Lu Nelson

Just How Bad Would an AI Bubble Be? The entire U.S. economy is being propped up by the promise of productivity gains that seem very far from materializing. By Rogé Karma

The capability-reliability gap might explain why generative AI has so far failed to deliver tangible results for businesses that use it. When researchers at MIT recently tracked the results of 300 publicly disclosed AI initiatives, they found that 95 percent of projects failed to deliver any boost to profits. A March report from McKinsey & Company found that 71 percent of companies reported using generative AI, and more than 80 percent of them reported that the technology had no “tangible impact” on earnings. In light of these trends, Gartner, a tech-consulting firm, recently declared that AI has entered the “trough of disillusionment” phase of technological development.

By one estimate, Meta, Amazon, Microsoft, Google, and Tesla will by the end of this year have collectively spent $560 billion on AI-related capital expenditures since the beginning of 2024 and have brought in just $35 billion in AI-related revenue. OpenAI and Anthropic are bringing in lots of revenue and are growing fast, but they are still nowhere near profitable. Their valuations—roughly $300 billion and $183 billion, respectively, and rising—are many multiples higher than their current revenues. (OpenAI projects about $13 billion in revenues this year; Anthropic, $2 billion to $4 billion.) Investors are betting heavily on the prospect that all of this spending will soon generate record-breaking profits. If that belief collapses, however, investors might start to sell en masse, causing the market to experience a large and painful correction.

The dot-com crash was bad, but it did not trigger a crisis. An AI-bubble crash could be different. AI-related investments have already surpassed the level that telecom hit at the peak of the dot-com boom as a share of the economy. In the first half of this year, business spending on AI added more to GDP growth than all consumer spending combined. Many experts believe that a major reason the U.S. economy has been able to weather tariffs and mass deportations without a recession is because all of this AI spending is acting, in the words of one economist, as a “massive private sector stimulus program.” An AI crash could lead broadly to less spending, fewer jobs, and slower growth, potentially dragging the economy into a recession. The economist Noah Smith argues that it could even lead to a financial crisis if the unregulated “private credit” loans funding much of the industry’s expansion all go bust at once.

It's seeming likelier and likelier that AI really could eliminate a lot of jobs very soon. Just not in the way boosters have suggested. www.theatlantic.com/economy/arch...

Ya think?

I've definitely had plenty of open tabs about l-tyrosine. Are you researching nootropics ?

I'm going to publish the entirety of an exchange I just had with the press office of the DOJ. I want you to see how they are talking about truth.

I once worked in a React codebase where the tech lead insisted on a policy of memoization, such that everyone was obliged to either tediously walk through the dependencies of every single component, or just defensively memoize everything. It was ridiculous and nightmarish.

Vibe code is legacy code Updates and articles from the Val Town team

Every time someone has asked me in the least year whether AI was going to take over all programming I’ve given some version of this argument. Now I can just link them to this 🙂.
Spoiler: no

blog.val.town/vibe-code

It turns out that "nobody" saw the Texas floods coming by design, this disaster has a culprit and their names are Donald Trump and Elon Musk. #3E #EndAutogenocide

I'm disappointed to find that Amazon does not seem to sell a razor-and-mirror kit for cutting lines of cocaine

I interpret this as a metaphor for the moment when we wake up from the collective hallucination that the LOTR movies are good

I would like to have a Git platform that has not "merge requests" or "pull requests" but "bung requests"

—like, "I just want to bung in this bit of code"

Diabolus Ex Machina This Is Not An Essay

A clear reminder that commercial chat LLMs —IMO, ChatGPT above all— are trained and tuned to engage you and keep you talking. Whether they might actually give you useful output is entirely secondary!
amandaguinzburg.substack.com/p/diabolus-e...

Accurate.

I reckon this has been happening in every industry associated with fun and “passion”—and will get even more hollowed out by AI. The next generation will need a different model of job satisfaction

A Ketamine Addict's Perspective On What Elon Musk Might Be Experiencing On Ketamine TW: Suicide

Fascinating and quite convincing
alisoncrosthwait.substack.com/p/a-ketamine...

This is happening right now.

www.theguardian.com/world/live/2...

It’s plain for all to see.

Impossible to communicate to people that once you decide that a group more vulnerable than you can be sacrificed in the name of your own safety or interests that you have already entered into an tacit agreement that you can also be sacrificed in the interests of a group less vulnerable than you.

It feels entirely possible that this a strategy of plausible deniability, in the form of plausible stupidity

Like that time when Canadian conservatives briefly renamed themselves the Canadian Reform Alliance Party (CRAP)

A "direct inverse correlation" between fluoride exposure and low IQ" means the more fluoride you get, the higher your IQ. What he's looking for is a "direct correlation," but he doesn't know what any of these words mean or how science works.

Let’s be clear. Trump's arrest of Judge Dugan in Milwaukee has nothing to do with immigration.

This latest attack is about one thing: unchecked power.

It is time for my colleagues in the Republican Party to stand up to his growing authoritarianism.

Donald Trump said: " I expect the legal profession to understand that the nation is not here for them, but they are here for the nation ...

From now on, I shall intervene in these cases and remove from office those judges who evidently do not understand the demand of the hour." JK it was Hitler

@t3.gg I'm curious whether you've considered offering data privacy guarantees as a premium feature on T3 Chat: AFAICT there are some inference providers that offer such guarantees, and to which you could ostensibly route requests?

Plato FTW: “One of the penalties of refusing to participate in politics is that you end up being governed by your inferiors.”

VS Code doesn’t support oklch() colors yet, but you can add support via this extension. It even includes a quick link to oklch.com.

marketplace.visualstudio.com/items?itemNa...

Kidnapping.

Capturing a person against their will without due process is called kidnapping.

Transporting them to another country without due process is called human trafficking.

A squalid extrajudicial prison for people found guilty of no crime is called a concentration camp.

I can only hope that they do

open.substack.com/pub/mikebroc...

Furthermore

bsky.app/profile/edua...

This simple and sober comparison implies something that I hadn't put my finger on before today:

Next.js often puts developers in to a mindset of complex orchestrations and optimizations, *before it is appropriate for their product* 🤦‍♂️

www.kylegill.com/essays/next-...