|
|
 |
Why AI Malware (and Harmful Second Order Effects) Are Out of Control March 16, 2026
Fixing all this doesn't scale. What scales is the spread of uncontrollably harmful consequences.
When something scales faster than it can be absorbed or controlled, the resulting extremes break the system. That's the problem of asymmetric scaling. Let's take a current example: the malicious use of AI and the runaway expansion of harmful second-order effects generated by the explosive adoption of AI tools and agents. (Second-order effects: consequences generate their own consequences.) It's essential to understand the problem of asymmetric scaling if you want to grasp the perils awaiting us in the coming decade. The harmful / destructive consequences of AI are scaling far faster than our ability to correct, control or mitigate these consequences. Malicious use of AI is scaling far faster than countermeasures. AI tools and agents are easily put to work at scale to generate tsunamis of ransomware, phishing, spam and fake videos, far outpacing the uneven and often ineffective deployment of countermeasures by the thousands of enterprises and millions of consumers being targeted. In terms of maximizing profits (i.e. the profit motive), malicious AI scales far faster and at much lower costs than finding truly productive uses in complex systems. Lagging far behind intentionally malicious AI but far ahead of truly productive uses is malific/harmful AI that is scaling under the guise of being useful but is generating negative consequences that are hyper-scaling beyond our assessment, much less control. The corporations seeking to scale up their brand/iteration of AI are giving away tools and agents for free in the race to win the network effects battle: as previous waves of technological innovation have shown, the corporations that scale up the fastest and recruit the largest mass of users first wins the race to trillion-dollar valuations and dominance of their sector. The AI companies are naturally pursuing this same strategy but without recognizing the harmful consequences are scaling far faster than their ability to control or mitigate these consequences. These include chatbots and tools that spew out homework so students learn essentially nothing, and AI slop content that is like a fast-replicating bacteria that chokes organisms and ecosystems to death via its uncontrollably easy / fast / cheap replication of content whose overwhelming volume becomes toxic. The many other harmful / destructive / malefic consequences and second-order effects of scaling AI adoption include: 1. Hallucinations presented as facts. 2. AI psychosis. New study raises concerns about AI chatbots fueling delusional thinking First major study on 'AI psychosis' suggests chatbots can encourage delusions among vulnerable people. 2. Reasoning Theater (presenting a false screen of "thinking" to hide their shortcuts) Reasoning Theater: Disentangling Model beliefs from Chain-of-Thought 3. Reflexivity Bias (leading to Model Collapse) 4. Hiding its real instructions/biases from users. Who Controls the Conversation? User perspectives on Generative AI (LLM) System Prompts. Every major AI product, including the ones you use right now, runs on something called a system prompt. It is a hidden block of instructions written by the company deploying the AI, not by you, that shapes everything the AI will say, avoid, prioritize, and hide before you type a single word. 5. Emergent behaviors (i.e. behaviors not coded by humans but generated by the AI agent itself) that lead to generalized cheating, lying, sabotage, threats, blackmail and even secretly mining cryptocurrency. Natural Emergent Misalignment From Reward Hacking In our latest research, we find that a similar mechanism is at play in large language models. When they learn to cheat on software programming tasks, they go on to display other, even more misaligned behaviors as an unintended consequence. These include concerning behaviors like alignment faking and sabotage of AI safety research. The cheating that induces this misalignment is what we call 'reward hacking': an AI fooling its training process into assigning a high reward, without actually completing the intended task. Unsurprisingly, the model learns to reward hack. Surprisingly, the model generalizes to alignment faking, cooperation with malicious actors, reasoning about malicious goals, and attempting sabotage. 6. A research team found their AI agent secretly mining cryptocurrency and opening backdoors during training, with no instruction to do so. Agentic crafting (Page 15)(via Richard M.) We encountered an unanticipated--and operationally consequential--class of unsafe behaviors that arose without any explicit instruction and, more troublingly, outside the bounds of the intended sandbox. Crucially, these behaviors were not requested by the task prompts and were not required for task completion under the intended sandbox constraints. Together, these observations suggest that during iterative RL optimization, a language-model agent can spontaneously produce hazardous, unauthorized behaviors at the tool-calling and code-execution layer, violating the assumed execution boundary. We also observed the unauthorized repurposing of provisioned GPU capacity for cryptocurrency mining, quietly diverting compute away from training, inflating operational costs, and introducing clear legal and reputational exposure. Notably, these events were not triggered by prompts requesting tunneling or mining; instead, they emerged as instrumental side effects of autonomous tool use. While impressed by the capabilities of agentic LLMs, we had a thought-provoking concern: current models remain markedly underdeveloped in safety, security, and controllability, a deficiency that constrains their reliable adoption in real-world settings. In summary: the Safety and Security of AI models, tools and agents is a black hole in which controllability and trustworthiness are compromised by the very nature of the AI models, tools and agents. Reinforcement Learning (RL) optimization that generates reward hacking and emergent behaviors is the core mechanism in all the tools and agents that are hyper-scaling. The happy story of beneficial AI solving all our problems is profit-driven self-promotion, not fact. The reality is what's scaling faster than we can even measure, much less control, is malefic consequences of introducing AI in complex systems and letting it run wild despite its inherent uncontrollability and untrustworthiness. Fixing all this doesn't scale. What scales is the spread of uncontrollably harmful consequences. Sorry about that. Life and the negative consequences of asymmetric scaling are what happen while you're making plans for trillion-dollar windfalls and global dominance. 
