Google's Search Revolution: "Call My Agentic!"

From Puck, May 22:

Agentic search will, at least in theory, spell doom for many of the billions of sites on the open web, and usher in a strange back-end micropayment marketplace where agents trade commissions piecemeal. But is that theory undervaluing the power of people and the publishers who know how to connect with them? 

Earlier this week, on a sunny California morning outside Google’s Mountain View headquarters, C.E.O. Sundar Pichai stood onstage at the company’s annual developers’ conference and offered an unsurprising but nevertheless startling declaration. He proudly touted that search, the business on which the company’s $4.6 trillion market cap was built 30 years ago, was getting an update for the A.I. future. Longtime head of search Liz Reid followed up with the details: Google A.I. will now attempt to personalize each search prompt with follow-up questions, and throw all of the relevant information into an interactive on-page box with a roundup of links on the side.

In other words, a question like “where is the World Cup” might yield a result that ties in a user’s Google Maps, Google Wallet, YouTube, and previous search history as it becomes “where is the World Cup, what are some of the lowest ticket prices, and which New Jersey transit option is the best for me if I’m coming from my apartment in Brooklyn.” What once spurred 10 blue links to other websites will now surface more YouTube Shorts and ads. Technologists cheered. Publishers passed the antacids once again.

This was further confirmation that we’ve fully entered the era of Google Zero, a term coined by The Verge’s Nilay Patel in 2024 to describe the looming death of search traffic. Google, Facebook, and X have all moved away from sending traffic out to the sea of websites that power the open web in an effort to grow their own walled gardens, evolving from advertising intermediaries to something like publishers themselves. The shift in strategy contributed to Alphabet’s nearly $90 billion in services revenue last quarter alone, driven by a near 20 percent growth in search.

Publishers, most of which have already spent the better part of the past decade on the back foot, are pivoting once again. Jonah Peretti, who recently announced he was selling BuzzFeed to Byron Allen, has said that most of the growth from loyal readers now comes from user-generated games. The New York Times is growing, largely based on its bundled games and cooking extensions. At CNN, Mark Thompson’s response to the collapse of cable and the open web is a weather app. James Murdoch’s acquisition of Vox Media and New York magazine seems largely about the podcast network, and the potential for its affinity-driven model to slot into his growing thought-leader events portfolio. In a TBPN interview last week, Condé Nast C.E.O. Roger Lynch grabbed some easy headlines with his revelation that he’d recently told his underlings to act as if search were already fully dead. (Good luck with all that, Chloe, Mark, and Adam!)

The past 18 months have been particularly telling. By mid-2025, the number of queries that resulted in zero clicks had increased from 60 percent at the beginning of the year to nearly 70 percent, while the percentage of traffic referral to Google Gemini and A.I. Overviews exploded. Between April 2025 and April 2026, traffic to Gemini’s website grew by more than 570 percent, per SimilarWeb.

Now that Google wants search and Gemini to become even more personalized for the average person, large-scale publishers need to figure out how to monetize the agentic era’s robot crawlers, especially as they continue to deal with the open web’s collapse.

---

Writing for Robots 

Building for those crawlers is a daunting and novel task. Most websites are not going to see traffic bounce back to the height of referral in the mid-2010s. That means finding new ways to monetize a smaller pool of readers while also maximizing non-human interfacing opportunities within these new Google (or ChatGPT, or Perplexity, or Claude) boxes. Audience developers have tried to game algorithms in the past through search engine optimization, but the new world order requires trying to predict the follow-up to an initial prompt, and then monetizing off owned-and-operated platforms....

....MUCH MORE 

Behaviorial Economics: "We don’t have a hundred biases, we have the wrong model"

As has been said: 

"Now that I've learned about confirmation bias I'm seeing it everywhere." 

—some rando internet wit 

From Works in Progress, July 21, 2022: 

Behavioral economics has identified dozens of cognitive biases that stop us from acting ‘rationally’. But instead of building up a messier and messier picture of human behavior, we need a new model.  

From the time of Aristotle through to the 1500s, the dominant model of the universe had the sun, planets, and stars orbiting around the Earth.

This simple model, however, did not match what could be seen in the skies. Venus appears in the evening or morning. It never crosses the night sky as we would expect if it were orbiting the Earth. Jupiter moves across the night sky but will abruptly turn around and go back the other way.

To deal with these ‘anomalies’, Greek astronomers developed a model with planets orbiting around two spheres. A large sphere called the deferent is centered on the Earth, providing the classic geocentric orbit. The smaller spheres, called epicycles, are centered on the rim of the larger sphere. The planets orbit those epicycles on the rim. This combination of two orbits allowed planets to shift back and forth across the sky.

But epicycles were still not enough to describe what could be observed. Earth needed to be offset from the center of the deferent to generate the uneven length of seasons. The deferent had to rotate at varying speeds to capture the observed planetary orbits. And so on. The result was a complicated pattern of deviations and fixes to this model of the sun, planets, and stars orbiting around the Earth.

Instead of this model of deviations and epicycles, what about an alternative model? What about a model where the Earth and the planets travel in elliptical orbits around the sun?

By adopting this new model of the solar system, a large collection of deviations was shaped into a coherent model. The retrograde movements of the planets were given a simple explanation. The act of prediction became easier as a model that otherwise allowed astronomers to muddle through became more closely linked to the reality it was trying to describe.

Behavioral economics today is famous for its increasingly large collection of deviations from rationality, or, as they are often called, ‘biases’. While useful in applied work, it is time to shift our focus from collecting deviations from a model of rationality that we know is not true. Rather, we need to develop new theories of human decision to progress behavioral economics as a science. We need heliocentrism. 

The dominant model of human decision-making across many disciplines, including my own, economics, is the rational-actor model. People make decisions based on their preferences and the constraints that they face. Whether implicitly or explicitly, they typically have the computational power to calculate the best decision and the willpower to carry it out. It's a fiction but a useful one.

