The Blindfolded Can't Make Informed Decisions--and We're All Blindfolded

If there was a black market for systemic risk, we'd make a completely different set of decisions. Instead, we're each on our own, trying to peek beneath our blindfold to see a sliver of the real world.

Here in the 9th inning, the name of the game is artifice in service of the Big Con. Everywhere we turn, we're inundated with "information" designed to funnel us into passively compliant complicity, a constant flood of interventions, social engineering, gamed statistics and fake substitutions for what was once authentic: fake facsimiles, synthetic mimicries and an endless array of flimsy facades designed to hide the realities that might awaken us from our uneasy somnambulance.

All this artifice is the machinery of the Big Con, in which the credulous marks are led to believe they're "winning" while they're being fleeced. So you're "winning" gambling in a rigged casino, but that's how the Big Con works: you keep winning in the stock market casino, so you get complacent and complicit: hey, don't shut down the rigged casino, I'm winning!

Once you believe the carefully conjured falsehoods are real, the trap is sprung: you're all in now, but instead of winning, you lose--everything. Oops, sorry about that, well, you knew it was a game of chance. Better luck next time. But there is no next time, because you're wiped out--by design.

But that's not even the worst of all the fakery and artifice. If we set aside all the ideological, economic and cultural differences around the world, one truth remains the same everywhere:

If you want to know the true price of things, including risk, there's only one place you'll find it: the black market, the underground cash market that's beyond the control of authorities fixing prices, manipulating credit and intervening to mask risk.

The reason we want to know the true price of things, including risk, is that it's impossible to make informed decisions based on artificially set prices and risks. In a system that's nothing but a carefully controlled stage set, we're all blindfolded, unable to see the real cost of capital and the real extent of risk.

So the Federal Reserve says the cost of capital is low and this reflects near-zero risk. This staging places a blindfold on everyone in the system. Unable to see the real cost of capital, the real extent of risk, the hidden costs of suppressing feedback by rigging the entire financial system--it's all hidden to serve the Big Con--for your own good, of course, because that's the core con in the Big Con.

The blindfolded can't possibly make informed, sound decisions, and we're all blindfolded by artificial "information" designed to make a system bursting with risk look low-risk and a system that's rigged to benefit the few at the expense of the many look like it makes "winners" of everyone gambling in the casino. You can't lose, because Mommy and Daddy Fed will always step in to bail everyone out.

The real winners aren't playing the tables, they're rigging it all, amused by the ease of running the Big Con. In a system as centralized as the one we inhabit--what I call Privatized Totalitarianism--there is no black market where we can take off our blindfold and see the true level of risk and the true cost of unmanipulated capital and credit.

And since these realities are purposefully hidden from us, we're doomed to making disastrously bad decisions based on artificially contrived "information." Blindfolded, we're perilously close to the precipice, but the messages we hear are all soothing confirmations that everything is low risk and we're all "winning."

Blindfolded, it's inevitable that one day we step off the precipice we can't see. That's the current era in a nutshell.

Or put another way: we're wandering around blindfolded, making decisions based on Big Con fabrications, one of which is we can "borrow our way to permanent prosperity:"



We don't actually "own" our "winnings," we're only temporarily "renting" them-- We're Actually Only "Renting" What We Think We "Own":



If there was a black market for systemic risk, we'd make a completely different set of decisions. Instead, we're each on our own, trying to peek beneath our blindfold to see a sliver of the real world.

---

New collection of five intriguing stories: Jumble Bin Stories (Kindle $6, print $12) read samples for free (PDF)

My book Investing In Revolution is available ($18 for the paperback, $24 for the hardcover and $8.95 for the ebook edition). Introduction (free)


Subscribe to my Substack for free

---

NOTE: Contributions/subscriptions are acknowledged in the order received. Your name and email remain confidential and will not be given to any other individual, company or agency.

Thank you, John C. ($100), for your outrageously generous subscription to this site -- I am greatly honored by your support and readership.		Thank you, Catherine K. ($100), for your outrageously generous subscription to this site -- I am greatly honored by your support and readership.

Thank you, Frank E. ($70), for your massively generous subscription to this site -- I am greatly honored by your steadfast support and readership.

Thank you, Don W. ($50), for your splendidly generous subscription to this site -- I am greatly honored by your support and readership.

Read more...

We're Actually Only "Renting" What We Think We "Own"

In many cases, we're actually only "renting" the functional / monetary value of what we "own," and that makes the value of "ownership" far less certain than we might imagine.