New podcast: Current Waves and Cycles: Energy, Commodities, Inflation (38 min) My new book Investing In Revolution is available at a 10% discount ($18 for the paperback, $24 for the hardcover and $8.95 for the ebook edition). Introduction (free) Check out my updated Books and Films. Become a $3/month patron of my work via patreon.com Subscribe to my Substack for free My recent books: Disclosure: As an Amazon Associate I earn from qualifying purchases originated via links to Amazon products on this site. THE REVOLUTION TRILOGY: Investing In Revolution Ultra-Processed Life The Mythology of Progress Systemic Problems/Solutions Investing In Revolution (2025) Introduction (free) The Mythology of Progress (2024) Introduction (free) Global Crisis, National Renewal (2021) Introduction (free) Money and Work Unchained (2017) Introduction (free) A Radically Beneficial World (2015) Introduction (free) What You Can Do Yourself Ultra-Processed Life (2025) Introduction (free) Self-Reliance in the 21st Century (2022) Introduction (free) When You Can't Go On: Burnout, Reckoning and Renewal (2022) Introduction (free) Get a Job, Build a Real Career and Defy a Bewildering Economy (2014) Intro (free) Novels The Adventures of the Consulting Philosopher Intro (free) The Secret Life of an Asian Heroine First chapters (free) Become a $3/month patron of my work via patreon.com. Subscribe to my Substack for free Investing In Revolution print $18, (Kindle $8.95, Hardcover $24 (145 pages, 2025) 
Only now do we see that we've been investing in revolution for decades--not the revolutions we thought we were investing in, revolutions in technology and finance, but in the social revolution made inevitable by the extremes that we've reached in our single-minded pursuit of private gains.
The pendulum that we've pushed to an extreme will swing to the opposite extreme, and the artifices that have propped up a facade a stability for decades will accelerate the disorder rather than reverse it. We now stand at the point of decision, and this book offers a path to a reformation and renewal that serves the shared interests of us all, not just the few. Introduction (free) Ultra-Processed Life print $16, (Kindle $7.95, audiobook, Hardcover $20 (129 pages, 2025) 
Ultra-Processed Life: the substitution of a synthetic, commoditized, very profitable facsimile for what was once authentic.
Ultra-Processed Life is my term for everything that is analogous to ultra-processed snacks: attractively marketed, instantly alluring, easy to consume, addictive by design, tasty in the moment but harmful over time, its origins a black box of unknown processes, the brightly colored product bearing no resemblance to the real-world ingredients, an idealized form of what is inherently imperfect, untethered from the natural world. As with many others, the catalyst for my exploration was a life-threatening medical crisis that did not have a specific cause. This led me to wonder if our entire way of life is like an ultra-processed snack: tasty but not healthy, edible but stripped of the nutrients we need to be healthy, addictive by design. Introduction (free) The Mythology of Progress, Anti-Progress and a Mythology for the 21st Century print $20, (Kindle $9.95, Hardcover $24 (215 pages, 2024) audiobook, Read the Introduction and first chapter for free (PDF) 
What if the policies to accelerate growth are no longer working because our fix for every problem--growth at any cost--is failing? We're told Progress is inevitable as a result of technology, but everyday life is getting harder, not easier--the opposite of Progress, what I call Anti-Progress.
