It’s Mad, Mad World

Perhaps you’ve heard of the Anna Karenina Principle, based on the following quote from Leo Tolstoy’s 1878 novel: “All happy families are alike; each unhappy family is unhappy in its own way.” Authors have cited the quote to explain why they are drawn to write novels of misfortune and tragedy rather than stories stuffed with bliss and contentment.

Beyond literature, the principle posits that a deficiency in any of several factors will condemn an endeavor or project to failure. Thus, according to the principle, a successful initiative is one in which every possible deficiency has been avoided.

Success and happiness are hard to achieve, and even more difficult to maintain. Moreover, nothing — certainly not unalloyed success — lasts forever. This, my friends, is one of the reasons why most news is bad news.

Each day, one after the other, innumerable newsworthy events, of perceived or real significance, occur. There’s no way to keep up, especially now that we live in a digital era where information travels at the speed of light and there’s a market, or at least an audience, for seemingly endless varieties of information.

You can’t consume, much less analyze, the entire information banquet, so you must pick only a few items from the menu. Even then, many choice cuts fall from the table. As I review a few reports that perhaps only resonate in the twisted canyons of my mind, allow me today to serve as your arbitrary and subjective news curator. I’m not an algorithm, alas, only a person, so consider the inherent limitations.

As I monitor what passes for news these days, I often find myself asking, “How did we get here?” The here to which I refer is not my physical location or yours, but to this particular point in human evolution where accelerating technological progress intersects with creeping irrationality and smug vacuity.

It’s a paradox, the latest in a long sequence of paradoxes. We should be gaining intelligence, standing atop the observation towers of information-technology’s superstructures. Instead, we seem to be regressing in our capacity to reason and to truly think as individuals. We’re overwhelmed and stultified by ostensibly helpful algorithms and, more recently, artificial intelligence. Perhaps our current state of affairs is no accident, and perhaps it is easily, or at least credibly, explained by the blithe complacency that attends automation. Being able to explain or rationalize the situation, however, doesn’t make it any more palatable.

AI’s Pointless Chess Tussle

Let me get down from my soapbox. There, that’s better. I’ll now dig into three news items. These aren’t necessarily the most substantive items I could find, but they’re acceptable candidates. What further recommends them is that they can be addressed briskly. Other items, on themes I might address in fuller scope at a later date, weren’t as amenable to concise dispatch.

First up, we learn that OpenAI’s o3 model comprehensive defeated xAI’s Grok 4 at chess, scoring a 4–0 victory in the final of the Kaggle AI Exhibition Tournament. Other AI competitors in the chess tournament included Anthropic, DeepSeek, Google, and Moonshot AI.

The cornermen of the two AI chess finalists were, of course, Sam Altman and Elon Musk, who have a history of cooperation, followed by bitter conflict, at OpenAI. If the competition had involved gratuitous insults, scored by international AI judges, I’m reasonably confident that Musk’s Grok would have triumphed splenetically.

Alas, it was chess competition, and I’m not sure what it proved. At one time, neural networks and AI were consistently capable of beating of humans at chess, but the tables, and the chess pieces atop them, have apparently turned, as this excerpt from an article in The Independent notes:

“Kasparov vs. Deep Blue was a huge spectacle in the ‘90s... What’s important from this tournament is to understand more about how these products think and reason,” Mats André Kristiansen, CEO and co-founder of Take Take Take, told The Independent.
In July, current world number one chess player Magnus Carlsen beat ChatGPT in an online chess match without losing a single piece.
Commentating on the final between Grok and o3, Mr Carlsen estimated that Grok had a rating of around 800, while OpenAI’s model was at around 1200. By comparison, Mr Carlsen’s peak rating is 2882. When asked directly by a user on X, Grok estimated its rating at around 1600-1800.

So, Grok has an inflated sense of its chess acumen. I wonder where it got that from? This AI chess tournament, as so much else in the AI world, seems partly to be a proxy-based pissing match between tech billionaires. Score this one for Altman, if you must, but any broader relevance is difficult to discern.

Trump’s Drive-By Attacks on CEOs

Donald Trump, you may have noticed, has opinions. He has no end of opinions, on almost every field of human endeavor — from entertainment to sports, from geopolitics to economics, from the culinary arts to the fine arts (though his appreciation of the latter extends only as far as gold trim and marble statuary), from business and technology to energy and natural resources. I think he has an opinion on everything. To his credit, he doesn’t let his ignorance constrain his scope.

We should not be surprised, then, to see him render a strident opinion on the stewardship of Intel’s current CEO, Lip-Bu Tan. Trump’s gripe is that Tan, through his past business entanglements, might be too close to China. I don’t know whether those charges are fair or accurate — though there’s a valid suspicion that Trump is using the attack for negotiating leverage — but I do know that it is wholly inappropriate for the president of the United States to publicly undermine the market valuation of a publicly listed company.

