chis

I'm not an expert on this. But isn't it the case that C++ code could have errors that span the entire codebase, like a setup in file A triggered by a bug in file B which is immensely far away on the import graph? A classic would be a use-after-free. To me that's the thing that Rust can help with, even if silly bugs aren't being written by AI.

The other thing is just that rewriting some old human-written codebase in Rust probably immediately catches many bugs. It would be hard to prompt the AI to properly scan for such bugs itself, they're lazy when working in that modality.

Well presumably now it’ll fall back to Sonnet 5.5 lol

This is a great approach. I remember even pre-AI having issues where, at a big state school, basically all the answers to homework could be found online. Sometimes I made the choice to copy a homework and tell myself I’d catch up later. But the classes where there was a scheduled quiz right away just forced me to keep up.

What if a piece of software were to exactly emulate a human brain. Would it be still be “just software” by your classification? What if a piece of software acted 20% like a human and 80% like an algorithm, where would that land?

They didn’t use a chatbot. It’s worth reading the article

What do you mean by this exactly, or have any sources? That's a bit cryptic

HN won't agree but that's a perfect example of an ethically grey product. It can be used for good, but you can easily trick such an AI into doing cyber attacks. Which again, maybe that's good! But other companies wouldn't be willing to risk their brand like that

The only edge Meta has at this point is their willingness to take risks and make unsafe, ethically grey AI products. I don't even mean this as some sort of anti-corporation hate speech, just an honest analysis. Their brand is so different from all the other big tech cos that they are in a unique position.

You can ask Meta Muse to take actions that clearly break other site's terms of service and it happily does it. I asked it to bot poker games and it just hopped right in to a table.

Personally I didn't find that guy very convincing. I think current frontier AI would already probably enable a non-expert to do gain-of-function work on a virus which could result in something like Covid.

And AI looks to keep getting better, and if it does then I don't see why it wouldn't become superhuman in virus design too.

I gave up on fable 5 after I asked it to critique my PR and it spit out a page of complete nonsense technical jargon. Like, to the point that I had to review the feedback with other models and try to parse what it was saying and ultimately it wasn't even right. Compare to Astra and Sol where I can almost forget there's a model and just speak/read naturally.

I think I should give 5.1 another chance but I am just so triggered by the way it talks after spending so long battling fable 5.

Also I'm starting to wonder if the latest round of models have finally saturated for my personal coding needs. I mean obviously not for taste and judgement, but those barely seem to improve with model generations. For just spitting out a 1000-line feature I've vaguely scoped out, Astra feels basically as good as I need.

> I don’t think AI is going to usher in an extinction event. In fact, even if nobody were to slow down, I really don’t think humanity would have much to worry about.

I mean sure. If you feel that way then your p doom is zero, and it makes sense to worry about things like market concentration or losing the fun of software engineering.

Chinese AI companies are a lot smaller in terms of capital, GPUs, and headcount as compared to American. The fact they’ve kept close so far is super impressive

GDP growth and inflation are different things. If inflation-adjusted GDP goes up, that means the average person is able to buy more things they want. Compare living in the US vs India today for the median citizen

Even at the scale of 100 people there is a coordination problem. If the 50 most conscientious researchers quit, then the 50 left behind would be the more aggressive group and be less interested in AI safety.

Personally I agree though. I would not be able to work on AI model development right now as I just don’t think it’s ethical. Fortunately for OAI/Ant I lack the relevant skillset anyways.

Nvidia is limited by the number of engineers they have. It might turn out that the AI market is so lucrative that it’s best to reallocate their gaming-focused engineers to AI.

Apple TV is at least a growth market for them, whereas gaming is sort of capped and clearly a tiny piece of nvidia’s revenue atm.

HN comments are just on a different planet lately lol

Fantastic answer thank you

> Opus

Are there cybersecurity concerns in the frontend? I would have thought you have to assume the client is untrusted and only do security work on the backend

The OG pros honestly sucked and had horrible reliability. The other models have treated me well

The connection type has nothing to do with it though. Modern bluetooth is basically transparent with 256kbps AAC, like you'd have to be an omega-expert listening in a silent room to identify it vs lossless. Certainly the difference you're noticing is just the headphone quality.

Honestly for me the Airpods 4 with ANC are one of the greatest feats of audio engineering ever. For such a tiny form factor they reproduce really great sound, big soundstage, AND they have good ANC without needing a closed earbud seal. They also have internal mics that listen to the response of your ear to adjust the sound profile, which you can hear happen in real time after bumping the earbuds. Obviously you can get better sound if you make different tradeoffs and wear over-ears, but they are absolutely cutting edge for what they are.

I think you really have to squint to call this a disproof lol

Certainly it's not possible, yeah. A better test might be to have the AI report either confirmed good, confirmed toxic, or unknown. Even an experienced mushroom hunter needs to humbly put many mushrooms into the "unknown" category.

what are you smoking

Sometimes I’m amazed anyone is willing to live without a git ui. The graph structure git uses is just such a natural fit for visual processing, plus the default git ux is borderline user-hostile until you master it.

What’s also funny is that early in my career I could tell other engineers looked down on me for using a git UI, perhaps assuming I was too stupid to use the terminal. I suspect that this mindset propagates git use to the next generation.

You’re still stuck paying for a 3 year contract for deprecating hardware. And the 256gb model which is still not enough is $230 a month. And it runs models 8 months behind frontier

This is just super unlikely to occur in the near term compared to some of these other risks. It's not like an instance of fable could just introspect into itself and pull out the weights. Model weights are stored encrypted and are highly protected, considering that they're targets for corporate and state espionage.

So reading more I guess the judge narrowly ruled that destructive scanning is legal because of the one to one replacement, but didn’t rule on whether or not non-destructive scanning would be legal. It’s still not tested if scanning in that way would count as transformative fair use.

Whatever judge ruled that it was legal to scan and use books if and only if you DESTROY a copy of the book was truly a moron. The law gets bent by judges all the time to be practical and fair, they should have used this opportunity to give a more sane ruling.

I think Fable is the beginning of Anthropic switching to training models as agent-first, tool second. It’s certainly the best model if you want something to work autonomously without supervision and don’t care to read the code. The code and writing is ugly but it can complete huge tasks and fix its own work.