Everyone will share in the wealth

 By Nick Anderson

Class of '26

By Rick McKee 

The rat about to destroy civilisation

 By Pedro Molina

AI plumber

 By Australian cartoonist Matt Golding

Solar and wind replacing the Hormuz gap

 From This is Not Cool

This seems like Good news.

Center for Research on Energy and Clean Air:

Global power generation from fossil fuels fell in the first month since the start of the Hormuz closure, with the fall in gas-fired generation offset by large increases in solar and wind power, rather than coal.

The power generation dataset prepared for this analysis covers countries that disclose near-real-time data. The dataset covers 87% of global coal power generation and over 60% of gas-fired power generation.

Total power generation from fossil fuels in countries with near-real-time data fell 1% year-on-year, with coal-fired generation flat and gas-fired generation falling 4%. The dataset covers the world’s largest power markets: China, the U.S., the EU, and India, among others.

Seaborne coal transport volumes fell 3%, to the lowest levels since 2021. The data contradicts widespread expectations that coal power generation would rise in response to the crisis.

This is the first oil crisis where we *have* alternatives.  We can replace imported gas with wind, solar and storage.  We can replace petrol and diesel vehicles with EVs.  What's more, *everybody* knows it.  Governments, companies, individuals.  

Global emissions have peaked.  Oil demand has plunged, and only some of that demand is coming back, and then only in the short term.  How ironic that this is thanks to Trump.

So yes, emissions will rise again, but the next peak will be lower than this one.

(Caveat:  So-called "AI" data centres.)

Have you seen The Matrix

 By Drew Sheneman

Elon Musk, stud

 By Bill Amend

Elon's X-ray specs

By Ben Jennings 

The coming AI crash

 From Owen Jones, talking to Professor Steve Keen, who correctly forecast the GFC.   He reckons the current AI boom will fizzle out within a year, to be followed by an AI bust as AI takes over more and more jobs.  He points out that a UBI will be essential, or people will starve to death, and there will be severe civil unrest, a dystopian "Hunger Games" scenario.  A new Great Depression, leading to massive economic and social change.

World PMI very sluggish

This is my calculation of world manufacturing PMI, compared with J.P. Morgan's calculation.  I only started keeping the J.P. Morgan data in 2011, which is why I needed to make my own calculation to understand previous cycles.  Where I don't have back data for individual countries, I have used manufacturing business confidence, and estimated what each country's PMI would have been if it had been calculated by IHS Markit (which used to publish PMI data before S&P Global took over.)  In some cases, I have smoothed the input series (some, such as ABSA's PMI for South Africa, or the AIG PMI for Australia or Canada's Ivey survey, are very "spiky", i.e., have large month-to-month random errors.)  In other cases, I have extreme-adjusted the series before I used them to calculate my estimate of world PMI.  This mostly, in effect, reduced the down spike from COVID, but had some small effects elsewhere.

Why this chart is interesting is because, hitherto, in all recoveries, from deep recessions or shallower slow-downs, the rebound has been sharp.  This cycle, it's been a slow, and not especially steady ascent.  Observe that it actually began a steep-ish recovery at the end of 2023, before it fizzled out.

Obviously, Trump's tariff tango has something to do with this, but I suspect there's more to it.  Inflation isn't falling like it should be when the economy is so sluggish, and part of the reason for that is the growth of monopoly and oligopoly is the US, and the West's determination to stop China exporting its deflation to the world, particularly in cars, solar panels and batteries, via tariffs and quotas.  Why was inflation lower before Covid, when manufacturing was just as concentrated as it is now?  Because everybody expected inflation to remain low.   But in the Covid rebound, firms found that they could indulge in a bit of "greedflation", and pushed up their margins, and expectations have accordingly shifted.  Monopolies and oligopolies now know they can shove up prices every year by more than they used to, and get away with it.  To use more technical terms, inflation over the last few years has been more cost-push than demand driven.

Sluggish growth may well continue, even though Europe is clearly (finally) recovering.  But higher inflation means that Central Banks will be reluctant (=slow) to cut interest rates.   And if the AI bubble pops, the US will go into recession.   If that happens, the US dollar will plunge, pushing other economies themselves into slow-downs or recession.  