Look, even if Trump had suspicions about Tan’s relationship with China, any concerns should have been conveyed privately to the proper authorities, legal and otherwise. There are well-established channels and processes for dealing with such matters.

By taking his grievances public, without first reaching out confidentially to Intel’s board or the proper investigative authorities, Trump has done nothing besides roil the markets, sending Intel’s shares further below ground. You might even describe what Trump is doing as akin to market manipulation, uncomfortably close to a form of extortion.

Consider that Trump presumably knows what he’s going to say publicly or state on social media before he says it, and that he presumably knows (or should know) that what he is about to say will affect the market. Perhaps he’s even told some of his acquaintances what he’s about to say or do. Those people could do any one of a number of things, some of them decidedly unethical, with that information.

Lest you think this is a one-off exception, and that Trump is usually a model of presidential decorum and probity, I ask you to think again. He has done exactly this sort of thing on several occasions, publicly assailing the CEOs of listed companies, knowing full well that his antics will influence the market. Such misbehavior invites suspicion.

People, this type of public verbal incontinence, invariably emitted as semi-literate bilge, isn’t normal for a U.S. president. It’s — and I know this will seem quaint in the current context — unbecoming of the office. Yes, yes, I know: Trump is not a typical president, but that doesn’t make his behavior right, and it begs the question of why so many people indulge antics that would be unacceptable from nearly anybody else.

Before somebody — I know you’re about there — jumps down my throat and counters with a vituperative volley of whataboutism, I request politely that you hold your fire. My objections aren’t partisan. I have no horse, donkey, or elephant in the race. My objections are to crass stupidity and crude venality. I can’t stop all the corruption and ignorance of in the world, but I don’t have to like it. Neither do you. In accepting the unacceptable, you normalize patently abnormal behavior.

I’ll say one more thing before I move to the next news item: Intel doesn’t need anybody outside its walls to beat it into submission. The company has made, and will continue to make, a hash of its affairs. Intel, including its board of directors and its executive braintrust, has indulged in more self-harm over the last decade or more than you’d see exhibited by a chronic alcoholic or a drug addict. The company is its own worst enemy, as I’ve documented here previously.

I have no vested interest, and definitely no investment, in Intel. Trump’s latest tirade doesn’t affect me financially or materially, but it feels very much like a case of piling on. Intel is already moribund. Trump doesn’t need to kick it while it’s down. If he wants to take a run at a publicly listed company, even though we’ve already determined that he should refrain from doing so, he can go back to picking on Apple or railing at Musk’s Tesla. They can handle the abuse.

AI’s CV Short on Personality

There’s still a lot of hysteria out there about AI’s alleged capacity to plunder and pillage every job in sight.

Folks, as I’ve said here before, that’s not going to happen. People will still have jobs to do, and plenty of them, for the foreseeable future, which admittedly isn’t as far out on the horizon as it used to be. Don’t get me wrong: AI will take jobs, including those in what are colloquially called white-collar professions. The jobs that are susceptible to AI, however, are restricted to closed, repetitive processes — like some accounting, HR, and software development functions, to name a few — that do not involve dynamic, intensive, reciprocal interpersonal exchanges.

There are countless jobs that will continue to require ongoing personal interaction, in sales, marketing, product development, engineering, and many other roles. Will AI perform tasks that once required human employees? Yes, it seems so. Will AI be applied to complement, enhance, or expedite the performance of tasks executed by humans? Yes, probability suggests that it will. There’s a limit, though, as to how far businesses and industries can take AI.

This excerpt from a Wall Street Journal commentary — in addition to an article, mentioned in a prior post, suggesting that we might have to turn to sex work en masse — is precisely the sort of maddening absurdity that caused me to begin this piece with my soapbox oration:

There are jobs AI likely won’t touch; Ellis offers Microsoft’s list of 20 such professions. They include floor sanders and finishers, roofers, motorboat operators, massage therapists and pile-driver operators: “The vast majority of the companies and businesses I see when I drive up and down Old Post Road don’t offer the services above.”

I’m flabbergasted. Seriously, my flabber is gasted. I don’t know what to say, except that the folks at Microsoft might be doing a little more than micro-dosing on hallucinogenics. The AI sales department at Microsoft needs to pull on the reins. AI isn’t a neutron bomb. It’s not going to empty all the office towers in New York, London, Tokyo, Toronto, or Beijing. You know what did that? COVID. And that was temporary.

Personally and professionally, people still like to deal with other people. They like to hear real-time verbal interactions and see and interpret non-verbal signals. I have a feeling — just a feeling, mind you — that we will rediscover and learn to cherish the eccentric and idiosyncratic characteristics of companionable human behavior, of uniquely human correspondence and quirks.

AI is a homogenous, probabilistic potpourri, a bland stew that regurgitates off-the-shelf ingredients. It has its uses, of course, but, as humans, we will still crave the flavor and spice of human interaction.

That’s enough for now. Enjoy the weekend. This week’s song is about defiant individualism, about what makes each of us, in our sensibilities and tastes, distinct and, yes, even unique.