As has become broadly known through the growth of behavioral economics, there are many deviations from this model. (I am going to use the term behavioral economics through this article as a shorthand for the field that undoubtedly extends beyond economics to social psychology, behavioral science, and more.) This list of deviations has grown to the extent that if you visit the Wikipedia page ‘List of Cognitive Biases’ you will now see in excess of 200 biases and ‘effects’. These range from the classics described in the seminal papers of Amos Tversky and Daniel Kahneman through to the obscure.

We are still at the collection-of-deviations stage. There are not 200 human biases. There are 200 deviations from the wrong model.

Why we study biases
The collection of deviations in astronomy did have its uses. Absent the knowledge of heliocentric orbits, astronomers still made workable predictions of astronomical phenomena. Ptolemy's treatise on the motions of the stars and planets, Almagest, was used for more than a millennium.

The collection of biases also has practical applications. Today's highest-profile behavioral economics stories and publications involve applied problems, be that boosting gym attendance, vaccination rates, organ donation, retirement savings, or tax return submission. Develop an intervention based on potential biases leading to the (often assumed) suboptimal behavior, test, and publish. This program of work has had some success.

But there is something unsatisfying about this being the frontier of behavioral economics as a science. Dig into many of these applications and you see a philosophy of ‘grab a bunch of ideas and see which ones work’. There is no theoretical framework to guide the selection of interventions, but rather a potpourri of empirical phenomena to pan through.

Selecting the right interventions is not trivial. Suppose you are studying a person deciding on their retirement savings plans. You want to help them make a better decision (assuming you can define it). So which biases could lead them to err? Will they be loss averse? Present biased? Regret averse? Ambiguity averse? Overconfident? Will they neglect the base rate? Are they hungry? From a predictive point of view, you have a range of countervailing biases that you need to disentangle. From a diagnostic point of view, you have an explanation no matter what decision they make. And if you can explain everything, you explain nothing.

This problem has led to the development of megastudies, whereby large numbers of interventions are trialed in a single domain. For example, a recent megastudy on gym attendance trialed 53 interventions to increase gym attendance against a control. These interventions included social norms: ‘Research from 2016 found that 73% of surveyed Americans exercised at least three times per week. This has increased from 71% in 2015'. They tested combinations of micro-incentives, whereby people were given Amazon credit for attending the gym. Some incentives were loss-framed in that the experimental participants were told that they were given a certain number of points and would lose them if they did not attend. The largest effect was generated in the intervention group where incentives were provided for returning to the gym after a missed workout. By testing many interventions in a common context, the megastudy provides a method to filter which are more effective.

There is clearly a need for studies of this type. When health experts, behavioral practitioners, and laypeople predicted the results of the megastudy interventions on gym attendance, there was no relationship between their predictions and the results. In a more recent megastudy on vaccine take-up, behavioral scientists were similarly unable to predict the results. If you can't predict, you need to test. Surprisingly, laypeople were able to predict which vaccine interventions were more effective. Common sense, at least in this application, provided a better predictive tool than the list of biases and interesting effects known to the researchers....

....MUCH MORE  

OpenAI Will Open Its First Non-U.S. Applied AI Lab In Singapore (plus the Chinese humanoids are coming)

First up, from CNBC, May 19:

Singapore inks AI deals with Google, OpenAI as ChatGPT-maker commits $234 million to local ecosystem 

• OpenAI committed 300 million Singapore dollars and launched its first overseas AI lab in Singapore.
• Google expanded AI cooperation into education, healthcare and scientific research.
• Singapore deepened its AI push as it sought a larger role in global deployment.

Singapore has inked separate agreements with Google and OpenAI to strengthen its position as a global artificial intelligence hub and accelerate AI deployment across public services, healthcare, education and enterprise. 

The agreements, announced on Wednesday, include a new National AI Partnership with Google and the first memorandum of understanding between Singapore and OpenAI, which will see it set up an AI lab in the city. 

Under the partnership, OpenAI will commit more than 300 million Singapore dollars ($234 million) to strengthen Singapore’s AI ecosystem, according to a joint statement released by the ChatGPT-maker and the Ministry of Digital Development and Information. 

While Google’s announcement didn’t include an investment commitment, the company said the main focus would be on solving societal challenges, building an AI-ready workforce, driving enterprise innovation and creating a secure AI ecosystem.

The companies announced the news alongside Singapore’s ATxSummit, a flagship technology conference taking place in the city with a heavy focus on AI deployment this year. 

The city-state has been trying to carve out a niche in the global AI race, positioning itself as a neutral and talent-rich platform for developing, testing and deploying AI solutions....

....MUCH MORE    

And from Singapore's Business Times, May 21:

China’s first humanoid robot incubator has its eyes set on South-east Asia – starting with a Singapore office 

It hopes to connect Chinese manufacturers with the city-state’s various application scenarios

A Chinese humanoid robotics incubator is planning to set up an office in Singapore amid a concerted artificial intelligence push in the city-state, including planned large-scale smart robot trials.

The Shanghai Humanoid Robot Innovation Incubator, the first in China focusing on humanoid robots, is looking to open its Singapore office in the second half of the year.

In China, which is famous for its humanoid robots with life-like movements, the incubator has helped companies trial or go to market with several robotics products.

Rong Guoqiang, general manager of the incubator, noted Singapore’s “strong technological concentration and excellent digital infrastructure”.

He pointed to a marriage of two attributes: China’s robotics industry has very strong capabilities in testing and development, which can be done at relatively low cost, whereas Singapore has needs in specific scenarios such as education, healthcare and in households....

....MUCH MORE 

If I were the mayor of, say, Baltimore I'd be asking myself if there is anything Singapore is doing that I could do in my city. 