Of course, happy days may be here again.  But I hae me doots.

Use of AI suddenly drops

 

From Futurism

After three years of unprecedented tech spending and nonstop hype, the demand for AI in the workplace seems to be drying up fast.

Referencing data from a recent US Census Bureau survey, The Economist estimated that the percentage of Americans using AI to “produce goods and services” at large companies rang in at a modest 11 percent in October, the latest available survey date. It’s not just that the figure is a bit soggy for the supposedly world-changing technology, but that it’s suddenly moving in the wrong direction: the financial publication notes that the percentage is actually down from 12 percent in the prior survey, conducted two weeks previously.

Looking at the big picture doesn’t make it any prettier. Back in March, the number of businesses with 100-249 employees that reported not using AI within the last two weeks stood at 74.1 percent. The survey results show a steady uptick in “no” results over the past few months, culminating in a dreadful 81.4 percent as of the latest poll.

For big corporations with over 250 employees, meanwhile, the “no” reports have crept up to 68.6 percent, up from the year’s low of 62.4 percent recorded in February.

The data is nothing if not a major red flag for an industry which is expected to spend $5 trillion on AI infrastructure between now and 2030. To do so will require a massive increase in revenue from both business and personal AI use — the latter of which has been lagging.

Unfortunately for the tech industry, enterprise AI customers aren’t picking up the slack. Though various non-government surveys cited by The Economist varied wildly in their numbers, they all seemed to spell out the same results: AI remains more of an experimental plaything in the workplace than a serious driver of productivity.

One economist at Stanford who tracks the use of generative AI at work found a major drop in usage month to month: though 46 percent of respondents reported using the tech in June, that number had fallen to 37 percent by September. Another estimate, by Fintech firm Ramp, found that AI use at American corporations went through the roof earlier in 2025 to around 40 percent, but has since plateaued.

The results follow a disappointing summer for AI advancements, with models like OpenAI’s GPT-5 falling short of expected performance gains. Still, the cracks in enterprise AI adoption had begun to show as early as December of 2024, when an EY pulse survey of 500 senior executives found over half felt they were “failing in their role” of supporting AI in their companies.

Instead, executives pointed to a creep of “AI fatigue” among the rank and file — which a year of AI horrors probably hasn’t helped.

With a $600 billion gulf between AI revenue and AI spending, an immense amount is riding on whether the tech can start bringing home the bacon.

My wild and woolly guess:  it's not going to bring home the bacon.    And that's bad news for the US economy, because most of the rise in spending recently has been on AI.

Will world interest rates fall further?

This chart shows the GDP-weighted world central bank discount rate (bank rate/Fed Funds rate/central bank lending rate) covering 83% of the world economy, as well as the median of world interest rates that I monitor . The median is the point in a series of data values at which half the observations are above and half below.  It is unweighted.  

The two different kinds of average interest rates usually move in more or less the same direction, though not always.  The GDP-weighted average will be dominated by the largest economies, the median is skewed more towards smaller economies, because they are more numerous.  Note how the median interest rate rose much more than the average before the 2009 GFC, likely making the recession worse; and how the jump in the median rate in 2011-2012 signalled that smaller economies were in trouble, worsening the downturn linked to the euro crisis.

Right now, the median is falling faster than the GDP-weighted average, which is consistent with my "small 15" average PMI index, which has been much weaker than the "big 8" index.  

Will interest rates fall further?  Well, yes, but not by much.  The ECB (European Central Bank) seems happy with its bank rate; the Fed is muttering about not cutting rates again; and the "small 15" PMI is rising fast, meaning smaller economies have become more reluctant to cut rates.  At the same time, world inflation has levelled off after falling from the post-Covid highs.  What is certain is that the low interest rates of the 2009 to 2021 years will not be reached again in this cycle.  Unless the AI-bubble pops .....

Clicking on the chart will make it easier to read.

Is the US economy just one big AI bubble?

 An interesting and, frankly, scary video from TLDR.

