seurimas

I assume this alludes to their prior work with models running on smart TVs being able to setup inference on additional smart TVs. The idea being that you can't be sure that, having gone through the whole deployment process, your hardware has not been subverted to run a different model. Any level of "robust" code you can imagine to prevent this in the traditional ways could be subverted on the fly by a capable model.

> This moves unobservability layer to the system that collects the proofs.

I'd say it's adding security through a secondary system, rather than displacing security to a secondary system. Security by layers.

Not to the degree needed for actually verifying the things this verifies. You could probably quite easily produce a non-Claude model that generates text that conforms to Claude watermarking.

I don't think the education barrier fully accounts for SWE salaries. It takes a particular disposition and other hard things, learned over years of doing, to actually keep do software engineering as a career.

AI changes the dispositions and difficulties, but it still favors a narrow selection of people to do the (still) difficult task of software engineering with agents.

You converse through electrical signals and muscle contractions, but that will not make you an expert in either.

Similarly, a language model trained to emulate speech, follow instructions, and solve problems will not implicitly have an understanding of its own statistics. The research doesn't back that up and it just doesn't make any sense if you think about it.

The only thing LLM produce that could be considered its "confidence" would be its token logit values, but even that is a sketchy connection.

Positional encoding is standard for transformers of all stripes. They introduce a seemingly novel, redundant positional encoding scheme. It's more difficult to train, but seems to enable producing multiple tokens at once (i.e. you could get an answer that is N tokens long in N/x steps instead N steps).

Interesting! 25 business days for a business is close to 41 calendar days, too. I wonder if there's some sort of common, human constant involved. 25 days of engaging with something to decide whether that something is worth keeping around. Maybe the median relationship length is 25 days, too...

Is that a joke, or are you just misreading the stats on that page? Reddit definitely isn't down 30% of the time.

Usually, you pay what they ask and, if you don't like the price, you don't use their work. It's very simple, but OpenAI skipping that step opens them up to courts deciding that price, if a price is actually owed. That price would probably be more than 0, which is what the authors have gotten so far.

Doesn't that kind of demonstrate the value being actively stolen from the creators, more than anything? Copyright law killed Napster, too. That doesn't mean applying copyright law was wrong.

Well, we didn't mind the mold and mildew as much, for starters.

The features vs versions graph is more than a little nonsensical, and its conclusions even more so.

> They should be stable and move slowly so that most of the time of your develoeprs is not spent fighting with their tools.

I can't speak to Go's tooling base, but Rust's is head and shoulders above any other ecosystem that I've had to work with. This is definitely more a point in Rust's favor than against.

I had the wrong path for the demo file. It ought to work now.

Point in fact, those were the rules. The DIF protected them past 250k. This was not an exceptional measure. Everything worked as intended.

I think the greatest idea for Rust scripting is just supporting WASM scripts. I've had some success toying with wasmer in bevy.