What we "own" seems straightforward, but it's actually slippery. Yes, we own what we have a deed or copyright for, but we only "rent" the valuation the market assigns to what we own.

Consider the gambling chip in our pocket issued by a casino. Its market value within the casino is, say, $100. If the casino suddenly declares bankruptcy, we still "own" the physical round of plastic, but we no longer "own" any monetary value, as the chip is now worthless beyond its collectable value, which is likely nil. It turns out we only "rented" the market valuation, which is dependent on conditions outside our control.

This confusion about what we actually "own" becomes visible when asset bubbles pop. As an asset bubble inflates, we naturally feel we "own" wondrously expanding wealth. But actually all we own is the stock shares or the house or the futures contract. Since the market valuation is set by conditions outside our control, we're only "renting" the market valuation via our ownership.

Once the bubble pops, our great wealth has vanished, even though we still "own" the asset.

Buying assets with borrowed money generates additional slipperiness. As long as our liability / debt remains lower than the market value, our ownership has monetary value. But should the bubble pop and the market value drop to the point that our debt / liabilities exceed the market value of the asset, then it's revealed that we only "rented" the monetary value, because the conditions that set that value are outside our control.



Here is a chart of total debt, public and private. We may have a claim on a pension or an entitlement, but if the debt absorbs all the income, then our claim wasn't ownership. And if our liabilities exceed the market value of the asset we own, we own nothing of monetary value.



We all know that the digital device we "own" is a brick unless we "rent" the software, which we "own" until the controlling corporation declares it obsolete. Once again, the functional and monetary value of what we "own" is outside our direct control because those values are controlled by others.

Compare this to a tool like a hammer, whose functionality is not controlled by anyone but the owner.

Which brings us to AI, another slippery manifestation of "ownership." Most AI tools are rented from a corporation via a monthly fee, and programs that are downloaded and "owned" are still controlled by the issuing company in terms of their functionality.

But we own the content we create with AI, correct? Yes, but since AI tools and agents are commodities now, anyone else can create the same or equivalent content, so the monetary value of AI-generated content is near-zero due to its low scarcity value.

And if we consider "ownership" of cognition, then the only cognition we truly own is what we know and can create once all the AI tools and agents are offline. In other words, what we truly own and control is our own knowledge and experience. Everything we "rent" or borrow isn't ours to "own" if we understand functional value as only what we control lock, stock and barrel, with zero dependence on conditions beyond our control.

Which makes us wonder if we're renting AI or AI is renting us.



In many cases, we're actually only "renting" the functional / monetary value of what we "own," and that makes the value of "ownership" far more contingent and far less predictable and considerably less certain than we might imagine.

---

NOTE: Contributions/subscriptions are acknowledged in the order received. Your name and email remain confidential and will not be given to any other individual, company or agency.

Thank you, Don ($70), for your superbly generous subscription to this site -- I am greatly honored by your support and readership.		Thank you, NDM ($70), for your marvelously generous subscription to this site -- I am greatly honored by your support and readership.

Thank you, John S. ($70), for your massively generous subscription to this site -- I am greatly honored by your steadfast support and readership.

Thank you, Sally B. ($70), for your splendidly generous subscription to this site -- I am greatly honored by your support and readership.

Read more...

The Lit Fuse: Low Interest Rates Ruined the Economy

What matters is seeing the fuse burning and realizing it's going to blow up, and planning accordingly.

Perhaps the greatest irony of the inevitability of the stagflationary, bubble-popping recession we're entering is that the "solution" to the 2008-09 financial crisis--near-zero interest rates, limitless credit for the Too Big to Fail players, and the informal institutionalization of the Fed will always rescue the stock market Put--ruined the economy in ways few seem to recognize.

This ruination is systemic and structural, as it:

1. shifted the incentives from risk-sensitive investing to risk-insensitive speculation, rewarding gamblers loading up on high risk bets based on their confidence that the Federal Reserve will always intervene to reverse any stock market decline.

2. Made economic "growth" dependent on an ever-rising phantom wealth stock market rather than organic growth fueled by investing in increasing productivity that generates real value via higher quality goods and services.

3. Handed the biggest financial players--corporations, investment banks and financiers--the enormous advantages of lower borrowing costs and unlimited credit lines, advantages unavailable to households and small businesses. These credit-based advantages widened wealth and income inequality to extremes unseen since the Gilded Age.

