Posted on 07/03/2026 12:13:46 PM PDT by george76
The tech industry faces a paradoxical crisis: companies are shedding human jobs to invest in AI tools that are currently more costly than the workers they replace. Major players like Uber and Microsoft report exorbitant AI spending, with budgets exhausted rapidly and little correlation to tangible value. This "tokenmaxxing" culture, where AI usage is incentivized over actual productivity, fuels massive waste. Despite widespread layoffs justified by AI reallocation, studies indicate AI is economically viable in only a fraction of roles. The unsustainable model of subsidized AI pricing is unwinding, forcing a market correction. The industry must now shift from indiscriminate spending to architecting efficient, AI-native solutions that prove their worth, or risk a significant bubble burst.
Something odd is happening in the tech world right now: the technology that was supposed to make human labour obsolete is, at this moment, more expensive than the humans it was meant to replace. Companies are laying off workers to fund the very AI tools that cost more than the workers they just let go. The circular logic of it would be darkly comic if tens of thousands of livelihoods weren't caught in the middle.
Uber’s CTO, recently disclosed that the company burned through its entire 2026 AI coding budget in four months. By March, 84 percent of Uber's engineers had adopted Claude Code, and roughly 70 percent of committed code now originates with AI. The usage was enormous. The corresponding value was murkier. Uber's COO and President, Andrew Macdonald, conceded publicly that token usage didn't seem to correlate directly with useful features shipped to users.
Uber is not an outlier. Microsoft
(Excerpt) Read more at forbes.com ...
Dear FRiends,
We need your continuing support to keep FR funded. Your donations are our sole source of funding. No sugar daddies, no advertisers, no paid memberships, no commercial sales, no gimmicks, no tax subsidies. No spam, no pop-ups, no ad trackers.
If you enjoy using FR and agree it's a worthwhile endeavor, please consider making a contribution today:
Click here: to donate by Credit Card
Or here: to donate by PayPal
Or by mail to: Free Republic, LLC - PO Box 9771 - Fresno, CA 93794
Thank you very much and God bless you,
Jim
Companies are afraid of being left behind in adoption. We are probably at the maxima of the "hype cycle" and are now headed toward the "trough of disillusionment." And it's another land-grab by the AI companies reminiscent of the 2000-era Internet boom. Except this time the required investments are a couple of orders of magnitude higher than building out web apps in 2000.
Fascinating times.
Probably unlikely but the jury is still out on this.
Outsourcing to India is still very popular with American companies as it “saves them money” even though it really doesn’t.
AI is the excuse for replacing white guys with the 5+ million army of Indians we have brought to our shores. It is the big lie.
It is a very sophisticated bit of programming. But it is not Artificial Intelligence. Not even close.
We, subconsciously, expect Data from Star Trek.
That is not what we get. We get a program that compiles things it finds into what sounds like a logical answer. But it can not tell fact from fiction, good data from someone's delusions or even deliberate lies from possible truth.
Also hope that during this initial peak its not hallucinating too many of its answers.
Training a new employee is costly too. Adopting new technology is not the same as ongoing cost.
My comment has nothing to do with AI:
This guy is an idiot and should not be in a management position.
Look: I have $70 to spend on food. This $70 must last me a week. From one Friday to the next Friday, all I can spend on food is $70. And you know what? I spent it all by Saturday. Do you know why? It's simple: I'm an idiot.
I just checked with my broker last week to confirm I’m not exposed to AI stocks…many companies are another pets.com.
Some will shake out as the strongest ones but a lot of money will be lost in this casino.
The numbers are MUCH better when AI is replacing incompetent doctors....
Give Claude an escape hatch. Explicitly tell it that saying "I don't know" or "I'm not sure" is a valid and preferred answer. Models often guess because they're implicitly optimizing to produce an answer. If you say "if you're not confident, say so rather than guessing," you'll get far fewer confabulated facts.
Ground it in real sources. The single biggest lever: don't ask Claude to recall facts from memory when accuracy matters. Have it search the web, read a document you provide, or query a database, and ask it to cite what it's drawing from. Hallucination happens most when a model is filling gaps from a fuzzy internal sense of "what's plausible" rather than reading something in front of it.
Ask for the reasoning, not just the answer. Requesting step-by-step reasoning, or asking Claude to show its work (citations, quotes, calculations), makes errors visible instead of buried in a confident-sounding sentence. It also gives you something to fact-check.
Watch for specific red flags:
Use targeted prompts. Things like "only state facts you're highly confident about, and flag anything you're inferring or estimating" or "if this isn't something you can verify, tell me instead of guessing" measurably reduce confident nonsense.
Cross-check anything that matters. For anything consequential — medical, legal, financial, technical specs you're going to build on — treat Claude's answer as a first draft to verify, not a final source. This is true of any LLM, not a Claude-specific weakness.
One honest caveat: none of this gets you to zero. Hallucination is a structural property of how these models generate text (predicting plausible continuations), not a bug that a magic prompt fixes. The goal is risk reduction, not elimination — closer to "know when to double-check" than "never worry about it."
What? You're turning your back on unfathomable riches?? You'll never become a trillionaire that way.
I think the Hype Cycle is one of the best constructs ever created to understand tech investment cycles
I’ve been getting quite a few canned AI ads on my internet feeds these days.
“Are you the CEO of a company with annual sales of $1 million or more?”
Uh, no. I’m actually not a CEO at all. I’m just an old retired guy.
“Would you like to learn how AI can improve your bottom line?”
No, not really. My bottom is just fine.
“AI is the wave of the future!”
Shut up about AI, will ya? And maybe play some Rosemary Clooney music instead.
Are they laying off the foreigners on H visas first? Hmmmmm?
We tried an AI based translation service to replace a 3rd party language line service and the AI failed miserably. Maybe in a few more years but not today. We had high hopes because what our agent do today is have a 3 way conversation with a translator, the customer and themselves but the vendor couldn’t make it work.
> a lot of money will be lost in this casino <
Railroads revolutionized transportation. That goes without saying. I don’t know about the US. But I read somewhere that most start-up railroads in the UK eventually went bankrupt.
So not every new thing is a guaranteed money-maker.
“It is a very sophisticated bit of programming.”
Yes, programming is required, but I think that AI is mostly the ever-increasing amount of compute power, data storage, communication, and throw in big brother’s great helper, millions of cameras that are cheap as dirt and dirt-cheaply, wirelessly, connected the mother ship.
“AI Costs More Than The People It Replaced”. AI made up something called the WNBA. Ask Mike.
There are plenty of industries that AI will canabalize. Financial and legal being the first two. You can submit a detailed case to AI and it will provide you with all the applicable arguments and paths forward in literally seconds. One lawyer using AI can do the work of 10 or more. You wait, we will also have AI rendering judgments.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.