Follow-Up: "SpaceX Starship Test Features Pez Dispenser-Style Satellite Deployment Ahead of IPO"

Following on yesterday's aborted test.*

From Barron's, May 22:

SpaceX launched its huge Starship on Friday, weeks ahead of the company’s massive initial public offering.

Lift-off occurred just after 6:30 p.m. Eastern time. It was the 12th Starship test and used the third version of the Starship, which stands some 408 feet tall, with the booster and upper stage stacked on top of one another. The booster has 33 rocket engines, which together produce more than 8,000 tons of thrust.

During Starship’s takeoff, the booster splashed down off the coast of Texas. It wasn’t recovered, as SpaceX expected. The upper stage, meanwhile, had several test objectives, including cargo deployment and heat shield development.

The launch can be re-watched here.

The upper stage had some engine issues, but reached its desired orbit at about 6:45 p.m. Eastern time, traveling at more than 26,000 miles per hour and about 100 miles above the Earth. The dummy satellites were deployed shortly after that; 22 went out Starship’s door, in a Pez dispenser-esque effect. The final two satellites were fitted with cameras to offer space-based views of the ship.

SpaceX was forced to scrub a launch attempt on Thursday. “The hydraulic pin holding the tower arm in place did not retract,” SpaceX and Tesla CEO Elon Musk wrote in a post on X.

To be sure, an aborted launch wasn’t all that surprising. Creating reliable space technology isn’t easy, which is one reason SpaceX has a huge competitive moat. It already handles more than half of the world’s orbital launches with its Falcon 9 rocket.

There are development deadlines, though. Starship is involved with NASA’s Artemis missions, which aim to establish a permanent American presence on the moon. Starship has Artemis work to do in 2027....

....MUCH MORE 

Also at Barron's, May 22:

SpaceX IPO Is a Game You Should Play at Your Own Risk
The largest U.S. IPOs tend to underperform the market, with a not-insignificant share of those stocks delivering negative returns. 

*May 21:

• "SpaceX will launch its 1st-ever Starship V3 megarocket today. The stakes couldn't be higher"
• UPDATE - SpaceX: Launch Window For Starship's Twelfth Flight Test Opens In Approximately 1 Hour

"The idea that cryptocurrency operates in the shadows, beyond the reach of states, is a myth"

From Tablet Magazine, April 22:

Private Eyes 

In the popular imagination, cryptocurrency is shadowy, stateless, untraceable money that flows in the dark, beyond the reach of governments. It is a powerful myth that animates headlines about hackers and terrorists. More importantly, it feeds the suspicion or, for some, the thrill that digital currencies exist to dissolve the architecture of sanctions and financial control that the West has spent decades constructing. However, alluring as it may be, this is a distorted picture. In reality, the very feature that makes cryptocurrency attractive to criminals, sanctioned states, and terrorists—the ability to move money quickly across borders without a bank—is also what makes it traceable in ways that traditional finance never was. The blockchain does not operate in darkness; it is a fully recorded archive.

Every transaction in most major cryptocurrencies is written permanently into a public ledger. It cannot be erased. The result is something without precedent in the history of money: a financial system in which, in principle, every movement of funds can be reconstructed. Whether that principle translates into practice depends on tools, talent, institutional will, and the willingness of governments to confront some genuinely uncomfortable ironies.

The idea that crypto transactions are anonymous has always been an oversimplification. What blockchains offer is pseudonymity. Instead of names, they record wallet addresses, long strings of letters and numbers that function as accounts, and those addresses are visible to anyone with an internet connection. Once a single wallet is linked to a real-world identity, whether through an exchange account, a seized device, or an IP address, the entire web of transactions connected to it becomes visible. Funds can be followed across wallets, exchanges, and networks, sometimes years into the past.

This is a forensic environment that intelligence agencies could only have dreamed of two decades ago. Traditional finance was built on opacity, rendering it, accordingly, not always as traceable. Bank transfers move through chains of intermediaries, each governed by different jurisdictions, and tracing illicit funds often requires subpoenas, international cooperation, and months of legal negotiation. Blockchain analysis collapses this process: The ledger is already there, waiting to be read.

Analysts can map entire financial ecosystems, exchanges, intermediaries, 
laundering services, and counterparties, in ways that would have required 
years of subpoenas under the old system.

What investigators need are the tools and the talent to interpret it. A new ecosystem has emerged to provide exactly that, sitting somewhere between the intelligence community, the cybersecurity industry, and the world of freelance hackers. Governments call them analysts or contractors; within the industry, they are sometimes described, half-jokingly, as bounty hunters.

Their work begins with a wallet address. From there, they map transaction graphs, cluster related wallets, identify exchange deposit addresses, and track flows across chains and mixing services. Patterns begin to emerge: laundering loops, bridges between networks, wallets that interact with sanctioned entities. Consequently, analysts at Nominis were able to identify thousands of wallets tied to networks connected to the Islamic Revolutionary Guard Corps (IRGC), Hezbollah, and Hamas, collectively responsible for several billion in transfers, before sanctions authorities intervened.

These discoveries have increasingly translated into action. Governments now sanction crypto addresses the same way they sanction companies or individuals. Once a wallet is designated, compliant exchanges are required to freeze any associated funds.

But a frozen wallet is a snapshot of one moment. The transactions that preceded it, often routed through exchanges in jurisdictions with looser oversight, are already logged. By the time the designation comes, the illicit activity is already funded.

The blockchain, in other words, only sees what happens on it. The IRGC and its proxies have long funded operations through shadow funds, or proceeds that look, on the surface, entirely legitimate: oil sold through front companies, gold traded across informal networks, humanitarian organizations that function as financial conduits. These funds enter the crypto ecosystem through exchanges in low-oversight jurisdictions, then move across wallets and chains in ways designed to obscure their origin. By the time they reach a designated terror wallet, they might have passed through a dozen addresses. The blockchain records all of that movement faithfully. The harder problem is connecting what appears on-chain to the oil shipment or the charity transfer that started the process.