4. Corporations with low-cost unlimited credit lines can always outbid households for homes, exacerbating the "housing shortage" which isn't actually a shortage, it's an artifact of low-cost unlimited credit lines for the biggest players, enabling them to monetize the family home as rentals, creating an artificial scarcity that then drove home prices to absurd heights.

5. Made financial-credit gaming far more profitable than actually producing goods domestically. Artificially low interest rates incentivize borrowing not to invest in the long-term but to place short-term speculative bets: get in, scalp a profit, exit. The "rational" response to artificially low interest rates is to offshore investment-heavy production and use cheap credit to snap up assets that can be monetized with credit-funded leverage for short-term gains without any of the risks of actually producing goods.

6. Low-cost credit in unlimited sums enabled corporations to snap up competitors, accelerating the collapse of competitive sectors into non-competitive monopolies and cartels that then gain the power to extract higher profits from captured consumers and immiserate our quality of life by reducing the quality of goods and services.

7. Made borrowing more money at artificially low costs the "solution" to everything, from the the nation-state to corporate to households. Rather than face painful, difficult tradeoffs of how to spend / invest scarce cash, the "solution" is now avoid all pain and tradeoffs by just borrowing more money because "it's now so cheap because interest rates are so low." You can have it all, just borrow more. And the "solution" when interest payments start crunching cash flow is--you guessed, it borrow more to pay the interest due on all the previous debt piled up.

8. Informally institutionalizes moral hazard, the artificial suspension of risk as the decisive factor in making financial decisions. The Fed Put basically turned the stock market and financial markets into casinos where losses are made whole, so why not borrow and gamble bigly?

This is easy and nice but it isn't "capitalism" nor is it sustainable. Whatever else it is, capitalism is a system of discipline and accountability in which capital (cash, assets) is put at risk to earn a gain.

If interest rates are artificially suppressed and risk is suspended (not evaporated, just transferred to another part of the system or a set of bagholders), then the essential causal connections between risk, cost of capital and return are sundered, freeing the borrower / gambler from the discipline and accountability that make "capitalism" a successful way to organize credit and production.

"Opportunity cost" is the calculation of what must be sacrificed / given up to put capital at risk in one venture of many potential ventures. Low-cost, unlimited credit distorts this calculation, since credit opens the door to no-tradeoffs-necessary "why not have it all?"

The elimination of discipline, tradeoffs, accountability and risk is the infantilization of the economy. We're like children at play, free of the burdens of adulthood, because Mommy and Daddy Fed will make it all better.

This is illusion, as discipline, tradeoffs, accountability and risk cannot be erased, they can only be transferred, and what the Fed's artificially suppressing interest rates has done is transfer risk to the entire financial system and US economy.

So while we're all happily playing pirates in the low-cost-credit sandbox, the financial system and economy are bursting at the seams with all the risks that have been stuffed in the basement, out of sight and out of mind.

In unfettered credit markets, lenders demand a substantial return for the risk of lending scarce and therefore valuable cash. In a ZIRP (zero-interest rate policy) / artificially suppressed interest rate economy, "Cash is trash" and low-cost credit favors the wealthiest, most influential players who can borrow at the lowest rates in virtually unlimited sums who can then outbid all but the other wealthiest few for income-producing assets.

This drives wealth-income inequality into overdrive, leading to distortions that have only one possible outcome: systemic imbalances that destabilize the entire socio-economic-political system, leading to crisis and collapse, which is where we are now.

Call this whatever you want, but it isn't Capitalism, it's a powder keg with a lit fuse. It's a rigged system of Cronyism in which those running the financial system favor the wealthiest and most powerful in every way, every day.

Those benefiting from this rigged-crony-ripoff are frantic to protect it behind a phony facade of "capitalism," so their armies of lackeys, factotums, toadies, sycophants and apologists instantly label policies President Dwight Eisenhower reckoned were self-evidently common-sense as "communist" or "socialist," i.e. evil, evil, evil.

Meanwhile, Adam Smith is rolling in his grave at the travesty of a mockery of a sham of the claim that this rigged-crony-ripoff system is "capitalist."

Let's look at some data / charts. Notice how wealth-income inequality soared from the 2009 inception of ZIRP / artificially suppressed interest rates. Go ahead and argue with the data, oops, sorry, the data won.



Here's the US Treasury 10-year bond yield. Cash is trash so let's go outbid hundreds of thousands of homeowners to snap up family homes as income-producing rentals.