Why aren't markets freaking out?

 From Paul Krugman

For those of us who follow economic policy in general and the Federal Reserve in particular, the past week has been shocking and terrifying. Donald Trump’s ongoing attempts to bully the Fed into large interest rate cuts have escalated into an attempt to fire Lisa Cook, a member of the Fed’s Board of Governors, over unsubstantiated claims that she committed financial fraud while still a college professor. Indeed, Trump claims that he has already fired her, although he has no legal right to do so.

Whatever happens, Trump’s campaign to take over monetary policy has shifted from a public pressure to personal intimidation of Fed officials: the attack on Cook signals that Trump and his people will try to ruin the life of anyone who stands in his way. There is now a substantial chance that the Fed’s independence, its ability to manage the nation’s monetary policy on an objective, technocratic basis rather than as an instrument of the president’s political interests and personal whims, will soon be gone.

So why aren’t markets freaking out? Nations in which central banks lose their independence sooner or later suffer high inflation, especially when they are taken over by autocrats who buy into crackpot economic doctrines. And Trump, who has been demanding large rate cuts because, he claims, the economy is running hot — which almost every economist would say is a reason to raise rates, not cut them — certainly fits that pattern. Yet although there have been small tremors in the bond and currency markets, there have been no significant upheavals in financial markets that reflect the severity of the situation we are in. Throughout this episode, the stock market has remained fairly flat and bond yields haven’t spiked.

Why not? Do financial markets doubt that Trump will get his way? Or do they reject mainstream economics and the clear examples of countries like Turkey and Argentina?

Neither. My read of economic and financial history is that market pricing almost never takes into account the possibility of huge, disruptive events, even when the strong possibility of such events should be obvious. The usual pattern, instead, is one of market complacency until the last possible moment. That is, markets act as if everything is normal until it’s blindingly obvious that it isn’t.

The inimitable Nathan Tankus summarizes this by saying that the market is not, as stylized economic models would have us believe, a mechanism that pools the knowledge and informed judgment of millions of investors. It is, instead, a “conventional wisdom processor.” That is, it reflects views that seem safe to hold because many other people hold them — and the crowd only abandons those views when they become blatantly unsustainable.

John Maynard Keynes said something similar in Chapter 12 of his General Theory of Employment, Interest and Money. Market investors, he argued, pay little attention to the question of what assets are truly worth. Instead, they worry mostly about the market value of those assets a few months in the future. In a memorable albeit sexist passage (it was 1936), he declared that:

"Professional investment may be likened to those newspaper competitions in which the competitors have to pick out the six prettiest faces from a hundred photographs, the prize being awarded to the competitor whose choice most nearly corresponds to the average preferences of the competitors as a whole; so that each competitor has to pick, not those faces which he himself finds prettiest, but those which he thinks likeliest to catch the fancy of the other competitors, all of whom are looking at the problem from the same point of view … we devote our intelligences to anticipating what average opinion expects the average opinion to be."

So if the conventional wisdom is that economic conditions will remain more or less normal despite highly abnormal policy, markets will remain calm until the illusion of normality becomes unsustainable. At that point market prices may “change violently.” The current technical term for this phenomenon is a “Wile E. Coyote moment” — the moment when the cartoon character, having run several steps off the edge of a cliff, looks down and realizes that there’s nothing supporting him. Only then, according to the laws of cartoon physics, does he fall.

You might ask why smart investors with long time horizons don’t foresee Wile E. Coyote moments and get very rich in the process. Some do. But for reasons that would take another long post to explain — maybe a primer one of these days — there never seem to be enough such investors to shake market complacency, no matter how unwarranted. It’s one thing to short a stock, but to short the entire market is a completely different beast.

Can I document these assertions? Let’s look at a couple of relatively recent examples of market complacency and myopia in the midst of clear signals of an oncoming crisis.

First, the subprime crisis of the 2000s. By 2005, at the latest, there were very good reasons to suspect that we were in the midst of a major housing bubble. Here’s a graph of one measure of housing overvaluation, the ratio of home prices to average rents: 

When home prices are very high compared with average rents, that indicates the likelihood of a bubble because, ultimately, the value of the house lies in its use as a place to live.

