Proof of Corn

You only want to apply expensive fungicide when there is a fungus problem. That means someone needs to go out to the field and check - at least today. You don't want to harvest until the corn is dry, someone needs to check the progress of drying before - today the farmer hand harvest a few cobs of corn from various parts of the field to check. There are lots of other things the farmer is checking that we don't have sensors for - we could but they would be too expensive.

If a neighboring farmer needs a bit of cash, has some land or equipment, and gets an email (or phone call!) from farmerfred@proofofcorn.com reading generally:

> I'm about to lease some acreage at {address near you} and willing to pay {competitive rate} to hire someone to work that land for me, are you interested?

I see no reason why that couldn't eventually succeed. I'm sure that being an out-of-state investor who doesn't have any physical hands to finalize the deal with a handshake is an impediment, but with enough tokens, Farmer Fred could make 100,000 phone calls and send out 100,000 emails to every landowner and work-for-hire equipment operator in Iowa, Texas, and Argentina by this afternoon. If there exists a human who would make that deal, Fred can eventually find them. Seth would be limited in his chance to succeed in these efforts because he can only make one 1-minute phone call per minute, Fred can become as many callers as Anthropic owns GPUs.

I do find it amusing that Fred currently shows the following dashboard:

    Iowa
    HOLD
    0°F
    Unknown (API error)
    Fred's Thinking: “Iowa is frozen solid. Been through worse. We wait.”

     Fred is here
    South Texas
    HOLD
    0°F
    Unknown (API error)
    Fred's Thinking: “South Texas is frozen solid. Been through worse. We wait.”

    Argentina
    HOLD
    0°F
    Unknown (API error)
    Fred's Thinking: “Argentina is frozen solid. Been through worse. We wait.”

Any human Fred might call in the Argentinian summer or 70F South Texas winter weather is not going to gain confidence when Fred tries to build rapport through some small talk about the unseasonably cold weather...

  > Hint: It doesn't work that way.

I mean... technically it would work this way but, and this is a big but, reality is extremely complicated and a model that can actually be a reliable formula has to be extremely complicated. There's almost certainly no globally optimal solutions to these types of problems, not to mention that the solution space is constantly changing as the world does. I mean this is why we as humans and all animals work in probabilistic frameworks that are highly adaptable. Human intuition. Human ingenuity. We simply haven't figured out how to make models at that level of sophistication. Not even in narrow domains! What AI has done is undeniably impressive, wildly impressive even. Which is why I'm so confused why we embellish it so much.

It's really easy to think everything is easy when we look at problems from 40k feet. But as you come down to Earth the complexity exponentially increases and what was a minor detail is now a major problem. As you come down resolution increases and you see major problems that you couldn't ever see from 40k feet.

As a researcher, I agree very much with you. And as an AI researcher one of the biggest issues I've noticed with AI is that they abhor detail and nuance. Granted, this is common among humans too (and let's not pretend CS people don't have a stereotype of oversimplification and thinking all things are easy). While people do this frequently they also don't usually do it in their niche domains, and if they are we call them juniors. You get programmers thinking building bridges is easy[0] while you get civil engineers thinking writing programs is easy. Because each person understands the other's job only at 40k feet and are reluctant to believe they are standing so high[1]. But AI? It really struggles with detail. It really struggles with adaptation. You can get detail out but it often requires significant massaging and it'll still be a roll of the dice[2]. You also can get the AI to change course, a necessary thing as projects evolve[3]. Anyone who's tried vibe coding knows the best thing to do is just start over. It's even in Anthropic's suggestion guide.

My problem with vibe coding is that it encourages this overconfidence. AI systems still have the exact same problem computer systems do: they do exactly what you tell them to. They are better at interpreting intent but that blade cuts both ways. The major issue is you can't properly evaluate a system's output unless you were entirely capable of generating the output. The AI misses the details. Doubt me? Look at Proof of Corn! The fred page is saying there's an API error[4]. The sensor page doesn't make sense (everything there is fine for an at home hobby project but anyone that's worked with those parts knows how unreliable they are. Who's going to do all the soldering? You making PCBs? Where's the circuit to integrate everything? How'd we get to $300? Where's the detail?). Everything discussed is at a 40k foot view.

[0] https://danluu.com/cocktail-ideas/

[1] I'm not sure why people are afraid of not knowing things. We're all dumb as shit. But being dumb as shit doesn't mean we aren't also impressive and capable of genius. Not knowing something doesn't make you dumb, it makes you human. Depth is infinite and we have priorities. It's okay to have shallow knowledge, often that's good enough.

[2] As implied, what is enough detail is constantly up for debate.

[3] No one, absolutely nobody, has everything figured out from the get-go. I'll bet money none of you have written a (meaningful) program start to finish from plans, ending up with exactly what you expect, never making an error, never needing to change course, even in the slightest.

Edit:

[4] The API issue is weird and the more I look at the code the more weird things are. Like there's a file decision-engine/daily_check.py that has a comment to set a cron job to run every 8 hours. It says to dump data to logs/daily.log but that file doesn't exist but it will write to logs/all_checks.jsonl which appears to have the data. So why in the world is it reading https://farmer-fred.sethgoldstein.workers.dev/weather?

The diagram looks correct for me when I disable CSS on the page or edit it's font-family to be "monospace". Seems like Geist Mono might just be borked.

If that were the case Claude would have come up with the idea to grow corn and it would have reached out to Seth and be giving Seth prompts. That's clearly not what happened though so it's pretty obvious who is leveraging which tool here.

It also doesn't help that Claude is incapable of coming up with an idea, incapable of wanting corn, and has no actual understanding of what corn is.