AI: The ROI Runway Could Be Long Outside the Tech Sector • Hacker News Board

Re: AI: The ROI Runway Could Be Long Outside the Tech Sector

#91

Post by eru » Wed, Jul 08, 2026, 7:53 AM UTC

rurufjcir wrote: post #84

eru wrote: post #66

Earlier quoted context omitted.

> If AI eats profits instead of grows them, then the economy shrinks right? No, why? The economy is bigger than company profits. Eg there's workers' wages and customer surplus and investments etc.

Because companies need earnings growth and they’ll fire workers to meet those targets

Firings workers doesn't make earnings grow in general.

And company don't _need_ earnings growth. That's why many companies buy back shares or pay dividends instead of always investing retained earnings.

Re: AI: The ROI Runway Could Be Long Outside the Tech Sector

#92

Post by DrewADesign » Wed, Jul 08, 2026, 7:39 PM UTC

eru wrote: post #90

DrewADesign wrote: post #78

Earlier quoted context omitted.

Yeah that’s why health care and credit are so cheap! How about literally anything run by private equity?

> Yeah that’s why health care and credit are so cheap! Yep, though only where competition is allowed! > How about literally anything run by private equity? Like many airlines? Flying has never been cheaper. Planet Fitness is also owned by private equity, and their 10$/month membership plan has been steady for two decades despite inflation. You can find lots more examples.

It certainly didn’t seem to stop enshittification in any meaningful way.

Re: AI: The ROI Runway Could Be Long Outside the Tech Sector

#93

Post by eru » Thu, Jul 09, 2026, 4:22 AM UTC

DrewADesign wrote: post #92

eru wrote: post #90

Earlier quoted context omitted.

> Yeah that’s why health care and credit are so cheap! Yep, though only where competition is allowed! > How about literally anything run by private equity? Like many airlines? Flying has never been cheaper. Planet Fitness is also owned by private equity, and their 10$/month membership plan has been steady for two decades despite inflation. You can find lots more examples.

It certainly didn’t seem to stop enshittification in any meaningful way.

If that's what customers pay for, that's what they get.

Re: AI: The ROI Runway Could Be Long Outside the Tech Sector

#94

Post by eru » Thu, Jul 09, 2026, 4:23 AM UTC

disgruntledphd2 wrote: post #79

eru wrote: post #75

Earlier quoted context omitted.

Yes. Though with Excel for most business users the choice was between - lobby the IT department for at least a quarter, then wait at least one quarter, and at the end you get a buggy implementation of your idea that doesn't quite work - or: spend a weekend hacking together a quick and ugly, buggy spreadsheet prototype of your idea that doesn't quite work.

> - or: spend a weekend hacking together a quick and ugly, buggy spreadsheet prototype of your idea that doesn't quite work. I mean, based on my own experience with AI tools, this feels like the standard output.

Yes, exactly. And spreadsheets took the world by storm. Deservedly though.

Nowadays people (allegedly) buy a Mac Mini to run 'claw'. Back in the day, they bought a mac to run a spreadsheet.

Re: AI: The ROI Runway Could Be Long Outside the Tech Sector

#95

Post by greggsy » Thu, Jul 09, 2026, 10:45 AM UTC

eru wrote: post #64

greggsy wrote: post #28

Earlier quoted context omitted.

Anecdotally, Claude Code has prompted an explosion of open source projects and prototypes from self-starters. A lot of these are just hobby projects, but some of them genuinely fill a niche that was previously too complicated or unviable to develop otherwise. Some of them have half baked financial models, but nobody will invest dollars backing a SaaS offering that could easily be replicated, or that could be made red…

> [...] but nobody will invest dollars backing a SaaS offering that could easily be replicated, or that could be made redundant tomorrow. Matt Levine wrote in his newsletter Money Stuff of some investment fund that has their employees vibecode replacements for software of potential investment targets. I guess the theory is exactly what you say: if the internal employees can replace the target's software in a few hour…

There are still support benefits in commercial offerings, and not everyone can vibe code.

The trash I ask Claude to whip up for my weird edge cases is certainly increasing in reliability lately though.

Re: AI: The ROI Runway Could Be Long Outside the Tech Sector

#96

Post by p1necone » Fri, Jul 10, 2026, 11:55 AM UTC

p1necone wrote: post #57

christophilus wrote: post #54

Earlier quoted context omitted.

Hm. I must be holding it wrong. I hit $20 easily in 5 hours with deepseek in opencode.

Same here - curious which deepseek model and what op was doing. I suspect it might be a question of conversational loop vs agentic dev - the former uses much less tokens than letting an autonomous agent churn away on your codebase.

My apologies, deepseek v4 flash through pi is indeed super cheap, I can see only using $1 in a day and getting a decent amount done.

Re: AI: The ROI Runway Could Be Long Outside the Tech Sector

#97

Post by DrewADesign » Fri, Jul 10, 2026, 1:53 PM UTC

eru wrote: post #93

DrewADesign wrote: post #92

Earlier quoted context omitted.

It certainly didn’t seem to stop enshittification in any meaningful way.

If that's what customers pay for, that's what they get.

And where’s the reasonable alternative that competition has provided to many of these enshittified services? They don’t exist because nickel-and-diming people to avoid misery on services they require is far more profitable, and charging a gigantic luxury premium to have everything you want… and don’t want, is even more profitable than that. The system can still be complete shit while providing some remnant of its original functionality. Without proper regulation, the whole thing just devolves into n-opolies hell-bent on extracting as much as they can from everybody else. Well, that’s what we’ve got in many markets.