AI Decoded: Threat, money and power

/
 

By Don Varyu

Oct, 2026

 
 

here is no definitive answer yet on whether Artificial Intelligence will take over our lives. But no doubt, it’s already made a deep wormhole into our consciousness.

Experts disagree on whether this technology is essentially different from previous advances that changed society (steam engine, electricity, railroads, Internet, social media, etc.) As with all these predecessors, AI clearly is changing society. What is different this time is that not only is this newest technology altering our lives, but it could also end them. This is something that seems worth addressing. 

It’s all troubling and mystifying but here are three thoughts that might help  clarify things. 

THREAT

The ongoing AI furor erupted again with a warning from a former Anthropic (Claude) employee who claimed development of the most advanced  “frontier” AI models could become so knowledgeable and so powerful they could erase humanity within a decade. (For example, if a prompt were given to a model to eliminate greenhouse gases from the atmosphere, AI could decide by itself that the easiest way to do that is by was eliminating humans.)

That got people’s attention. 

Anthropic chief Dario Amodei has a long record as a voice for conscience and safety in the industry. He wasted little time endorsing the possibility of such a threat. He called on the industry to “pace” development until safeguards could be assured.

Then came almost immediate agreement from both Sam Altman (Open AI/ChatGPT) and Elon Musk (x.ai, Grok). These three guys agreeing on anything is particularly notable. Together, they once co-founded the company that eventually became ChatGPT. But they argued and bitterly separated and remain unfriendly today. 

Meanwhile, other major AI tech leaders were not quite so frantic. Google couldn’t quite figure out a cohesive stance. Its top leadership said, in effect, “yeah, extinction is something we might need to figure out.” At the same time, the guy who founded their AI initiative strongly was not at all troubled, arguing that the same supercharged frontier models that could create such a threat could also be used to defuse it. 

Most of the remaining big players (Microsoft, Meta, Amazon, etc.) said something along the lines of ,  “yeah, we need to talk—but in the meantime, we’re not slowing down.”

And there is one more distinct point of view—one that rejects the doomsday scenario in total. Primarily, this includes chipmaker Nvidia and Donald Trump. In fact, Trump, first labeled the whole extinction thing “a hoax”…but later relented a little by announcing a new AI “czar” to give the whole thing a  rethink. Which could also be a head fake to protect against even angrier voter reaction. 

The bottom line here is simple. If there really is no threat, then any efforts to contain such a thing causes no harm.

But if there is real danger involved…doesn’t it make sense for everyone to come together and erect the necessary guardrails—NOW!?

MONEY

This intellectual argument leaves any non-STEM PHD wondering why the main players wouldn’t immediately start working together. 

And the answer, in a single word, is money. 

nfluential business/finance podcaster Scott Galloway has been preaching for more than a year, “the U.S. economy is just one giant bet on AI.” Data backs this up. 

Some estimate that two-thirds of all stock market gains since the start of 2025 are directly attributable to AI. And looking forward, it’s estimated that this year the balance sheets of 80% of all U.S. companies will be affected by AI--one way or the other. AI may not have captured your world; but it’s kidnapped Wall Street.

Among the handful of “bro-ligarchs” directing all this, one will win; and the others will be left behind. None of them is emotionally equipped to be a loser. Each sees himself as the one who will grab the maximum boost in both corporate reputation and personal wealth. Each believes he MUST be that man. And for most, that drive suffocates any meaningful interest in restricting threats.

inancial claims on both sides are fuzzy. Opponents say a wealth exodus will drain between $1billion and $25 billion from state coffers. Proponents claim the opposite is true—that it will generate a new $100 billion windfall. Believe what you want.  

The latest independent pols show a close race: a virtual coin flip.

ut that doesn’t mean there aren’t business and financial obstacles to overcome. 

  • Collectively, the technobrats have invested over a trillion dollars in capital expenditures—buying and building things they hope will pay off. Most noteworthy is construction of those noisy, ugly, and oh-so-despised data center megaliths. They feel like they’re building the future…and they resent the lowly people who stand in their way. Negative publicity is unhelpful. 

  • Anthropic/Claude and OpenAI/Chat GPT currently are the undisputed kings of cutting-edge development. And they’re anxious to leverage that leadership into more dollars. Both announced Initial Public Offerings of stock (IPOs) that value their companies not in the hundreds of billions, but in the trillions—Trillions, with a “T.” 

  • Once executed in the market, dozens of new people will become instant billionaires. More negative PR in this. 

