Re:DeepSeek: X2 Speed for WASM with SIMD
by dboreham on Tuesday January 28, 2025 @12:33PM
(#42855244)
I've wondered about this too. The LLM could just write machine code. But now a human can't easily review it. But perhaps TDD makes that ok. But now the tests need to be written in a human readable language so they can be checked. Or do they? And if the LLM is always right why does the code need to be tested?
Re:DeepSeek: X2 Speed for WASM with SIMD
by nkozyra on Tuesday January 28, 2025 @01:53PM
(#42856337)
At a certain point I don't see why a human needs to be in the loop at all. But I suppose that's the most dystopian part of it all.
Maybe the human has the money.
The LLM might be terrible at writing machine code directly. The kinds of mistakes I see GPT-4 making in Python, PostScript, or JS would be a much bigger problem in machine code. It "gets confused" and "makes mistakes" in ways very similar to humans. I haven't had a chance to try DeepSeek R1 yet.
"Development" is effectively translating abstractions of an intended operation to machine language.
What I find kind of funny about the current state is we're using large language models to, like, spit out React or Python code. This use case is obviously an optimization to WASM, so a little closer to the metal, but at what point to programs (effectively suites of operations) just cut out the middleman entirely?