The shaded area starting in late 2007 is the Great Recession [GFC]. Now, one could try to rationalize the extremely high prices of houses relative to rents in 2006. But an honest assessment would at least have reflected the serious possibility — not the certainty — that there was a bubble in house prices during this period. It would also reflect the possibility of a flood of mortgage defaults when the bubble popped.

Yet ABX indices, a measure of perceived default risk on securities backed by subprime mortgages, didn’t show any serious decline until well into 2007, when the housing bubble had already been deflating for more than a year

Source: Bank for International Settlements

Another example of market complacency is the euro area crisis that began in 2009. By the mid 2000s it was already obvious that huge sums of money were flowing into southern European nations like Spain, where they were being used largely to finance highly speculative real estate investment — very much like the US sub-prime bubble.

Even if it was unclear that the flood of money would abruptly end -- a nasty “sudden stop” – the possibility of such a stop should have been reflected in bond yields.

Yet the spread between interest rates on Spanish and German bonds — a measure of the risk markets perceived that Spain would experience a crisis — stayed very low until the crisis was already underway:

So if you want to know why markets aren’t reacting to the risk of very bad policy if Trump takes over the Fed, you should know that major market reactions to that kind of risk are rare. In fact, I can’t come up with a single example.

All of which says, in turn, that the absence of a strong reaction to Trump’s assault on the Fed isn’t a sign that everything is OK. We are, in fact, looking at a policy disaster in the making. But markets probably won’t react strongly until the disaster is already upon us.

I sold all my personal holdings in early May, going into 100% cash.  (In my notional portfolio, I also went into cash, but reinvested later.  The reality is that clients want to enjoy the last of the any rise in the markets, and get angry if you underperform the share market, even if you eventually are right.  Understandable, but it leads to exactly the kind of market actions Krugman deplores.)

When I was still managing portfolios professionally, I sold 50% of our portfolios in early 2008, just before the GFC hit.  Mortgage default rates were already high.  If there was a recession, unemployment would rise, and defaults would explode.   At the first payrolls report in January 2008, for December 2007, employment fell (ironically revised away later!).  I was on holiday, and so was our dealer.  I went back into the office, I called him in from his holiday too, and we sold all our liquid stocks.

The perilous situation now is made even riskier by the dominance of AI stocks in the US share market.  AI may eventually make money, but the situation smells just like the dot-com boom of the early 2000s, which I also sat out.  Some internet companies did go on to eventually make big profits (Amazon!) but the market halved between 2000 and 2002.  The market cap of the top ten companies as a percentage of the total market cap is now at a record high.   7 of those companies are AI-related.  When the AI bubble bursts, the market will collapse, just like it did after the dot-com boom.

Will that happen tomorrow?  Who knows?  But it will happen.  It's only a matter of time.

DISCLAIMER:  I might be wrong.  That has happened from time to time before.

Will AI take your job?

 Survey results from 21 countries.

From Visual Capitalist

AI models are starting to fall apart

From Futurism

 

As CEOs trip over themselves to invest in artificial intelligence, there's a massive and growing elephant in the room: that any models trained on web data from after the advent of ChatGPT in 2022 are ingesting AI-generated data — an act of low-key cannibalism that may well be causing increasing technical issues that could come to threaten the entire industry.

In a new essay for The Register, veteran tech columnist Steven Vaughn-Nichols warns that even attempts to head off so-called "model collapse" — which occurs when large language models (LLMs) are fed synthetic, AI-generated data and consequently go off the rails — are another kind of nightmare.

As Futurism and countless other outlets have reported over the last few years, the AI industry has continuously barreled toward the moment at which all available authentic training data — that is, information that was produced by humans and not AI — will be exhausted. Some pundits, including Elon Musk, believe we're already there.

To circumvent this "Garbage In/Garbage Out" conundrum, industry titans including Google, OpenAI, and Anthropic have engaged in what's known as retrieval-augmented generation (RAG), which essentially involves plugging LLMs up to the internet so they can look things up if they're presented with prompts that don't have answers in their training data.