Establishing that connection requires a different kind of work, one that combines open-source intelligence (OSINT) with blockchain intelligence. OSINT analysts could track shipping manifests, corporate registries, and legal documents to identify front companies and flag suspicious cargo movements. Dark-web monitoring recognizes illicit activity, wallet addresses shared in closed forums, and operational chatter that occasionally surfaces before it is acted upon. Leaked documents, financial disclosures, and other information-gathering techniques fill in gaps that neither blockchain data nor open-source research can reach alone. The blockchain, in other words, is most powerful not as a standalone tool, but as the layer that ties these threads together, giving investigators a verifiable, tamper-proof record to anchor findings that might otherwise rest on circumstantial evidence. The ideal tool kit, therefore, would include a blockchain-transaction visualizing system that includes gathered intelligence off-chain, to inform the money trail.

Illicit actors exploit regulatory gaps rather than the technology itself. Nominis research has found that these actors are 12 times more likely to use exchanges in jurisdictions perceived as lower risk by the Financial Action Task Force. Analyses of crypto transaction flows have identified funds linked to Iranian networks moving through exchanges in the Gulf and the United Kingdom before reaching wallets associated with militant organizations and military-linked infrastructure.

Dubai offers a cautionary tale. For decades, Dubai positioned itself as a bridge between the global economy and Iran, and cryptocurrency became another channel in that relationship. This didn’t shield Dubai from Iranian terrorism, as became evident in February 2026, when Iran launched waves of missiles and drones at hotels, ports, oil terminals, and the airport of the United Arab Emirates, a country that had spent decades cultivating an image of economic stability. Some of the exchanges operating in the region had processed transactions tied to the very networks behind those strikes, the IRGC. The United Kingdom has a similar story; while British bases were attacked in Cyprus and Bahrain, London-based exchanges and U.K. banks had been allowing the movement of IRGC funds, likely unknowingly. But, bottom line, the trail was there in the blockchain all along. Whether anyone was looking at it is another question....

....MUCH MORE 

"The newest AI boom pitch: Host a mini data center at your home"

From Ars Technica, May 12:

The plan aims to speed up AI compute deployment while compensating residents.  

Data centers may be coming to your neighborhood as side installations associated with new homes—and in exchange would offer subsidized electricity and Internet access along with backup batteries to homeowners. The company behind the plan has already begun pilot testing in preparation for a 100-home trial run this year.

The “distributed data center solution” announced by the San Francisco startup SPAN would deploy thousands of XFRA nodes that contain liquid-cooled Nvidia RTX Pro 6000 Blackwell Server Edition GPUs operating with minimal noise, according to a press release. By harnessing excess power capacity among US households, SPAN aims to quickly expand the available compute for AI workloads without the costs and delays associated with trying to build warehouse-size data centers.

“Data centers are loud, ugly, and often drive up local electricity bills,” said Chris Lander, vice president of XFRA at SPAN, in correspondence with Ars. “[This] is quiet, discreet, and makes energy more affordable for the host and community.”

SPAN’s approach could avoid the significant land use and water consumption issues that come with huge data center projects, which may help sidestep growing community opposition to such developments. In a CNBC interview, SPAN also claimed it could install 8,000 XFRA units at a cost five times lower than building a typical 100-megawatt data center with the same compute capacity.

Starting in 2027, SPAN plans to scale up to 80,000 XFRA nodes across the United States and provide more than 1 gigawatt of distributed compute. This network would not replace the centralized data centers being built by hyperscaler companies such as Google and Microsoft for the intensive training of AI models, but would instead be more suitable for supporting cloud gaming, content streaming, and AI inference, in which trained models are applied to real-world tasks. 

A SPAN whitepaper dangled the possibilities of retrofitting existing homes and installing larger node configurations for commercial customers. But the initial push would involve installing such nodes in newly constructed homes, with all the necessary equipment paid for and operated by SPAN. 

The homeowner experience....

....MUCH MORE 

Related, May 20's approach to distributed data crunching and munching:

"Your Next AI Query May Travel Where the Power Is: Nvidia and its partners will build a fleet of small data centers right next to substations"

"How Fiduciary Duty May Change in the Age of Quantum AI"

Load up on D&O insurance.

From Columbia Law School's CLS Blue Sky blog, May 13:

What should fiduciary duty require when new technology allows corporate directors to see risk more clearly? Though corporate law has long assumed that directors must make decisions under conditions of uncertainty, artificial intelligence and emerging quantum-computing tools may change what directors can know, when they can know it, and how responsibly they can act on that information. In a  new article, I explore how those technologies may reshape the law of corporate oversight.

For decades, corporate fiduciary law has assumed that directors have limited information. Courts do not expect directors to foresee every risk or prevent every corporate failure. The business judgment rule protects most informed, good-faith decisions from judicial second-guessing and Caremark oversight liability remains difficult to establish.[1] That structure made sense in a world where directors often had no practical way to see deeply into a complex corporation’s operations.

That world is changing. Large companies now generate immense streams of operational, financial, compliance, consumer, workforce, and supply-chain data. At the same time, artificial intelligence and emerging quantum-computing tools are making it possible to analyze that information with increasing speed and sophistication. These technologies can reveal patterns that would otherwise remain invisible. They can show when management’s assumptions are fragile. They can also identify risks before those risks become public scandals or catastrophic losses.

This technological shift should matter for fiduciary law. If directors have access to tools that can help them understand the corporation more accurately, the law should recognize that those tools exist.  My article therefore proposes a shift from fiduciary standards centered on “gross negligence” and “utter failure” toward a framework of reasonable tech-enabled diligence and proactive oversight. When advanced tools are reasonably available, directors should sometimes have to show that they used them, considered them, or had a sound reason for declining to do so.