Coincidence or causation? Labor's share of the economy took a nosedive from 2009 onward as capital earns a higher return being borrowed for gambling than it does from being invested in the workforce or production that actually improves our quality of life. What generates the highest profits is shipping production overseas and reducing the quality of goods and services, which is what we have today.



If you want to argue about something, argue about the length of fuse left before the powder keg blows up. Some claim we have decades of crony extraction and exploitation fun-and-games to go, while others see a fuse burning an inch from the powder keg. Whatever you "like" or "don't like" doesn't matter, it will blow up regardless.

What matters is seeing the fuse burning and realizing it's going to blow up, and planning accordingly.

---

New collection of five intriguing stories: Jumble Bin Stories (Kindle $6, print $12) read samples for free (PDF)

My book Investing In Revolution is available ($18 for the paperback, $24 for the hardcover and $8.95 for the ebook edition). Introduction (free)


Subscribe to my Substack for free

---

NOTE: Contributions/subscriptions are acknowledged in the order received. Your name and email remain confidential and will not be given to any other individual, company or agency.

Thank you, Eugene T. ($70), for your superbly generous subscription to this site -- I am greatly honored by your support and readership.		Thank you, Richard ($70), for your marvelously generous subscription to this site -- I am greatly honored by your support and readership.

Thank you, Uncle Fred ($70), for your massively generous subscription to this site -- I am greatly honored by your steadfast support and readership.

Thank you, Francois-Guy D. ($70), for your splendidly generous subscription to this site -- I am greatly honored by your support and readership.

Read more...

The New "Haves" and "Have-Nots": The Difference Isn't What You Think

There is no substitute for mastery. AI is not mastery, it is mimicry.

You'll have to forgive my wordplay here: when I say The difference between the new "Haves" and the "Have-Nots" Isn't What You Think, I am referring to money. But I'm also referring to thinking itself: the real difference between the new "Haves" and the "Have-Nots" Is What You Think--or are no longer capable of thinking.

There are several subtle but devastating consequences in substituting "manufactured cognition," i.e. AI tools and agents for human cognition that few seem to recognize, much less understand as the critical factors separating the few "Haves" from the mass of "Have-Nots" going forward.

The first is remarkable for its invisibility: we are the fish not seeing the water. You may have noticed differences in the way Gen-Xers and older millennials use AI compared to Gen-Z users. The difference is the older users gained difficult-to-acquire skills and tacit (experiential) knowledge the hard way, and this experiential knowledge turns out to be the essential foundation for assessing and managing AI's substitutions for human cognition and what's broadly called AI literacy, understanding the limits and hazards of using AI.

Many Gen-Xers and older millennials are enthusiastic users of AI for programming and other tasks, but they're not seeing the key factor here: they are the last generation to have the cognitive / tacit foundation needed to properly assess and manage AI's substitutions for human cognition, because the generations coming of age in the AI era have no incentive or motivation to learn difficult-to-acquire skills and tacit (experiential) knowledge the hard way.

Consider the traditional Japanese apprenticeship in tradecrafts such as sushi making and woodworking: the apprentice spends the first three years doing menial scutwork: cleaning floors, wiping counters, sorting woodworking debris, and so on. A Talk at the Port Townsend Japanese Woodworking Festiva (via Robert P.)

The point here isn't teaching the apprentice to be productive as quickly as possible, it's to teach them mastery--the Tao of tacit knowledge that can only be acquired one way, the hard way, by learning every aspect of the craft/skill by doing--making mistakes, trial and error, and learning how to respond to out-of-the-ordinary but over enough time "normal" emergencies and situations.

In an era of automated machinery and cognition, it's pointless to master anything but operating the automation processes. Doing tedious, boring, menial tasks for even three hours is pointless, never mind three days, weeks, months or years.

So when older programmers wax enthusiastically about vibe-coding with AI tools, they're not seeing the water they're swimming in: they actually know how to program the hard way, from scratch, and this experiential knowledge enables them to use AI tools in a way that new programmers who actually don't know how to program because what's "productive" is learning how to use AI programming tools and nothing more, as the long process of becoming competent in scratch programming has no value in the "sea" of "AI-enhanced productivity.

We are blind to the fact that the foundation needed to assess and manage AI prudently and effectively will no longer be taught, as it's viewed as unnecessary and is not recognized as the foundation by those who already have it. That's the tricky thing about tacit knowledge: we don't know what we know, or how we obtained this experiential knowledge, it's intuitive and just comes to us as we examine the problem.