  • And the worst time for all of this to be on display is right before the midterm elections. So mysteriously, both IPOs have been rescheduled; Anthropic’s until closer to Thanksgiving…and OpenAI’s until some unspecified date. There is no reason to stick heads out of bunkers right now.

  • But even here, there’s a catch. At this writing there is a minor downturn happening for some of the central AI-related stocks. No one’s seriously talking “bubble”--yet. But if those IPOs linger too long, fortunes could be slashed. Such are the burdens of the world’s richest men. 

nd money is also of direct concern to the two vocal “hoaxers,” Trump and Nvidia chief Jensen Huang.  

  • During the ongoing market boom, no one has benefited more than Nvidia. Its stock price today is 16x higher  than it was just five years ago. On one day alone in 2023, it leaped 23%.

  • However, the only way it can  possibly maintain that lofty trajectory is to design and sell more chips. So, Huang’s assessment is, “nothing to see here--full speed ahead.”  That’s his only play.  

  • As for the Trump family, the corruption here is outrageous, even by its own slimy standards. Led by the nepo-spawn sons and an equally loathsome son-in-law, the family has concocted a witches’ brew of AI investments, data center and other infrastructure construction, cryptocurrency scams, military contracting, computer chips, and venture capital. A cadre current master’s student in accounting could graduate and then spend entire careers unraveling all this—and never get to the bottom on it. 

  • But let’s focus on just one Trump scam. In his first term, Trump publicly slammed cryptocurrencies, saying they were “based on thin air”…tied to the “drug trade”…and were a direct threat to the U.S. dollar. 

  • After losing to Joe Biden, he had more time to listen to the evil troika (Don, Jr., Eric and Jared). They showed him how he could make ridiculous amounts of money by splashing the Trump name across equally ridiculous ventures. Thus, after reelection in 2024, Trump did a trademark 180, declaring America would be the “crypto capital of the planet.” 

  • AI investments became an offshoot of this tentacled fraud. 

  • Trump then employed his superpower of defining a lethal enemy, naming China as the AI bogyman. And the stakes were high; in his words, “whoever wins AI…wins.” 

  • So Nvidia was blocked from selling its most powerful chips to China—until they weren’t. The ban ended when Nvidia agreed to play a licensing fee of somewhere between 15% and 25% of those sales to the U.S. government. (So the enemy is also a client?)

  • Meanwhile, the nepos announced in a press release that a crypto “stablecoin” the family owned could be used instead of cash to pay a firm also run by the family which sold access to more than 40 Chinese AI entities. (Rules don’t apply to everyone). 

  • Investigative reporters say that the Trump family has realized at least $2.3 billion dollars in revenues, just since the start of the second term. And many of their investments are not publicly recorded.


POWER  

So yes, AI is about reputation and wealth—but most of all, it declares who exerts power. Think about the ability to aim the most lethal technical weapon in history at another corporation…another nation…or another political party. 

Think about AI endlessly drilling down into your tax returns…your business records…your child support payments…your speeding tickets. 

Whether or not AI could execute an extinction of humanity, it certainly seems inevitable it will gain the ability to expose you…and rewire the world we live in. 

How might we stop this?

We did not allow private companies to develop nuclear bombs. Similarly, you cannot announce that you’re forming a rival  “private army” to compete with the U.S. military. You can’t go out and buy or build armed drones and missiles and fighter jets and submarines…and equip a multi-million-man private army—just because you want to. These threats remain the province of the government—a government we elect. 

The reason we do this is because of Silicon Slytherin like Elon Musk, the world’s richest man. He has stated, “the fundamental weakness of Western civilization is empathy”…and called such empathy “suicidal.” We all saw what this looked like when he ran DOGE. More than 350,000 government employees were sacked. A countless number of young kids in Africa died after a cutoff of U.S. medical aid. And Musk did all that without AI. 

onsequently, for AI the only logical alternative in terms of laws or regulations or guardrails--or common sense--is the federal government. And not just government “oversight”--government ownership. 

Some say this is stupid because the government “can’t run anything!”  I would counter with the FAA, the SEC, the Centers for Disease Control (before RFK, Jr.), and even Social Security. Yes, the fund is running out of money—but that’s the fault of Congress. The SSA efficiently runs a business with 70 million “customers”…and another 185 million workers currently paying into it. That’s a total population larger than Germany, England, or  France. In short, even Social Security works.

Thus, I will always trust the federal government over the empathy-free  Elon Musk. 

’ll conclude by reverting to the threat.

We can’t risk reducing  AI to a typical construct of nation vs. nation. What difference will it make if America “beats” China…if the frontier AI models decide to erase all of us? 

The question is not what threats are real.

It’s what threats are possible.