> Zig is particularly irrelevant compared to Rust in the LLM era.

This. If we continue on this LLM/AI trajectory, and you're not constrained by your platform to use a specific language, why would you choose C or Zig or C++ over Rust? You won't be reading or writing the code yourself, LLMs will, and they benefit from a strong type system that catches a lot of issues, meaning fewer tests are needed.

Why can’t the LLMs avoid the memory issues entirely?

In the general case, there can be cases where it's simply straight up undecidable. (Think of a while loop going for as long as it finds a counterexample to insert your unsolved math problem of your choice) and call a free after. Sure, it's a very contrived example but similar "basically impenetrable barriers" do happen.

But the more common case is simply that each line of code gets some limited "thinking", whether it's inside the model's weights, a thinking block, or a whole turn. Only repeated turns can reason for arbitrary depths, so it's just not necessarily a good use of any of these buckets to spend on something we have a good and efficient algorithm for. Plus any of the former are stochastic which may accidentally ignore true issues (just like humans).

Because LLMs are statistical. They can, with a given probability, avoid memory issues, but cannot prove the non-existance of memory issues in the way that that borrow checkers and other formal methods can.

Are all programming languages which initially to be ergonomic to humans (Rust included) irrelevant in the "LLM era"?

All the programming languages that didn't already have enough inertia when Claude Code came out will be irrelevant.

But... but... they said about so powerful AGI/ASI, and turns out it needs "inertia"/support data and cannot improvise via any isoteric language or write directly machine code.

Poor AI

/s

Yeah, I think AI-generated assembly/machine code will be more and more common. The human-oriented programming language popularity ranks will be pretty much "fixed" from 2025 onwards.

No /s

> Zig is just trying to be a middle ground, which we've established doesn't work

We did?

Zig's C compiler is super handy for cross-compiling Rust

That's overly confident. When I see Rust code, I'm immediately reminded of Scala and how it was supposed to take over the JVM world, but did not, and Kotlin, a much simpler language, is getting more and more popular. You might claim that LLMs have no problem with that kind of language complexity, but people still need to review stuff. I have no doubt that Rust will reach C++ level of popularity, but that still leaves a lot of space for other low-level languages.

Scala "failed" for two reasons: Firstly, (like Haskell) it quickly exceeded its weirdness budget, and cultivated a community that embraced and sustained that weirdness. Secondly, a number of prominent individuals publicly expressed political views that made it hard for engineers in "mainstream" organisations to loudly promote Scala.

I think Rust is doing much better on both fronts. It has always been a more pragmatic language than Scala, and the way it is developed supports that, rather than pushing the weirdness budget higher. The overall political lean of the Rust community is also more amenable to support in mainstream institutions + companies.