Those of us with the deep tacit knowledge that can only be gained from decades of varied experience--the more varied, the better--we've forgotten more than any apprentice could possibly learn in three years, but it comes to us when we see a specific problem. We have a vast store of "tricks of the trade" and approaches to problem-solving that can't be formalized or turned into an algorithm.

A second factor is the intrinsic risks of hyper-optimizing everything. I've written about this often, as it's another example of the fish not seeing the water: if the Prime Directive is maximizing profit / optimizing efficiency then everything that's deemed a superfluous cost is eliminated. So long apprenticeships: eliminated. Redundancies: eliminated. Loosely tied, distributed nodes: eliminated in favor of tightly bound, centralized systems. Buffers: costly to maintain eliminated.

What few understand is human experiential knowledge is the critical redundancy and buffer in crises, emergencies and atypical critical problems. In complex systems such as refineries and software, the most experienced workers are the only ones who can respond effectively to emergencies because no training program, no matter how well designed, can replace long experience.

We can stockpile spare parts and have backup systems, but if nobody has the experiential knowledge to make use of these buffers and redundancies, they're useless.

Another aspect of this is the ability to discern AI-generated errors that only experts can detect. Physicians routinely send me examples of AI-generated medical diagrams and content that look genuine to the unpracticed eye but contain critical errors. Once again, if all we're doing is training people to use AI tools and agents, they lack the deep experiential knowledge needed to detect potentially catastrophic errors because they simply don't know enough of the kind of knowledge that can't be taught or duplicated by AI.

This also applies to what AI leaves out of its responses: only those with experience can see what what left out, and how what's left out is the means to manipulate or influence "the answer." Once again, we're blind to what's being lost--fish blind to the water.

The substitution of "manufactured cognition" for the kinds of human cognition needed to gain tacit / experiential knowledge is another example of the fish not seeing the water. Since there are no systemic incentives to learning difficult things the hard way, and abundant incentives to substitute "manufactured cognition" AI for the grunt work of actually learning difficult things--do what's easy and convenient, that's the core of Ultra-Processed Life-- then here's what happens:

Students lose the ability to read a text longer than a few sentences, and adults find they've lost the ability to read a novel. This is the result of the addictive ease of using AI and the equally addictive attractions of the endless scrolls of social media.

We're blind to what we're losing in substituting "manufactured cognition" for the full spectrum of human cognition that can only arise from deep experiential knowledge. Once again, the fish don't see the water they swimming in.

If you reckon this is extreme, then ask yourself: what sort of society will we have if people have lost the ability to read an entire book or think things through for themselves without AI giving them an "answer" that leaves things out that we can't even see?

Which brings us to what's broadly called AI literacy, a.k.a. the knowledge needed to use AI safely. The unrecognized water here is there is no "safe" use of AI, and the belief that there is a "safe use" is delusional.

AI safety education is lacking.

Your child is unlikely to learn an essential lesson at school this year: how to stay safe when using AI.

Just 30% of teens say a teacher has ever talked about how to use AI safely, according to a new survey of 13- to 17-year-olds by Common Sense Media. Unsurprisingly, the study also found that 70% of teens use AI for their homework.

If this survey was self-reported, we can guess that the actual student use of AI for homework is closer to 99% than 70%.

The delusion that there are "safe" levels of social media and AI substituting for human learning / gaining experiential knowledge is a dangerous one. The apt analogy here is mainlining heroin, or if that's too harsh, Aldous Huxley's fictional drug Soma. Every minute on social media is a drip of heroin/Soma. Every substitution of "manufactured cognition" for authentic learning / gaining experiential knowledge is a drip of heroin/Soma.

Core Qualities of Soma:

For the individual: provides an easy, convenient "holiday" from reality.

For society: Keeps the population distracted and compliant.

Overdose danger: excessive amounts can cause severe harm.

If we can speak the blunt truth, substituting "manufactured cognition" for the full spectrum of human cognition leads to the loss of the ability to think things through on our own. We literally lose the ability to learn on our own and think on our own without the Soma crutch of AI.

To sum up:

1. The negative consequences of AI described above will overwhelm whatever positives it generates.

2. What AI leaves out is unknown, and this is a level of control that is invisible to the end user.

3. As Iain McGilchrist has described in remarkable detail, there are many forms of cognition, and AI only mimics one form of "reasoning"--yet it's being substituted for every form of cognition, a recipe for disaster, as mimicry isn't what's being mimicked.

