Atomic Machines 用 AI 造微型器件：首款产品 PrimeSwitch 与“Matter Compiler”
ChatGPT for hardware? Atomic Machines is using AI to build tiny devices, with a bigger ambition… describe an invention and have a factory manufacture it. Its first product, PrimeSwitch, is a fingertip-sized power switch the company says opens in 50 microseconds while conducting electricity efficiently through metal. Behind it is the “Matter Compiler,” a manufacturing system that builds and tests components while generating data to improve future designs. Engineers still move parts and help assemble devices, so this isn’t fully autonomous manufacturing yet. But the goal is something like ChatGPT for physical inventions, where your prompt could eventually become a machine that works.
Atomic Machines 正用 AI 制造微型器件，首款产品 PrimeSwitch 是 fingertip 大小的电源开关，公司称其 50 微秒内开启并能通过金属高效导电。其背后的“Matter Compiler”制造系统在建造和测试组件的同时生成数据以改进后续设计。目前工程师仍需搬运零件并协助组装，尚未实现完全自主制造，但目标是让描述一项发明就能由工厂造出可工作的机器。
粉丝 8.6万
The latest in AI and robotics, explained Founder & CEO of The Rundown AI Zurich 🇨🇭 / NYC 🇺🇸
- 浏览
- 1842
- 点赞
- 122
- 评论
- 2
- 分享
- 0
- 收藏
- 28
数字来自平台，为最近一次抓取时的数据。
以下是视频口播的自动转写，可能有错字。
So this is a new startup building tiny devices with ai and its goal is for you to describe an invention and have a factory build it for you.
It's called atomic machines and its first product is a power switch small enough to sit in your fingertip. That sounds simple enough until you understand the trade-off.
See conventional mechanical switches conduct electricity efficiently but take milliseconds to open silicon switches respond much faster but waste electricity.
When generating heat atomic says it's prime switch carries current through metal and opens in just 50 milliseconds. But the bigger story is how it's made.
The company calls its manufacturing system a matter compiler. Basically, it's ai learns from thousands of physical experiments, setting how materials behave when heated, combined and assembled.
Those designs feed into a chain of 30 machines that build and test the components, and each build generates data that can help improve the next one.
The ambition is to produce different microscopic devices by changing the code, potentially enabling tiny pumps and motors for things like surgical robots.
That said, there still is a major gap between that vision and today's current factory. Engineers still have to move components between machines and help assemble them themselves.
But the goal is something like ChatGPT for physical inventions. You describe what you want, and instead of getting an answer, you get an actual machine.
If you want to stay in the loop with the latest in ai and robotics, follow for more.
来源：TikTok · Rowan Cheung · tiktok.com