ChatGPT for PCB design

ChatGPT is good at the half of PCB design that is judgement, and structurally unable to do the half that is geometry. Knowing which is which saves a lot of wasted effort.

What ChatGPT does well

• +ChatGPT is genuinely useful for picking parts: describe what the circuit must do and it will suggest a regulator, a microcontroller or a sensor family worth considering.
• +It can explain a topology, sanity-check a power budget, or talk you through why a decoupling capacitor belongs close to a supply pin.
• +It can draft a rough netlist in text form, which is a reasonable starting point for a design you intend to verify yourself.

What it cannot do, and why

• −ChatGPT has no geometry engine. It cannot place components on a board with real dimensions, or produce copper that respects clearance.
• −It cannot route a multi-layer board. Routing is a search problem over physical space with hard constraints, not a text-generation problem.
• −It cannot run a design rule check. Without one there is no way to know whether the result is manufacturable.
• −It cannot output fabrication files. Gerber, Excellon drill and pick-and-place data are exact formats that a fab will reject if they are approximated.

The netlist looks finished. It usually is not.

A netlist is a list of which pin connects to which. It is text, so a language model produces it fluently, and it reads like the hard part is done. Language models draft netlists in our own pipeline every day. The findings below come from the models we run, not from ChatGPT itself, but none of these failures depend on which model you use. They come from asking any text generator to recall pin-level facts.

The same pin on two nets

In 4 of 6 boards we sampled, the model's netlist listed at least one pin on two different nets. That is a short circuit. The pattern was always the same: a power pin listed on its rail, then listed again on a signal net. On an ESP32 board it was the pull-up resistor, on both +3V3 and EN. Read as prose, each line looks reasonable.

Pinouts from memory

A battery charger IC, the MCP73831, came back as an 8-pin part with a "VPROG" pin. The real part is a 5-pin SOT-23-5. A wrong pin map produces a board that cannot work, and nothing about the text tells you so.

The value field names the wrong thing

Asked for an Arduino Nano soldered to a board, the model wrote the Nano's value as "ATmega328P", the chip the Nano carries. Anything that trusts that field builds the board around a bare 32-pin chip instead of the module.

Same prompt, different circuit

Temperature 0 does not make a hosted model repeatable. Two identical runs of the same prompt gave us 48 nets and then 50 on one board, and 13 and then 11 on another. If you re-ask for the netlist, expect a slightly different circuit, and check the one you actually build.

How to use ChatGPT for PCB design anyway

1. Let it choose. Describe the board, ask for parts and the reason for each. This is the judgement half, and it is good at it.
2. Ask for the netlist, then distrust it. Treat it as a draft.
3. Check every pin against the datasheet, not against the model's memory. Especially power, enable, reset and anything on a module.
4. Check that each pin is on exactly one net. A ten-line script does this, and it catches the most common short.
5. Hand the result to a tool that has geometry and rules: placement, routing, DRC and Gerber export are computed, not written.

The split that actually works

If you use ChatGPT for hardware, use it where it is reliable: talking through what the circuit should do, which regulator suits a 9 V input, whether a part is still in production. Treat the netlist it drafts as a sketch, because a connection list that reads well is not the same as one that matches the real pins.

PCBEditor keeps that division. The model proposes parts and reads your request; the connections are formed from each part's actual pin data, a deterministic router lays the copper, and a rule engine checks it. The board passes or fails the same checks whichever model planned it.