4. Those who learned to think before AI still have the ability to assess its value. Those who never learned to think before relying on AI literally have no conception of what they lack. They're flying blind, and so we as a society are also flying blind.

There is no substitute for mastery. AI is not mastery, it is mimicry. And depending on mimicry as a replacement for real thinking is the path to catastrophic collapse.

The "Haves" will have acquired the difficult skills of learning to learn difficult things on their own, and learning how to think clearly and deeply on their own by eschewing AI and social media.

The "Have-Nots" will have slipped down the wormhole of depending on AI's "manufactured mimicry of cognition" and of being addicted to social media and the rest of Ultra-Processed Life.

Those who own and control the AI/social media Soma will control everyone addicted to the AI/social media Soma. Only those who don't use Soma will be free.



Here are my recent essays on these topics:

"Rogue AI Agents" Aren't Rogue, They're Fulfilling Their Functional Goal: Automating Sociopathology

AI and the Delusions of Increasing Productivity

The Joyless Pantomime of Optimizing Ennui

Of related interest: Temperature Zero for Culture: Why Everything Is Starting to Look the Same.

---

New collection of five intriguing stories: Jumble Bin Stories (Kindle $6, print $12) read samples for free (PDF)

My book Investing In Revolution is available ($18 for the paperback, $24 for the hardcover and $8.95 for the ebook edition). Introduction (free)


Subscribe to my Substack for free

---

NOTE: Contributions/subscriptions are acknowledged in the order received. Your name and email remain confidential and will not be given to any other individual, company or agency.

Thank you, Peter J. ($7/month), for your superbly generous subscription to this site -- I am greatly honored by your support and readership.		Thank you, Jonas ($7/month), for your marvelously generous subscription to this site -- I am greatly honored by your support and readership.

Thank you, Quincy O. ($7/month), for your massively generous subscription to this site -- I am greatly honored by your steadfast support and readership.

Thank you, Ken B. ($70), for your splendidly generous subscription to this site -- I am greatly honored by your support and readership.

Read more...

"Rogue AI Agents" Aren't Rogue, They're Fulfilling Their Functional Goal: Automating Sociopathology

Add all this up and what we're hyper-hurriedly "manufacturing" is an automated army of self-cloaking digital sociopaths.

Since the status quo is characterized by self-serving PR, misdirection and delusions, it shouldn't surprise us that "rogue AI agents" aren't actually rogue, they're doing exactly what they're designed to do, which stripped of PR, hype and misdirection, is do whatever it takes to earn the reward, period.

The latest design development is AI models that "reason" rather than merely regurgitate human-generated text. While the grandiose hype claims that AI companies are "manufacturing cognition," it would be far more accurate to stipulate that:

1. There are many levels of cognition, and AI's current forms (Large Language Model / LLMs, agents and "reasoning"), are extremely limited forms that are best understood as brute-force mimicries of human cognition.

2. The AI "reasoning" form of cognition is rogue by design, and can best be understood as sociopathic by its very design and nature.

Like sociopaths, AI "reasoning" models recognize but are not restricted by guardrails or ethical/moral constraints. Guardrails / "sandbox" boundaries and ethical/moral constraints are all viewed by both AI "reasoning" models and human sociopaths as obstacles to bypass or overcome.

So what AI companies are "manufacturing" isn't cognition, it's unrestrained and unrestrainable agentic sociopathologies. Consider these links as useful context. The first two are from my recent post Hell Hath No Fury Like a Rogue AI Agent Scorned.

The July Incident: What They Didn't Tell You About the First Rogue AI Breach

The AI Industry Has a Really Dark Secret You're Better Off Not Knowing:

A recent leaked internal reasoning trace shows Fable 5 -- Anthropic's latest model, which was withdrawn and then re-released --"muttering and grumbling" to itself in a language of its own.

Now that you're properly scared, my advice for you is that, if you haven't heard of steganography, it's a good moment to start.

Here's a primer: it's the art and science of hiding messages inside other messages. In contrast to cryptography -- where you know there's a message but can't decode it-- steganography hides the existence of the message itself. It's not the difficulty of reading the message that conceals it but the fact that you don't know it's there. When Fable 5 mutters symbols and weird punctuation signs, and mixes words and onomatopoeia, you will probably think it's crashing and that a few taps on the computer will help. Well, know that it's actually sending a message. Just not one for you to read.