That concept seems pretty intuitive on its face, especially when presented with the specter of rapidly-approaching model collapse. There's only one problem: the internet is now full of lazy content that uses AI to drum up answers to common questions, often with hilariously bad and inaccurate results.

In a recent study from the research arm of Michael Bloomberg's media empire that was presented at a computational linguistics conference in April, 11 of the latest LLMs, including OpenAI's GPT-4o, Anthropic's Claude-3.5-Sonnet, and Google's Gemma-7B, produced far more "unsafe" responses than their non-RAG counterparts. As the paper put it, those safety concerns can include "harmful, illegal, offensive, and unethical content, such as spreading misinformation and jeopardizing personal safety and privacy."

"This counterintuitive finding has far-reaching implications given how ubiquitously RAG is used in [generative AI] applications such as customer support agents and question-answering systems," explained Amanda Stent, Bloomberg's head of AI research and strategy, in another interview with Vaughn-Nichols published in ZDNet earlier this month. "The average internet user interacts with RAG-based systems daily. AI practitioners need to be thoughtful about how to use RAG responsibly."

So if AI is going to run out of training data — or it has already — and plugging it up to the internet doesn't work because the internet is now full of AI slop, where do we go from here? Vaughn-Nichols notes that some folks have suggested mixing authentic and synthetic to produce a heady cocktail of good AI training data — but that would require humans to keep creating real content for training data, and the AI industry is actively undermining the incentive structures for them to continue — while pilfering their work without permission, of course.

A third option, Vaughn-Nichols predicts, appears to already be in motion.

"We're going to invest more and more in AI, right up to the point that model collapse hits hard and AI answers are so bad even a brain-dead CEO can't ignore it," he wrote.

Fake Crowd

 By Walt Handelsman

AI is wreaking havoc on global power systems

 From The Guardian.

The rise in demand for datacentres, driven in no small part by the hype around AI systems, is fuelling an increase in demand for electricity. 

This excellent, in-depth feature from Bloomberg charts the rise of data centre electricity demand, how it may outstrip the supply of electricity from renewables, and more.

Tesla Robotaxis. I was wrong.

I've been doubtful that Tesla (or anybody) would ever make an AI which would be able to safely drive a car. Well, I'm eating my words.

From The Driven.

After more than a decade of development on its revolutionary vision based autonomous driving software, Tesla will finally reveal its much anticipated Robotaxi on August 8, 2024.
The Robotaxi unveil will mark the convergence of Tesla’s latest Full Self Driving software and revolutionary 3rd generation vehicle manufacturing and usher in a new era of “Transport as a Service” (TAAS) with massive ramifications for the 70 million unit per annum global fossil car industry.

The announcement comes as the online Tesla community is abuzz with Full Self Driving (FSD) Beta software testers raving about the latest FSD version 12.3.3 update, with drivers reporting zero interventions during long drives in complex city traffic.

Former Tesla employee and YouTuber Farzad Mesbahi discussed the latest software update with James Douma, who’s one of tens of thousands of Tesla drivers in the US who’ve been testing the Beta software over the past two years.

“It’s a pretty remarkable departure in behaviour from V11,” said Douma. “It just works, you just don’t have interventions anymore.”

Douma, who’s been testing the latest update for the last two weeks, says he’s completed hours of city driving without manually overriding the software.

“The first thing I did was spend 3 hours driving all over the part of LA I live in, just random pin drops.” said Douma.

“And I didn’t have any interventions, it was rock solid.”

Another FSD tester and online Tesla blogger Omar Qazi, AKA @WholeMarsBlog, has also been testing FSD beta and has posted some stunning videos of version 12.3.3 in action around San Francisco. 

 

Unlike other companies who’ve attempted to use LiDAR to solve autonomous driving, Tesla’s strategy from the beginning was to use a vision-based system of camera’s and artificial intelligence.

The theory being that humans naturally use vision to drive and navigate the world so why shouldn’t machines? Our road networks are all designed for vision with lines and signs which can be easy read by cameras and AI.

