A Japanese tech blogger and engineer used Google’s Gemini AI bot to help him design an x86 motherboard for a 1987-era Intel CPU, the chip that spawned the x86 architecture. The project showed that generative AI is best when augmenting human reasoning, as Ikejima relied on his own logic and expertise to complete the task.
Initially, Ikejima used Gemini to assist with circuit design, which was fine-tuned by hand using KiCad and Python scripts. He also asked the bot to help create a base software for the Raspberry Pi Pico microcontroller, which acted as a control, debugging, and memory interface for the V30 chip.
The project had its share of setbacks, including mistakes in circuit design that could have caused electrical issues. However, with Gemini’s suggestions, Ikejima was able to correct these errors and eventually successfully boot up the system with MS-DOS’ COMMAND.COM.
Ikejima’s use of Gemini highlights the importance of combining human expertise with AI assistance, as both are necessary for complex tasks like this one. While Gemini proved its limitations in certain areas, it ultimately helped Ikejima complete a challenging project that demonstrated the potential of generative AI when used effectively.
Source: https://www.tomshardware.com/raspberry-pi/raspberry-pi-projects/tech-tinkerer-gets-gemini-to-help-him-design-an-x86-motherboard-from-scratch-bot-help-was-impressive-but-project-still-required-human-awareness-and-intervention