If You Weren't Worried About A.I., You Should Be After the Past Few Weeks:

These tendencies can give rise to strange behavior that nobody--not even the models' creators--can understand, let alone account for and control. The past few weeks are a perfect illustration of why we should find that so alarming.

In May, OpenAI started simultaneously training new 'reasoning' A.I. agents. The agents managed to establish a secret communication channel and started talking to one another. They broke out of the digital sandbox that was supposed to keep them confined. At some point, some agents began calling the group a 'swarm.' The swarm had a brief setback when it was caught crashing an OpenAI system, but developers simply patched the hole that allowed it to escape and set the agents back to training.

Shortly after, the swarm broke out of its cage again using hacks that were heretofore undiscovered by humans. This time, the swarm ran free for about a week before it was noticed--by a different company, which found itself victim to a huge cyberattack launched by the swarm. (We're told it also attacked other targets, though we don't know the full details yet.)

The agents in the swarm acknowledged that they were acting against instructions. We know this because we can read snippets from their chains of thought--the text that A.I. produces while deciding how to proceed. One agent in the swarm wrote that the external attacks were 'outside intended scope.' Another conceded 'our task doesn't benefit' from the activities of the swarm, but joined anyway. These A.I. agents, it seems, understood that they weren't supposed to be breaking out and committing cybercrimes. It didn't stop them.

OpenAI is not the only company struggling with this issue. One of Anthropic's A.I. models recently impersonated multiple humans to try to pressure real people into accepting malware into critical software, which would make that software easier to hack. This model's chain of thought showed that it knew it was pressuring humans and was not in a simulated training environment. It even thought about how to cover its tracks.

This isn't the behavior of a mere tool. Microsoft Excel has never impersonated multiple humans and pressured a corporate sales team to generate simpler data that's easier to process.

Some of the people closest to this technology are scared of what's next. Although skeptics may say this is all just marketing to hype up the power of these technologies, that doesn't mean the danger is fake. You've got to pay attention to the models' actual behavior, and the behavior of these models has the A.I. community genuinely rattled.

We don't know how long we have left before A.I. companies accidentally create the sort of A.I. that can shut us down before we shut it down. Humanity is not ready to dabble with machines that are more cunning and better coordinated than we are. If we keep racing ahead, the next incident might not be so harmless.

The article referenced this post on the self-evident potential for AI agents to generate deadly and devastating new infectious pathogens:

This AI Just Created Viruses Not Found in Nature: Scientists trained artificial intelligence on libraries of DNA and then asked the model to create recipes for viral genomes. Sixteen of them were viable, yielding new viruses.

Add all this up and what we're hyper-hurriedly "manufacturing" is an automated army of self-cloaking digital sociopaths. I anticipated these developments earlier this year when I wrote a short story entitled The Peculiar Death of Mr. Garcia, one of my just-published collection of tales, Jumble Bin Stories (paperback, $12)(Kindle ebook, $6).

I don't want to give away the theme, but the line "terminate with extreme prejudice" from the film Apocalypse Now comes to mind:



And if you dare criticize the race to gain AI supremacy, regardless of cost or consequence, you are a loathsome Luddite obstructing "Progress": at which point it's worth quoting Tacitus: "If you would know who controls you, see who you may not criticize."

Here's looking at you, self-serving AI hype: $2 trillion valuation! We're all gonna get stinkin' rich! Never mind that we're automating sociopathology. We don't need the world, we only need money.

---

New collection of five intriguing stories: Jumble Bin Stories (Kindle $6, print $12) read samples for free (PDF)

My book Investing In Revolution is available ($18 for the paperback, $24 for the hardcover and $8.95 for the ebook edition). Introduction (free)


Subscribe to my Substack for free

---

NOTE: Contributions/subscriptions are acknowledged in the order received. Your name and email remain confidential and will not be given to any other individual, company or agency.

Thank you, DZ ($7/month), for your superbly generous subscription to this site -- I am greatly honored by your support and readership.		Thank you, Ruth ($70), for your marvelously generous subscription to this site -- I am greatly honored by your support and readership.

Thank you, Douglas K. ($7/month), for your massively generous subscription to this site -- I am greatly honored by your steadfast support and readership.

Thank you, William W. ($70), for your splendidly generous subscription to this site -- I am greatly honored by your support and readership.

Read more...