprised the world this past Sunday by showing his feat of bringing ChatGPT to an old laptop running MS-DOS from 1984.

A feat full of technical obstacles

The team has a 4.77 MHz Intel 8088 CPU and 640 KB of RAM, which today we find laughable. Yeo wanted to go further by overcoming technical hurdles greater than the one he faced by releasing a Slack client to Windows 3.1, an operating system released in 1992.

Making use of the new ChatGPT API, Yeo knew how to face the lack of a native network, the lack of processing speed of the operating system, among other drawbacks, in order to be able to make his ChatGPT client capable of being functional in this old equipment.

It also took advantage of Open Watcom C/C++, a modern software compiler, running on Windows 11 64-bit, which is also capable of targeting 16-bit DOS platforms, while virtualizing MS-DOS via VirtualBox on Windows for testing. the client throughout the development, taking the resulting binary finished the process to the old laptop.

Source code available for third-party enhancements

Yeo can be satisfied with having achieved a better result than expected, to the point of having released the source code to be open to contributions from third parties that make it possible to improve it, where those interested can access it by searching for doschgpt on GitHub.

On this GitHub page, he also shows in more detail the technical challenges he had to face and how he did it, which serves as an experience to keep in mind for possible initiatives that may come from other enthusiasts.

This is a clear example that nothing is impossible, where sooner or later someone always ends up achieving something unusual and unthinkable from the start, and knowing how far enthusiasts can also go with time and patience.

Link/Image Credit: Source code on GitHub