After a while you pick up on the warning signs and just bail early without any guilt about false positives. It's really the only sustainable strategy in a world where it takes 5 seconds to absorb 5 minutes of your attention span.

It's a familiar example of the perennial "[THING] could be solved overnight if [PERSON_OR_GROUP] would just start taking [THING] seriously" trope.

The damage would come much later, well beyond the point where it could be simply pulled out and replaced without spending massive amounts of money and would also basically necessitate training an entire new generation of engineers.

Then the AI giants would start appearing vulnerable like cigarette companies in the 90s while an AI Superfund and interstate class action are being planned but Sam Altman would already be a centitrillionaire at that point so it would be someone else's problem.

I use AI tools daily (because they feel like they're helping me) but it's not exactly hard to imagine scenarios where an explosion of slop piling up plus harm to learning by outsourcing all thinking results in systemic damage that actually slows the pace of technological progress given enough time.

History of new technologies tend to average into a positive trend over a long enough time scale but that doesn't mean there aren't individual ups and downs. Including WTF moments looking back at what now seems like baffling decision-making with benefit of hindsight.

(Not the exact same chart but similar idea, I guess it's sort of a meme: https://imgur.com/a/YrNGYOR)

So I looked at the most recent CC release notes on Github and the majority look like this:

  Fixed /clear not resetting the terminal tab title after a conversation
  Fixed session title chip from /rename disappearing while a permission or other dialog is active
  Fixed agent panel below the prompt being hidden when subagents are running (regression in 2.1.122)
  Fixed external-editor handoff (Ctrl+G) blanking the conversation history above the prompt
  Fixed /context dumping its rendered ASCII visualization grid into the conversation, wasting ~1.6k tokens per call
  Fixed OAuth refresh race after wake-from-sleep that could log out all running sessions
  Fixed 1-hour prompt cache TTL being silently downgraded to 5 minutes
  Fixed cache-miss warning appearing spuriously after /clear or compaction when changing /effort or /model

But hey, all that churn looks great on charts being circulated on social media as free advertising for their flagship product (and consequently the company's valuation) so never mind, LGTM!

They could even use one of many headless CMSs combined with a static generator. Claude Code in the hands of non-technical users deploying to prod regularly seems like one of the worst possible ways to do it (except for the "cool" value telling people about it).

At my company the internal devs don't even have access to wherever the company site is hosted, it's a WordPress CMS and marketing can make updates safely with a couple clicks and zero day-to-day development oversight required. IT just helps keep the box updated but otherwise it's entirely their own thing.

But last month I picked it up again and it has crushed everything I've thrown at it. As Codex limits tighten on the Plus plan it's been my main fallback and doesn't even feel like a downgrade when I switch over. Haven't hit a single loop so far using it nearly every day for several weeks so that problem seems solved finally, thank god.

I've been using it in the auto router mode and haven't felt the need to manually lock in the bigger model yet. It's incredibly snappy which I realized I really appreciate vs. waiting around endlessly for minutes each turn, but I've read other people's experiences needing to manually select the Pro model so YMMV.

Off the top of my head, I'd guess the message is closer to an observation about being disconnected from history in the modern world leading to vaguely defined feelings of angst and alienation.

  > Agents (if implemented well) are an order of magnitude more effective at resolving issues compared to a call centre worker who is reading off a script and churn within 9 months

But it seems like companies don't want to do this part, possibly because of fears that someone will trick the agent into giving them a refund or something. Or because the actual goal is to optimize for fewer costly refunds/cancellations/policy exceptions etc.

So for whatever reason, they stay stuck in that useless local maxima while simultaneously making traditional help increasingly difficult to get ahold of when needed for an overall net worse experience as a customer.

After they killed that and then stopped handing out free model access to users of every Cline fork for weeks following model releases, vibe coder hype moved back to Chinese models for cost and the SOTA models for quality.

https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...

https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

But at the same time.... I had been doing nearly everything the iPhone could do in terms of raw functionality (plus plenty of stuff that took 1+ years to land on iPhones) on multiple different Windows Mobile and Palm smartphones pre-iPhone.

Saying pre-iPhone smartphones don't count because "ugly nerdphone with gross keyboard" is just as ridiculous as a "iPhone was overhyped and no better than existing smartphones" claim.

Apple created a device category within smartphones that then consumed and became what we now think of as a "smartphone" after iPhone and Android together strangled the first movers.

Like, the famous Steve Jobs "an iPod, a phone, an internet communicator" line was just listing standard smartphone features by that point. More or less the definition of a smartphone in fact.

And, lest you think generating "600,000 lines of production code in 60 days" [2] is potentially problematic, has also fully solved the primary failure modes of AI coding identified by Andrej Karpathy, once and for all: "Karpathy's four failure modes? Already covered." [1]

As someone who has experienced mania, including with a programming bent specifically, it's hard not to raise an eyebrow at the idiosyncratic human-y bits of his thinking floating up from the sea of em-dashes and it's not X it's Y in his manifestos.

Plus volunteering this [3] in an interview:

“I sleep, like, four hours a night right now,” he told his interviewer, fellow VC Bill Gurley, during an onstage interview Saturday. “I have cyber psychosis, but I think a third of the CEOs that I know have it as well,” he joked about his current AI obsession. (Tan’s assistant confirmed to us that he was joking. ...)

It’s like I was able to re-create my startup that took $10 million in VC capital and 10 people, and I worked on that for two years, and I took anti-narcoleptics — I remember, you know, sort of being on modafinil...

[1] https://github.com/garrytan/gstack

[2] https://github.com/garrytan/gstack/blob/main/docs/ON_THE_LOC...

[3] https://techcrunch.com/2026/03/17/why-garry-tans-claude-code...

Like, thanks, really useful stuff (and definitely worth the creepy vibes to include that).

It sounds like Github Actions is the first choice, if it's unavailable (or if Github blocks GhostBox in the future), are each of the alternatives viable as a more or less drop-in replacement? Or would there be loss of functionality?

Those are the questions I had when reading through the site so I think some basic technical docs would go a long way to help people understand the project and decide to give it a try. I like the cute/whimsical branding but I'll admit to doing a little internal eye-roll when I clicked that link expecting technical specifics and instead read:

  > GitHub Actions is only the first place ghosts come from. There are strange little pockets of temporary compute all over the internet. Ghostbox makes them feel like one small machine. 

[1] https://www.ghost.charity/#gft

  > They were always in harm's way. The war could have waited, and Iran could have doubled or tripled its missile stockpile and then they really would have been in harm's way. 

The missing ingredient has never been how many missiles Iran has stockpiled, it was external military action from someone like the US that gave them the window to assert that control.

The US didn't do the world any favors by getting it out of the way sooner or something, that's just absurd apoligism for a poorly planned war of choice that has obviously been a net negative for basically the entire world.

It would be like if the US nuked China and then shrugged after they predictably retaliated saying it just proved the threat from their stockpile that had always existed.

Previously a quick scan of comment history would make it obvious you're looking at an LLM, now you're stuck arguing over a one off comment where they can get away with benefit of the doubt.

  > Importantly, journalists in media, classically inept at any economic analysis, implied that 10% tariff = 10% RRP rise. They never corrected themselves, nor for the economists who falsely claimed the economy would collapse.

It's also pretty vague without any examples of what specifically deserves corrections.