parliament32

AI slop, I'm sure the contents are interesting but after a few paragraphs of the LLM "tone" I gave up.

I'm still daily-ing a Timbuk2 from a decade ago and it looks fantastic, no issues whatsoever.

> There seems to be some implicit feeling that everything ought to be getting better and cheaper than it used to be.

But don't we see this everywhere, all the time? Pull up any of the recent Claude Code threads about the product's declining quality and you'll see at least a handful of well-upvoted comments about how text generators are definitely going to get cheaper while simultaneously getting "better" over time.

AI hype -> layoffs -> AI underperforms -> ????

Hilariously, it's the exact same playbook as the big third-world-country-outsourcing hype from a few years ago.

I presume it's from here:

> Notification Center shows your notifications history, allowing you to scroll back and see what you've missed.

https://support.apple.com/en-ca/108781

Note that although Android has a similar "notification history" feature, it's disabled by default and requires opt-in.

There's a fallacy here around how software is fungible. WordPress hasn't made web developers obsolete, despite everybody having access to a $5/mo WYSIWYG-and-domains-and-hosting-bundle environment; quite the opposite, in fact.

I'm seeing the parent's point along these lines: "me and all my friends are starting businesses being the middlemen between WordPress and (people who want websites)". It's not that it won't work, it's just a shit business model.

Only a small sliver of the world has to worry about health insurance. Job security, maybe.

I think the biggest component is all the crap that comes with running a business.. accounting, sales, budgets and planning, regulatory concerns, office/site management, the list goes on forever. I'm an engineer, I want to do this and leave the other jobs to people who specialize at those, not run around trying to spin a dozen plates at once. I'm sure there's a tidbit more money to be made but it's just not worth it for me.

Now, if someone can make a vibe-business platform where AI handles all the drudgery and I can stick to the tech.. that might be worth talking about.

I think a closer analogy is paying for a personal trainer vs working out yourself. Some people find value in that, but not many.

How could your "business" ever make money if any idiot with a $20 CC subscription can recreate it in a weekend? And no, "I can prompt better than them" is not a differentiator.

We (the two of us) do fine with 95% Costco shopping, and I have similar "she won't commit to eating something specific for a week" restrictions. The only real tricks are get better at storage[1], and get better at cooking[2]. FWIW, the 2% rebate on the executive membership always covers my membership renewal price plus $50-$100 off a shop.

[1] Yes, you need somewhere to stuff 24 rolls of paper towels etc. I ended up building more shelves in the voids at the top of closets and the like. Ladder-access only but it works out.

[2] Stop doing 32-ingredient cooks with baby bok choy and white tomatoes and whatever other exotic instagram reels crap. Buy beef, chicken, fish, then portion and freeze (a vacuum-sealer is not necessary but it helps). Buy a standard 4-5 vegetables and a couple fruits. Potatoes and rice for carbs. Then figure out a list of recipes you can make from those ingredients -- I promise they can be combined damn near indefinitely.

Your shed saga is a fantastic writeup, thank you for sharing.

Good checklist, but

> The actual development happens on a rented server

Why not Hyper-V or libvirt/KVM? VM escapes aren't a thing in real life (or VMs from hyperscalers wouldn't exist), so why deal with additional cost, latency, and third-party trust when you could just run it yourself?

Both the article and repo[1] are slop.

[1] In the repo, the "reproduce" is just a bunch of print statements about what would happen, the bug isn't actually triggered: https://github.com/juxt/agc-lgyro-lock-leak-bug/blob/c378438...

The wall of slop after the single human paragraph, you mean? Text generator output isn't data.. it's at best unreliable, and at worst entirely fabricated.

Internally, we measured a regular developer consuming about $40/workday of compute on the $200/mo plan thresholds.

> There's no such thing as a "slop vulnerability"

https://daniel.haxx.se/blog/2025/07/14/death-by-a-thousand-s...

See the list at the bottom of the post for examples.

It was never about cyber capability. It's a liability transfer framework.

If a service provider has a control that says "we use firewalls on all network access points, and configure those firewalls to CIS benchmark whatever", and a third-party signs off with "yes we checked, they have the firewalls, and they're configured properly", you now have two parties you can sue when a security incident caused by lack of firewalls causes you material damage.

Your org's cyber insurance will also go down if you can say "all our vendors have third-party attested compliance, and we do annual compliance reviews".

> This is surprising given the excellent capabilities of GPT-5.2

The real surprise is that someone writing a paper on LLMs doesn't understand the baseline capabilities of a hallucinatory text generator (with tool use disabled).

Powered by what, exactly? Neither lugging around a few tons of fuel or painting a bullseye on yourself via solar panel array sound practical.

This wasn't immediately obvious to me, but it's important to note this unit is remotely controlled. The article made it sound autonomous. Further, the unit went back to base nightly (for maintenance / battery swaps I assume).

https://devdroid.tech/en/catalog/droid-tw

Emitting a shrug and "AI made me do it" is cheaper.

Google doesn't have to fight context windows. They can cache and store an AI response to a Google query without having to worry about much other than locale etc. You can't do that a dozen messages into an LLM conversation.

But longer sessions means more queries, which means more costs, which makes the problem even worse, right?

How could this possibly work? Google is profitable because they can insert ~4 ads into a search query. An LLM query costs about 2 orders of magnitude more resources to run than a Google search, so.. I'm not seeing it unless OAI can figure out how to shoehorn 400 ads per prompt into the interface somehow.

If anyone is replaceable by AI, executives are first in line. Make "decisions" based on expert input, give presentations, sit in meetings and on calls. No liability, no concrete "work product" to speak of, so why not?

> can anyone on the product side actually predict traffic

Hypothetically, could you not? If you engineer a bridge you have no idea what kind of traffic it'll see. But you know the maximum allowable weight for a truck of X length is Y tons and factoring in your span you have a good idea of what the max load will be. And if the numbers don't line up, you add in load limits or whatever else to make them match. Your bridge might end up processing 1 truck per hour but that's ultimately irrelevant compared to max throughput/load.

Likewise, systems in regulated industries have strict controls for how many concurrent connections they're allowed to handle[1], enforced with edge network systems, and are expected to do load testing up to these numbers to ensure the service can handle the traffic. There are entire products built around this concept[2]. You could absolutely do this, you just choose not to.

[1] See NIST 800-53 control SC-7 (3)

[2] https://learn.microsoft.com/en-us/azure/app-testing/load-tes...

In the midst of the 4chan v. Ofcom civil suit? Interesting.

https://www.courtlistener.com/docket/71209929/4chan-communit...

You remind me of those guys who swear they have a "system" at the casino.

Fascinating read. What's curious though, is the claim in section 2.3.0.1:

> Each task runs in its own sandbox. If an agent crashes, gets stuck, or damages its files, the failure is contained within that sandbox and does not interfere with other tasks on the same machine. ROCK also restricts each sandbox’s network access with per-sandbox policies, limiting the impact of misbehaving or compromised agents.

How could any of the above (probing resources, SSH tunnels, etc) be possible in a sandbox with network egress controls?

..so they copy/paste your message into Claude and send you back a +2000, -1500 version 3 minutes later. And now you get to go hunting for issues again.