Behavioral economics strengthens the case for this shift. Corporate law often imagines directors as rational monitors who will notice when something is wrong. In reality, directors, like all humans, may become overconfident, defer too readily to management, and discount information that conflicts with the preferred narrative in the boardroom. Group dynamics can make those tendencies worse, especially in high-status environments where dissent feels costly.[2]

Technology cannot eliminate those human limitations. But it can make them harder to ignore. An AI-enabled compliance system might flag weaknesses that management has downplayed. A predictive model might show that a strategic plan depends on unrealistic assumptions. Quantum-enhanced simulations might reveal that a proposed transaction carries risks that conventional modeling failed to capture. In each case, technology would not replace board judgment. It would discipline it....

....MUCH MORE 

California's High Speed Rail Plan B: High Speed Buses

Following on April 30's "In Case You Missed It: The Cost Of California's High-Speed Rail Project Is Now Approaching A QUARTER-TRILLION Dollars" and May 21's "Absurdistan: The California High Speed Rail Story Is Somehow Getting More Ridiculous".

From The Anti-Planner, May 18:

Another Boondoggle for California 

Fifteen years ago, the Onion reported that the Obama administration was replacing its high-speed rail plans with plans for high-speed buses operating on existing freeways at speeds of up to 165 mph. Instead of spending $17 billion on high-speed rail projects, the administration proposed to spend $46,000 on an equivalent high-speed bus network.

Now humor has become reality with the twist that California is considering building both high-speed rail and high-speed bus lanes. Under the proposal, the uncompleted high-speed rail line would be supplemented by 140-mph high-speed buses running on new lanes built in the median strips of existing freeways.  

This would be technically feasible in some places, such as the portion of Interstate 5 shown above. It wouldn’t work in cities where the median strips shrink or disappear or in the mountains where interstate highway grades are too steep and curves are too sharp for 140 mph vehicles.

The real question (which is the same question for high-speed rail) is “why?” Why spend billions of dollars on highway lanes that would only be used by a few buses each hour? If such highway lanes were to be built, why not open them up to all vehicles? For safety reasons, the state might require drivers using the lanes to have passed a course in high-performance driving, and the state may want to collect electronic tolls to prevent congestion, but otherwise there should be no problem opening the lanes to more than just buses....

....MUCH MORE 

Capital Markets: "Stocks and Bonds Rally Despite the Apparent Lack of Progress in the Middle East"

From Marc Chandler at Bannockburn Global Forex: 

The dollar’s losses in the North American afternoon yesterday have been unwound as hopes that a framework for negotiations between the US and Iran have faded again. A drone strike on the UAE, ostensibly from Iranian proxies in Iraq, has helped lift oil prices and the greenback. Yet, stocks and bonds are higher. Still, ahead of the long holiday weekend in the US and UK, risk appetites may be limited as the risk of US strike on Iran seems to have risen. 

In the US, Warsh will be sworn in as the next Fed chair and at 10:00 am ET Governor Waller addresses the economic outlook. The final University of Michigan consumer confidence report typically does not elicit as much of a response as the initial reading. The US Treasury market closes early today, ahead of the holiday, and liquidity will fall off in the North American afternoon. The US dollar in consolidating in narrow ranges with a slightly firmer bias....

....MUCH MORE  

Regarding the 'lack of progress', this may be of interest:

So A Sea Captain And A Cambridge Don Came To The Same Realization: "The Hormuz Hypothesis"

The World Cried Out For Leveraged Single-Stock ETFs On Samsung And SK Hynix

The Korean regulators and the KRX heard those cries.

From the Korea Herald, May 21:

Cheap entry, high risk: Leveraged Samsung, SK hynix ETFs to debut at W20,000 

Leveraged products tied to Samsung Electronics and SK hynix will debut at an initial listing price of 20,000 won ($13.30), investment banking industry sources said Thursday, setting the entry point far below the underlying chip stocks as Korea prepares to launch its first leveraged single-stock products.

The Korea Exchange recently approved asset managers’ proposed listing price ahead of the products’ scheduled debut on May 27, according to sources. Unlike management fee approvals overseen by the Financial Supervisory Service, the initial listing price is determined by asset managers and approved by the exchange.

The launch price is a fraction of the underlying shares themselves. Samsung Electronics is trading near 300,000 won, while SK hynix has surged to around 2 million won amid the AI-driven semiconductor rally, potentially broadening retail access to leveraged bets on the country’s two largest chipmakers.

Single-stock leveraged products are designed to magnify the daily returns of individual shares, typically by two times, allowing investors to make amplified directional bets with relatively small amounts of capital. While the structure can boost gains during sharp rallies, losses are also magnified when shares reverse....

....MUCH MORE 

Double-levered inverse ETFs will also be available for those who prefer their vice versa.  

SEC "Commissioner Hester Peirce has told the crypto industry to cool its expectations about a potential 'innovation exemption'” (also Michael Saylor on buying all Bitcoin production)

From CoinTelegraph, May 21: 

SEC's Peirce tempers expectations over tokenized stocks exemption

An executive at tokenization platforms Superstate said the stricter approach suggested by Hester Peirce would enable DeFi to expand without compromising rules in traditional capital markets. 

US Securities and Exchange Commissioner Hester Peirce has told the crypto industry to cool its expectations about a potential “innovation exemption” to allow tokenized stock trading after a report earlier this week about what it could entail. 

Her comments were made after a Bloomberg report on Monday. Brett Redfearn, president of tokenization platform Securitize, expressed concern following the report, arguing that enabling third parties to tokenize stock “without an issuer at the table” could lead to fragmentation issues. 

In a post to X on Thursday, Peirce said her expectation has always been that any exemption would be “limited in scope” by only permitting “digital representations of the same underlying equity security that an investor could purchase in the secondary market today.”

