这个案例展示了AI如何让非技术背景的人也能实现农业自动化，做农业科技或关注AI落地场景的开发者、农民和创业者值得一看，可以直接借鉴思路。
日本北部一位从未学过农业的西兰花农民Hiroki Tomiyasu，利用OpenAI的Codex和ChatGPT构建了定制工具，自动化管理100公顷的西兰花、南瓜、大葱和大豆农场。他通过Codex搭建了温室控制系统，用文本指令控制通风口升降，并在农场群聊中设置机器人管理操作。他还用ChatGPT拍照识别作物病害，实时判断是否需要干预，并利用卫星植被数据分析每块地的需求。甚至让ChatGPT绘制自制控制箱的接线图，Images 2.0生成了带日文注释的完整图纸。Tomiyasu形容体验“就像身边总有一位超天才工程师”，展示了AI在农业领域的实际应用潜力。
"I hired one engineer for the farm. His name is Codex." Awesome profile from @OpenAI on Hiroki Tomi...
"I hired one engineer for the farm. His name is Codex." Awesome profile from @OpenAI on Hiroki Tomiyasu, a broccoli farmer in northern Japan who automated his own farm with AI. Tomiyasu never studied agriculture, but runs 100 hectares of broccoli, pumpkins, green onions, and soybeans using custom tools built with Codex and ChatGPT. Some of the coolest use cases: - Built a greenhouse control system with Codex to raise and lower vents via text, plus a bot in the farm group chat to manage operations - Takes photos of crops to identify diseases, with ChatGPT triaging whether they need intervention right in the field - Pulls live satellite vegetation data onto maps of his actual fields to analyze and decide what each plot needs - Asked ChatGPT to draw the wiring diagram for his homemade control box, and Images 2.0 returned one fully annotated in Japanese Tomiyasu says the experience "feels like having an ultra-talented engineer always by your side." You can just build things 💬 11 🔄 2 ❤️ 12 👀 1116 📊 11 ⚡