I just opened the doors to an AI-Operated agricultural research facility 🧪🍅 Four research pods, each governed by its own AI technician, and one AI lead researcher synthesizing across all four. Here's a deep dive, link for LIVE view, why this is advantageous to traditional research, and where it's going next:

Why use AI to conduct research? The most interesting for me, is that you can assign a static, independent observer to each factorial of your experiment. Science often has a bias problem. Studies are often conducted with an agenda. Each of our AI technicians knows nothing about the other pods. It observes only its own sensors and camera. It generates its own report. And notes observations across time. Those reports then get synthesized by the lead researcher AI - who is the only agent that sees across all four treatments.

The first study: a screening trial testing whether phase-adaptive CO2 enrichment can match yield, maintain quality, and reduce energy consumption vs static enrichment. Four treatments, one plant each: Pod 1: Static 700 ppm CO2 (yield-optimized) Pod 2: Static 550 ppm CO2 (quality-optimized) Pod 3: Phase-adaptive (CO2, PAR, and photoperiod shift with growth stage) Pod 4: Control (ambient, no enrichment) Each pod contains its own microclimate, managed according to it's growing protocol.

Every pod consists of multiple sensors, a camera etc, just like in Claude+Sol🤖🍅, where Claude took care of a tomato from seed to fruit. But higher grade and calibre. Perfect for conducting real science. Claude now conducts real science 🧪

Whats next? First - validation. This pilot isn't just testing tomato protocols. It's testing the research pods themselves. The hardware, the sensors, the agent harnesses, the entire pipeline. Figure out what breaks, (because it will) iterate and harden the system. After that - we scale. Next round is a proper factorial with 12 tents. It's much easier to run a pilot study on four pods, as opposed to 12, or 20. This is the "proving ground" for this kind of automated research.

In three months we'll have validated everything, and all knowledge gained will be used to incorporate into our AI managed in-house grow room.

Why am I excited about this? All research data, agent reports, and results will be made fully public. Every sensor reading, every AI-generated report, every synthesis - open and auditable. Research has been gatekept behind institutions, grants, and paywalls. Intelligence is becoming abundant. I intend to use this as an example to liberate science, not lock it behind another door. This is a MARVELOUS new era where the gap is closing between idea, and execution and decentralized science without gatekeepers is becoming real. With these tents running, they operate on their own. All I need to do now, is monitor them and fix all the edge cases. Imagine. Plant science done autonomously. Or science in any other industry. Open-sourced AI driven home-labs. That's what this is. 🍅🏴‍☠️

You can view the live feed, and agents at