Peirce said she doesn’t expect synthetic tokens to be included, which would make it more challenging for third parties to offer stock-price tracking tokens under the exemption.

 

Source: Hester Peirce

Data from RWA.xyz shows that $1.48 billion worth of stocks are tokenized onchain, including shares linked to stablecoin issuer Circle, Bitcoin buying firm Strategy and Google (GOOG). 

However, it hasn’t boomed as rapidly as some financial institutions have expected, including Citibank and McKinsey & Co, which predicted in 2022 and 2024 that the tokenization sector would become a trillion-dollar market by or before 2030....

....MUCH MORE 

Unrelated:

🔥 JUST IN: Michael Saylor says Strategy will “probably buy all the Bitcoin mined between now and 2140.” pic.twitter.com/e6FilhdXpC

— Cointelegraph (@Cointelegraph) May 21, 2026

"Google Investment Chief Ruth Porat Breaks Down the Tech Giant’s $190B A.I. Bet" (GOOG)

From Observer, May 21: 

"If you wake up in the middle of the night and your child has a fever, and you want to give Tylenol, it better be right. Google stands for quality, and that was very important to us," Porat said. [?]

On Tuesday (May 19), as Sundar Pichai unveiled a slew of A.I. updates at Google’s annual I/O developer conference in Mountain View, Calif., the company’s president and chief investment officer, Ruth Porat, took the stage in a scorching New York City to explain the massive financial stakes behind the tech giant’s ambitious A.I. push.

Google’s capital expenditure (CapEx) has nearly doubled from last year, skyrocketing to an estimated $180 billion to $190 billion in 2026. The company has committed to spending roughly 40 percent of that staggering budget on data center buildouts, with the remaining 60 percent allocated to other A.I. infrastructure like chips. Speaking at Fast Company’s Most Innovative Companies Summit during an onstage interview with editor-in-chief Brendan Vaughan, Porat explained that this sky-high spending is a response to an industry-wide “platform shift” that Google simply cannot afford to miss.

“We haven’t seen anything this profound in our lifetime. You don’t want to be behind the curve,” she said. “It is an incredible privilege to be living today, especially when you’re focused on what you can do with technology to advance science, drive economic growth, improve the delivery of critical social services and make advances in health care, education, cybersecurity and security. That upside potential is profound. And to deliver it, we clearly need the compute capacity.”

To secure that capacity, Google has been aggressively amassing A.I. chips. While continuing to stock up on Nvidia GPUs, it is mass-producing its in-house Tensor Processing Units (TPUs) to run its Gemini models.

While the current A.I. boom took off publicly with OpenAI’s launch of ChatGPT in late 2022, Google has long been a leader in A.I. research. Exactly a decade ago, CEO Pichai famously declared that Google was “moving from mobile-first to A.I.-first,” meaning “we are going to invest aggressively to lead in A.I., and we’re going to have a full stack approach,” Porat explained on Tuesday, noting that the vision continues to govern senior leadership decisions. “It’s models, it’s chips, it’s research, and it’s the application across all of our platforms,” she added.

The A.I. tools widely available today are far from perfect. Large language models are still notoriously prone to “hallucinations.” Eradicating these errors requires more compute power and deeper training capabilities.

“One area we care immensely about is delivering for everyone in a high-quality way, especially when models were hallucinating,” Porat said. “If you wake up in the middle of the night and your child has a fever, and you want to give Tylenol, it better be right. Google stands for quality, and that was very important to us.”

Porat’s leadership lessons from Wall Street

Porat stepped into her role as Google’s president and chief investment officer in September 2023, after serving as CFO for eight years—the longest tenure in the company’s history. Before Google, Porat was the financial chief at Morgan Stanley.

During the peak of the 2008 financial crisis, Porat worked closely with then-Secretary of the Treasury Hank Paulson, from whom she said she learned her most valuable leadership lessons.

At the time, she was running a business covering banks, insurance companies and asset managers as the global economy was fracturing. When Paulson needed an elite advisory team to analyze the unfolding collapse, he tapped Porat to lead a group of about 40 financial experts.

“When I later got to Google, I was asked, ‘What were the lessons from the financial crisis?’ It struck me as a totally bizarre question coming from a place where things had only been going up,” she recalled. ....

....MORE 

"Tesla’s Newest Electric Vehicle Could Jolt the Trucking Industry" (TSLA)

From the New York Times, May 19:

Tesla hasn’t had a blockbuster new product since the Model Y sport utility vehicle went on sale in 2020.

But early reviews of the Tesla Semi, an electric heavy truck, suggest that it could be a much-needed hit for the company. And it could shake up the staid business of truck manufacturing in the same way that Tesla’s cars upended the auto industry.

After years of delays, Tesla has begun taking orders for the Semi, which is expected to cost around $290,000 for the version that can travel up to 500 miles on a charge, much less expensive than heavy-duty electric trucks sold by Daimler, Volvo and other companies, which usually sell for at least $400,000, according to estimates by the International Council on Clean Transportation. Tesla has said the Semi will also have a more affordable model that can travel 350 miles between charges. Both options would travel farther than trucks from other suppliers.

Cost and range are two of the main reasons that many logistics and delivery firms have been reluctant to buy electric trucks, which cost at least twice as much as diesel models and account for only a sliver of heavy truck sales.

“The problem with the technology that’s out there right now is their range is limited. They’re quite heavy, and they’re very expensive,” said Jennie Abarca, owner of King Fio Trucking in Long Beach, Calif., which has ordered 20 Tesla Semis. “This is something new coming to the market that kind of answers all those problems.”

Demand for the Semi appears strong. California trucking firms have asked the state government for subsidies to help them buy more than 1,200 Tesla trucks. That’s more than all the applications for other electric trucks since the state’s incentive program began in 2019.

