vntok

Hackers rejoice!

How so? If it's not local, it's not yours, a third party owns it.

Why would you expect that third party specifically trained their model to be more aligned to you and your needs than to them and their (business) needs?

Nah we don't do this "communicate" thing in here.

From the same guidelines:

Please don't post insinuations about astroturfing, shilling, brigading, foreign agents, and the like. It degrades discussion and is usually mistaken. If you're worried about abuse, email hn@ycombinator.com and we'll look at the data.

Dishwasher? I hardly know her!

This issue does not come from the coding assistant. In fact, humans will occasionally do the exact same thing as long as their tools have enough permissions to deploy to both local and prod from the same environment.

Fix it by separating the tools (different non-interconnected VMs, etc for dev/qa/preprod/prod environments) and the permissions (different accounts, sessions, tokens, etc for the run/debug/test/deploy loops).

I'm running QubesOS, my various qubes reboot about a dozen times a day.

Indeed, the more accurate way to say it is that people in the US don't care enough about mass school shootings to do something about it besides thought and prayers.

That's even more expensive.

You must have very unrepresentative customers. What will they use?

Thank you for your service.

The cooling towers are right there.

Wait till you learn the bank routinely uses CSV as a non-ironic exchange format! That definitely is stochastic.

Look up "prepared statements", it's pretty well documented.

Spreading malware to your website's visitors is wild and illegal in most jurisdictions. I certainly wouldn't confess about it online.

Well in addition to what you wrote, the marketing manager ALSO wasn't tracking any ad-related marketing performance indicator (CTR, CR, etc.) in any measurable way for very long periods of time... or they would have caught it almost immediately ("wow ad spend, CTR and CR have all suddenly gone down to 0/0% and have been staying there for days on all our campaigns! What's up with that?").

His restraint alone is commandable.

His team is basically him and two other humans, powering an ambitious well-known project so successful an industry titan ended up acquihiring him/them. That's pretty lean, no?

Well, for starters the program you wrote is wrong (very unreliable) 100% of the time (very predictable)... so you just got your answer I guess.

In any case, most -if not nearly all- of the top-100 LLM will answer your prompt with some code that does what you intended the first program to do. Only they'll actually code it properly of course.

Are LLMs that super reliable in their output already with all the guardrails around?

Well, what is your definition of "super reliable in the output", and is it a quantifiable/measurable target or just a feeling?

Is it "more than humans", "more than senior developers", "almost perfect", "perfect"?

It might behave differently than specified and a human is required to validate every output carefully or else.

Sure, just like meatbag developers. All the security flaws AI finds today were introduced years/decades ago by humans and haven't been found (that we know) by humans in ages.

Or maybe you can buy some stuff while visiting on holidays?

Ad blockers do absolutely nothing against fake reviews, fraudulent claims, sponsored articles and influencers manipulating you into buying stuff.

But they do have wildly successful products.

They also have failures, then again most companies have failures as well at all points in the product cycle.

Definitely comes over as salty. Naming major flaws has been a tradition for decades. Remember Heartbleed? It had a site and a logo :) Shellshock, Meltdown, Spectre as well. A few more: https://github.com/hannob/vulns

This site though is pretty useful; first it serves as a central location to point people to with short links in chats/emails/whatever, then it has a quick visual explainer and a link to the detailed technical report for those who want more info. Pretty neat.

Last but not least, buying the domain must have taken 5 minutes, prompting the page must have taken 30 minutes and posting it on HN must have taken 1 minute. So it certainly wasn't a lot of work in the grand scheme of things and probably did not deter the team from doing other important things.

How is this relevant? Human General Intelligence has a lot of limitations as well and we have managed to do lots.

If your Worker starts returning 503s at 02:00 UTC, finding the root cause: be it an R2 bucket, a misconfigured route, or a hidden rate limit, you’re opening half a dozen tabs and hoping you recognize the pattern. Most developers don't have a teammate who knows the entire platform standing over their shoulder at 2 a.m. Agent Lee does.

But it won’t just troubleshoot for you at 2 a.m. Agent Lee will also fix the problem for you on the spot.

Want to hack a site? Just bombard it with security scans at 2 a.m, causing a sudden increase in errors, then lean back and wait a few minutes for the Cloudflare Agent to helpfully disable the security rules for you to drive through.

Say you study some piece of software. And it happens that it has an automated suite of tests. And it happens that some files aren't covered by the test suite. And you happen to find a bug in one of those files that were not covered.

Would you publish a blog post titled "the XXX test suite proved there was no bug. And then I found one"?

It would be a bit silly, right?

Yes, and "his obsessive drive for perfection" as you put it is what would make him "rolling in his grave if he could see the software quality of the products that Apple releases today" as the parent put it.

The counter-hacker double-keyboarding sequence was inspiring.

From the thread:

Q: Why the heck did you hyperlink [the malware installer]?

A: If someone reads this and they still click the download then they kind of deserve the virus tbh