OpenAI’s first piece of hardware is here, and it’s not a smartphone or a smart home gadget — it’s a $230 macropad called Micro, built with boutique keyboard maker Work Louder. The device is a six-key programmable pad designed to control ChatGPT and OpenAI’s coding tool Codex, complete with a dedicated voice dictation button. The packaging is sleek and, as multiple reviewers have noted, aggressively Apple-coded. The hardware itself is substantial enough to moonlight as a paperweight.
But the question hanging over Micro isn’t about build quality. It’s about who this is actually for. OpenAI is pitching the device at heavy ChatGPT users and developers, but the early reception from that exact crowd has been chilly. Reddit threads are littered with dismissive takes — one user called it “a prank and not a real product” — and the independent outlet Aftermath titled its review “OpenAI’s expensive macropad feels engineered to piss me off specifically.” The core complaint: cheaper DIY macro pads and off-the-shelf alternatives do the same thing without the learning curve.
That learning curve is real. Users need to memorize what specific ChatGPT sessions are mapped to which frosted “agent” keys, and then decode a color-coded light system where white means idle, blue means thinking, green means done, and red signals an error. The dictation feature — hold a button, talk, tap send — is genuinely convenient once you’re set up, and toggling between projects with a physical button does feel efficient. But it requires a week of programming and memorization to get there, which raises an obvious question: why not just use your keyboard and mouse?
The whole thing lands at an awkward moment for OpenAI’s hardware ambitions. Apple is suing the company over alleged trade theft, and reports are swirling about a smart home device built by former Apple engineers. Micro, by contrast, feels like a niche experiment — a fun, clicky desk toy for the ChatGPT power user with money to burn, and a head-scratcher for everyone else.