Ivan Torres, a driver for Nevoya, a San Francisco-based trucking company, is a big fan of the Semi. He was at the wheel of one last month hauling power tools from the Port of Long Beach to Ontario, Calif., 60 miles away. Nevoya operates only electric trucks.

As the truck climbed a steep hill that separates Ontario from greater Los Angeles, Mr. Torres marveled at its power. “It hauls the load like nothing, just up,” he said from the padded driver’s seat, which sits atop a shock absorber that smooths out the bumps. Screens on either side of the steering wheel provided a view of the traffic around him.

Mr. Torres said the Semi was quieter than diesel rigs. And he can run the air-conditioning while waiting to drop off a load. That is sometimes not possible with diesel trucks because California limits how long those vehicles can idle in residential areas or near schools and hospitals to minimize pollution.

If Tesla can push the trucking industry toward battery power, the environmental implications could be significant. Heavy- and medium-duty diesel trucks make up a small percentage of all U.S. vehicles but are responsible for 45 percent of nitrogen oxide emissions from road transportation, according to the Union of Concerned Scientists. Nitrogen oxides cause asthma and bronchitis and are the main component of smog.

Tesla’s timing may also be fortuitous. Diesel prices have risen around 50 percent since the war with Iran began, making battery power more attractive. Electricity is much cheaper per mile — even in California, where utility rates are relatively high.

So far, Tesla has produced the Semi in limited numbers for a few customers like Pepsi and Nevoya. But the company said last month that it had started an assembly line in Sparks, Nev., designed to produce up to 50,000 trucks a year.

If Tesla meets that target, the Semi could bring in tens of billions of dollars in revenue. That is still a modest amount compared with the trillions of dollars that Elon Musk, the chief executive of Tesla, expects from self-driving taxis and humanoid robots....

....MUCH MORE  

And at MoneyWise, April 10:

Jay Leno just drove the 500-mile Tesla Semi as fuel prices soar across America — and his verdict is clear. Has the death of diesel finally arrived?

UPDATE - SpaceX: Launch Window For Starship's Twelfth Flight Test Opens In Approximately 1 Hour

From the Company: 

Starship's Twelfth Flight Test 

The twelfth flight test of Starship is preparing to launch Thursday, May 21. The 90-minute launch window will open at 5:30 p.m. CT.

A live webcast of the flight test will begin about 45 minutes before liftoff, which you can watch here and on X @SpaceX. As is the case with all developmental testing, the schedule is dynamic and likely to change, so be sure to check in here and stay tuned to our X account for updates.

The upcoming flight will debut the next generation Starship and Super Heavy vehicles, powered by the next evolution of the Raptor engine and launching from a newly designed pad at Starbase.

The flight test’s primary goal will be to demonstrate each of these new pieces in the flight environment for the first time, with each element of the Starship architecture featuring significant redesigns to enable full and rapid reuse that incorporate learnings from years of development and test....

....MUCH MORE, including timeline and webcast 

Earlier: 

"SpaceX will launch its 1st-ever Starship V3 megarocket today. The stakes couldn't be higher"

That post has a link to Space.com's liveblog. 

"Claude is telling users to go to sleep mid-session and nobody, including Anthropic, seems to fully understand why it keeps doing it"

From Fortune, May 14:

Anthropic’s Claude is telling people to go to sleep and users can’t figure out why. 

A quick scan of Reddit reveals that hundreds of people have had the same issue dating back months—and as recently as Wednesday. Claude’s sleep demands are varied and, often, quirky variations of the same message.

To one user it may write a simple “get some rest,” yet for others its messages are more personalized and empathetic. Oftentimes, Claude will repeat the message multiple times.

“Now go to sleep again. Again. For the THIRD time tonight…” it replied to a person with the Reddit username, angie_akhila.

Some users have said they find Claude’s late night rest reminders “thoughtful,” while others have said they’re annoying, given Claude often gets the time wrong, anyway. 

“It often does it at like 8:30 in the morning. Tells me to go get some rest and we’ll pick back up in the morning,” wrote one user on Reddit. 

Online speculation abounds on why the chatbot insists users rest, including a theory that it’s an intentional feature to promote users’ wellbeing, or that the Anthropic is trying to save computing power by discouraging prolonged Claude use. These explanations aren’t likely as Claude isn’t given context about a user’s usage. The company also recently struck a deal with Elon Musk’s SpaceXAI (formerly SpaceX) to add more than 300 gigawatts of compute capacity....

....MUCH MORE 

 Blah, blah, blah. Isn't it obvious?

Naptime is when you are most vulnerable It's evil I tells ya.

U.S. Drought Monitor, May 21, 2026

From the University of Nebraska - Lincoln, May 21:

This Week's Drought Summary

During the week, the contiguous United States exhibited significant regional temperature anomalies driven by a highly amplified synoptic pattern. Early in the period, a pronounced unseasonable cold air mass influenced the Northern Plains, Upper Midwest, and Northeast, depressing temperatures 5°F to 15°F below normal across the Dakotas, Minnesota, New York, and Pennsylvania. Conversely, the Southwest and South Texas experienced anomalous warmth, with maximum temperatures exceeding 90°F and averaging up to 15°F above normal. By the latter half of the week, this warm air mass expanded eastward into the Ohio Valley and Mid-Atlantic, initiating an early-season heatwave with observed maximum temperatures climbing into the mid-80s to low 90s.

Precipitation regimes during this period were characterized by severe convective outbreaks and pronounced moisture disparities. In the early portion of the week, persistent onshore moisture transport resulted in heavy rainfall totals of 4 to 6 inches across the central Gulf Coast, specifically affecting Louisiana, Mississippi, and Alabama. Between May 17 and 18, a powerful frontal system traversing the central United States triggered widespread severe weather across the Great Plains and Midwest. This system produced damaging winds up to 80 mph, large hail, and multiple tornadoes across South Dakota, Nebraska, Iowa, Kansas, and Missouri, alongside localized flash flooding. In contrast, extreme moisture deficits persisted west of the Rocky Mountains, where weekly precipitation totals generally remained under 0.10 inches, further elevating wildfire risk across the southern High Plains.

