But I guess I also don't fly much, and I never had to deal with delays or rebooking with them.

Though again, making no judgment as to whether or not it should be allowed. I just think it should be a law, and not casting modern values on the 1700s era founders' words.

If we don't want the government to be able to do that, we should pass laws to that effect.

We're being told that Mythos is such a big step change in capability that it needs to be kept secret and carefully controlled because a wide release could threaten cybersecurity everywhere. That does not really hold water if a barely simpler harness can do the same stuff at a lower price and is available to all of us.

The burning question to me, at least, is how many false positives each approach generated, and the degree of their falseness (e.g. "valid but not exploitable" vs. "not valid"). It's not super useful if it's generating way more noise than signal.

That's not too surprising for those of us who have been working with these things, either. All kinds of simpler use cases are manageable with harnesses but not reliably by LLMs on their own.

I think you're misrepresenting what they're doing here.

The Mythos findings themselves were produced with a harness that split it by file, as you noted. The harness from OP split each file into chunks of each file, and had the LLM review each chunk individually.

That's just a difference in the harness. We don't yet have full details about the harness Mythos used, but using a different harness is totally fair game. I think you're inferring that they pointed it directly at the vulnerability, and they implicitly did, but only in the same way they did with Mythos. Both approaches are chunking the codebase into smaller parts and having the LLM analyze each one individually.

[1] https://en.wikipedia.org/wiki/NAT64

I guess what I'd really like is a middleware box or something that I could put on my home network, but would then still give the same user experience as the normal app. I don't want to have to log into some web interface and manually add firewall rules after I find something not working. I like the pop-ups that tell you exactly when you're trying to do something that is blocked, and allow you to either add a rule or not.

I'm probably straddling some gray area between consumer-focused and enterprise-focused feature sets, but it would be neat.

Though I guess if that would work, they'd just use those loud suction toilets they use on airplanes.

That said, I'm sure there's gotta be a better way to solve it with less copper. And I think they did something like that with CyberTruck.

I think it factors into why public perception is increasingly anti-AI. It'd be one thing if people were losing jobs, but on the other hand, their daily chores were done by a robot. Instead, people are losing (or fearing losing) their jobs, while increasingly having to fight with AI chatbots for customer support and similar cost-center use cases.

It's like AI is the "high fructose corn syrup" of tech. Nobody's arguing the output is better--it's just a lot cheaper and faster to get there, so that's its legacy. Making things cheaper and worse.

I was essentially told that I could update the mailing address but going through the steps for that process would result in the name on my account being changed to the legal name. And so today, it still has my parents' mailing address. Thankfully they haven't moved.

An example of a poorly thought-out implementation elsewhere does not exclude the possibility of coming up with a better one than humans coordinating with their mouths over radio.