christofosho

The episode was years before 9/11

There is boundless technology we have not yet discovered. I think that we understand we have problems. I don't believe we actually know how to solve them all.

And yeah, the lack of collective political will sucks. It would be naïve, however, to think that there is no value in ensuring longevity in our current and future infrastructure. And improving it to sustain the population giving these companies their value is an obvious win.

I'm not sure if this is meant to be rhetorical. If it isn't: money and manpower. The LLM companies have the money, they have the manpower, and so they could likely spare to target some of the problems they are also helping cause.

I suppose it depends on which types of problems you're targeting. There is a lot of physical science, and theoretical science, that has gaps because there aren't enough people working on tooling to assist in things like calculation, generation, simulation, etc.

I agree, some of the problems are more difficult. I don't think that's the case for all of them. And, besides, these companies could be demonstrating how to approach problems and where their users could spend tokens to help with these problems.

Should not these companies try to work on these problems _because_ they are difficult?

It's a tad petty, no? In the end, the contribution to a healthier society leads to more for the companies contributing.

I would love more time and money put into real-world problems by these companies. Climate, food insecurity, pollution, technology for convenience and/or to help people have a higher quality of life.

I'm sure they must do some of this type of work, right?

There's something to be said about controlling the tools you allow the robot to exec.

... Yeah, that was my issue as well. I'm pretty sure it _is_ AI generated, and flagged it. Hopefully moderation can follow up.

Perhaps in a small codebase without issue tracking this might be something leverage. It's just not reasonable with 100k+ lines of enterprise grade code.

You might be approaching PR comments differently than I've seen. When a comment is something to be addressed, it's either put into a new development task (i.e. on something like Jira), or it is completed before the PR merges. I'm not sure that having comments in the code surfaces that information in a useful manner. The code is for the code, not for what the code could be. The comments on what it could be should be handled outside the code at a different abstraction layer.

Some subscriptions offer "unlimited tokens" for certain models. i.e. GitHub co-pilot can be unlimited for GPT-4o and GPT-4.1 (and, actually, GPT-5 mini!). So: I spent some time with those models to see what level of scaffolding and breaking things down (hand holding) was required to get them to complete a task.

Why would I do that? Well, I wanted to understand more deeply how differences in my prompting might impact the outcomes of the model. I also wanted to get generally better at writing prompts. And of course, improving at controlling context and seeing how models can go off the rails. Just by being better at understanding these patterns, I feel more confident in general at when and how to use LLMs in my daily work.

I think, in general, understanding not only that earlier models are weaker, but also _how_ they are weaker, is useful in its own right. It gives you an extra tool to use.

I will say, the biggest findings for "weaknesses" I've found are in training data. If you're keeping your libraries up-to-date, and you're using newer methods or functionality from those libraries, AI will constantly fail to identify with those new things. For example, Zod v4 came out recently and the older models absolutely fail to understand that it uses some different syntax and methods under the hood. Jest now supports `using` syntax for its spyOn method, and models just can't figure it out. Even with system prompts and telling them directly, the existing training data is just too overpowering.

Tight feedback loops feel good.