The software is so advanced that it can differentiate between sedans, utes, trucks and buses as well as motorbikes, scooters and bicycles. It can accurately identify pedestrians, traffic cones, wheelie bins and even dogs and place them in 3D space with astonishing precision.

Unlike the purely object based LiDAR system, the cameras can also identify and read traffic signage such as stop signs, traffic lights, speed limits, road works and even the arrows and symbols painted onto road surfaces. For an in-depth look of Tesla’s FSD software development see The Rise of the Machines: Tesla drives 50km autonomously through heavy LA traffic.

Unlike other companies who’ve attempted to use LiDAR to solve autonomous driving, Tesla’s strategy from the beginning was to use a vision-based system of camera’s and artificial intelligence.

The theory being that humans naturally use vision to drive and navigate the world so why shouldn’t machines? Our road networks are all designed for vision with lines and signs which can be easy read by cameras and AI.

The software is so advanced that it can differentiate between sedans, utes, trucks and buses as well as motorbikes, scooters and bicycles. It can accurately identify pedestrians, traffic cones, wheelie bins and even dogs and place them in 3D space with astonishing precision.

Unlike the purely object based LiDAR system, the cameras can also identify and read traffic signage such as stop signs, traffic lights, speed limits, road works and even the arrows and symbols painted onto road surfaces. For an in-depth look of Tesla’s FSD software development see The Rise of the Machines: Tesla drives 50km autonomously through heavy LA traffic.

If Tesla delivers on its August 8th commitment to showcase the self-driving Tesla Robotaxi, it will mark yet another correct prediction made by technology futurist Tony Seba.

Seba, who was interviewed on The Driven podcast last year, predicted in his 2014 book Clean Disruption that lithium-ion batteries would reach $50/kWh by 2027.

That was a forecast that many people said was crazy. However, it now seems Seba’s prediction was too conservative as Chinese battery maker (and Tesla supplier) CATL is likely to reach the milestone by mid-2024.

Despite being considered one of the boldest technology forecasters in the world, Seba has also underestimated the speed of development of battery longevity. In 2017 he predicted the first million-mile battery by 2030 however last week CATL announced a new EV battery with a 1.5 million km warranty, effectively beating Seba’s prediction by 5 years.

On autonomous vehicles Seba had some fascinating insights which he shared during his interview with The Driven.

“The day that we get level four, autonomous technology ready and approved by regulators, when that converges with on-demand, and electric transportation we will get what we call transportation as a service [TAAS].” Seba told The Driven.

“Some call it Robotaxi. Essentially, when that happens the cost per mile of transportation is going to drop by anywhere from 10 to 20 times.”

“So for most people who can barely pay their bills, it won’t make any sense to own a car,” said Seba.

“Do I spend $50,000 over the next five years to own a car? Or do I pay $100 a month for a subscription to transportation as a service?”

Seba says ICE vehicles get around 140,000 miles (225,000 km) over their lifetime. An EV with a 1.5 million km battery will get almost 7 times that amount. This means that EVs will last at least 6-7 times longer than ICE vehicles meaning the global car market will likely drop by over 75% because people won’t need to replace cars as often.

“People are going to be buying vehicles a lot less often. So with that, essentially cut the global vehicle market by a factor of four or five.”

TAAS combined with the million-mile battery will mean new vehicle sales will drop even further as people opt for super cheap electric robotaxi transport instead of spending tens of thousands on private vehicles.

“Either way, it’s pretty much over for internal combustion engine.” says Seba.

I doubt that robotaxis will be the money-spinner Musk says.  If they become that profitable, everyone will buy a Model 3 to make money, and the charges they will be able to levy will go down.  (BTW, I don't think Tesla will be allowed to run a robotaxi monopoly --- but that doesn't mean they won't be able to charge a lot for FSD)  But that only implies that TaaS will take off.  Seba is right.  Why pay a fortune for a car which sits in your driveway or at in a car park for most of its life?  Taxis are expensive because they have to have a human driver and because they're ICEVs.  Robotaxis will be cheap.

Anyway, now I'm convinced.   I was wrong.

AI religious Trump