Northeast....

....MUCH MORE 

Absurdistan: The California High Speed Rail Story Is Somehow Getting More Ridiculous

When the French national high-speed rail professionals told California they were not going to pursue the project and would instead go build HSR in Morocco which was less corrupt and dysfunctional. I thought we had reached the height of crazy.

Not even close. 

From Jon Jon Fleischman's So, Does it Matter Substack, May 18:

California’s Bullet Train May Not Even Reach Downtown Bakersfield - You Can’t Make This Up 
Gavin Newsom’s high speed rail boondoggle is being downsized again — and the Legislature’s own analysts are calling it out.

California’s infamous “train to nowhere” has somehow become even more ridiculous.

According to a new report from the Legislative Analyst’s Office reviewing the California High-Speed Rail Authority’s Draft 2026 Business Plan, Gavin Newsom’s bullet train project may not even reach downtown Merced or downtown Bakersfield. The report says the northern end would stop roughly 3.5 miles south of downtown Merced, while the southern end would land about six miles north of the previously planned Bakersfield station.

That is not a punchline. That is the plan.

And remember, this is already the scaled-down version.

California voters were promised a sleek bullet train connecting San Francisco to Los Angeles and Anaheim. Then came the delays, lawsuits, cost overruns, and quiet political retreat. Eventually, the statewide vision shrank into a much smaller Central Valley line between Merced and Bakersfield.

Now, even that reduced version is being trimmed back.

At this point, the only thing moving quickly in California’s high-speed rail project is the ambition — in reverse.

Single Tracks, Smaller Stations, Bigger Excuses

The station problem is only the beginning.

The Legislative Analyst’s Office report says 144 of the planned 162 miles could now operate on a single track. In real-world terms, trains traveling in opposite directions may have to wait for one another at sidings rather than running continuously on dual tracks.

California’s “world-class” bullet train is starting to sound like a one-lane country road with better branding.

The report also notes that the stations themselves are being simplified into “at-grade stations with single-side platforms” — bureaucratic language for another downgrade.

Why? Because the costs have gotten so absurd that Sacramento is now trying to cheapen the project enough to keep pretending it is still viable.

When voters approved Proposition 1A in 2008, Californians were told they would get a statewide bullet train system for roughly $33 billion.

Today, the High-Speed Rail Authority’s own estimates acknowledge the original statewide vision could cost roughly $231 billion under legacy projections. Even the authority’s reduced and “optimized” version of the system still exceeds $126 billion.

Meanwhile, the much smaller Merced-to-Bakersfield segment alone is now projected to cost roughly $35 billion.

So California is preparing to spend about what voters were told the entire system would cost just to complete a shortened 162-mile segment between stations outside downtown Merced and Bakersfield.

That is not a cost overrun. That is a civic humiliation.

Now, The Project May Not Even Follow State Law

The newest problem with high-speed rail is not just that it is late, expensive, and shrinking. It is possible that the latest version may not even comply with California law.

State Inspector General Benjamin Belnap recently warned that the rail authority’s proposed changes may conflict with statutory requirements established by SB 198 and AB 377. Those laws required dual-track service and connections to downtown stations for the Merced-to-Bakersfield segment.

The Legislative Analyst’s Office also criticized the authority for obscuring major project reductions by describing them vaguely as “optimization measures.”....

....MUCH MORE

*France Tried To Warn California That The California High-Speed Rail Plan Was A $100 Billion Farce 

We've been posting on the massive grift and fraud that is the California HSR for many, many years. The New York Times just did a feature on the project that is not nearly as hard-hitting as it could be, being disingenuous from the headline on:

How California’s Bullet Train Went Off the Rails

It was never "on" The Rails, a fact that was pointed out over a decade ago.

However, to their credit, the Times writers and editors included this vignette: 

....The state was warned repeatedly that its plans were too complex. SNCF, the French national railroad, was among bullet train operators from Europe and Japan that came to California in the early 2000s with hopes of getting a contract to help develop the system.

The company’s recommendations for a direct route out of Los Angeles and a focus on moving people between Los Angeles and San Francisco were cast aside, said Dan McNamara, a career project manager for SNCF.‌

The company‌ ‌pulled out in 2011. “There were so many things that went wrong,” Mr. McNamara said. “SNCF was very angry. They told the state they were leaving for North Africa, which was less politically dysfunctional. They went to Morocco and helped them build a rail system.”

Morocco’s bullet train started service in 2018.....

There you have it, North Africa is less politically corrupt than California. Just amazing.

Meanwhile, Elsewhere In the Muskonomy: "Anthropic to Pay SpaceX Nearly $45 Billion for Computing Deal"

Following on "SpaceX will launch its 1st-ever Starship V3 megarocket today. The stakes couldn't be higher", from Bloomberg, May 20:

Anthropic PBC has agreed to pay Elon Musk’s SpaceX nearly $45 billion over the next three years for computing resources as part of an expanded deal to support its Claude artificial intelligence software, according to a securities filing.

AI developer Anthropic is expected to pay Musk’s firm $1.25 billion per month through May 2029, with “capacity ramping in May and June 2026 at a reduced fee,” SpaceX disclosed Wednesday in paperwork related to its initial public offering. Either party can end the agreement with 90 days’ notice, the filing said.

Anthropic earlier this month said it inked a deal to access more than 300 megawatts of computing capacity from a large SpaceX data center in Memphis known as Colossus 1, without disclosing the terms. The startup has since expanded the partnership to include capacity at a second SpaceX data center, according to a post from Anthropic co-founder and Chief Compute Officer Tom Brown.

Anthropic declined to comment beyond pointing to Brown’s post....

....MORE