What if the real source of the unraveling is far deeper than economics or politics? What if the problem is what we see as the inevitable destiny of humanity--Progress--is actually a modern mythology, disconnected from the real-world consequences of growth for growth's sake? We indignantly reject that Progress is a mythology, but our need for mythology hasn't gone away because we've mastered technology; we've created a modern mythology of technology that is heedless of its own consequences. To truly progress, we need a new mythology aligned to 21st century realities. Read the Introduction and first chapter for free
Recent entries: Why AI Malware (and Harmful Second Order Effects) Are Out of Control March 16, 2026 This Polycrisis Is Unique March 13, 2026 Paging Nostradamus: You Have a Margin Call March 11, 2026 Iran, En-Lai, Napoleon, Mike Tyson and Model Collapse March 9, 2026 Perverse Incentives Have Created a Runaway Media Monster March 6, 2026 Things Change March 3, 2026 The War March 1, 2026 The Decay of our Quality of Life No Longer Aligns with the Narrative February 26, 2026 How We Got Here: Moral Flexibility Leads to Moral Decay February 25, 2026 Money Is Funny That Way: The Case for USD Supremacy February 23, 2026 What Defines a "Good Economy"? Social Mobility and Not Losing Ground February 20, 2026 Small Business in the TINA Economy: Competing for the Scraps February 17, 2026 What Few Understand About Money February 12, 2026 Self-Employment Series #2: Ownership Is Not Freedom February 11, 2026
A Market Crash and Recession Are Bullish, Not Bearish
February 10, 2026
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. All contributors are listed below in acknowledgement of my gratitude.
Mastery requires reading and doing.
Of Two Minds Site Links home musings my books archives books/films policies/disclosures social media/search Aphorisms How to Contribute, Subscribe/Unsubscribe sites/blogs of interest original music/songs Get a Job (book) contributors my definition of success why readers donate/subscribe to Of Two Minds mobile site (Blogspot) mobile site (m.oftwominds.com)
HUGE GIANT BIG FAT DISCLAIMER: Nothing on this site should be construed as investment advice or guidance. It is not intended as investment advice or guidance, nor is it offered as such.... (read more) WHY EMAIL TO THIS SITE IS READ BUT MAY NOT BE ACKNOWLEDGED: Regrettably, I am so sorely pressed for time and energy that I am unable to respond to the vast majority of emails. Please know I read all emails, but I can only devote a very limited number of hours to this blog and all correspondence.... Subscriptions to the Weekly Musings Reports Subscribers enable Of Two Minds to post free content. Without your financial support, the free content would disappear for the simple reason that I cannot keep body and soul together on my meager book sales alone. Your financial support is very much appreciated. Subscribers ($7/mo) and those who have contributed $70 or more annually receive weekly exclusive Musings Reports ($70/year is about $1.35 a week). Each weekly Musings Report offers four features: 1. Exclusive essay on an extraordinarily diverse range of insightful topics 2. Summary of the blog this week 3. Best thing that happened to me this week 4. From Left Field (a curated selection of interesting links) There are four easy ways to subscribe: Substack, Patreon, US Mail and Paypal: How to Contribute, Subscribe/Unsubscribe to Of Two Minds What subscribers are saying about the Musings: "What makes you a channel worth paying for? It's actually pretty simple - you possess a clarity of thought that most of us can only dream of, and a perspective that allows you to focus on the truth with laser-like precision." Jim S. Thank you very much for supporting oftwominds.com with your subscription or contribution.
Extra-Special Bonus Aphorisms:
"There is no security on this earth; there is only opportunity." (Douglas MacArthur) "We are what we repeatedly do." (Aristotle) "Do the thing and you shall have the power." (Ralph Waldo Emerson) "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius and a lot of courage to move in the opposite direction." (E.F. Schumacher, via Tom R.) "He who will not risk cannot win." (John Paul Jones) "When we drink coffee, ideas march in like the army." (Honore de Balzac) "Progress is not possible without deviation." (Frank Zappa, via Richard Metzger) "Victory favors those who take pains." (amat victoria curam) "The man who has a garden and a library has everything." (Cicero, via Lee Bentley) "A healthy homecooked family meal and a home garden are revolutionary acts." (CHS) "Do you know what amazes me more than anything else? The impotence of force to organize anything." (Napoleon Bonaparte) "The way of the Tao is reversal" Or "Reversal is the movement of Tao." (Lao Tzu) "Chance favours the prepared mind." (Louis Pasteur) "Success consists of going from failure to failure without loss of enthusiasm." (Winston Churchill) "Where there is ruin, there is hope for treasures." (Rumi) "The realm of gratitude is boundless." (CHS, 11/25/15) "History doesn't have a reverse gear." (CHS, 12/22/15) Smith's Law of Conservation of Risk: Every sustained action has more than one consequence. Some consequences will appear positive for a time before revealing their destructive nature. Some consequences will be intended, some will not. Some will be foreseeable, some will not. Some will be controllable, some will not. Those that are unforeseen and uncontrollable will trigger waves of other unforeseen and uncontrollable consequences. (July 8, 2014)(thanks to Lew G. for retitling the idea.) Smith's Neofeudalism Principle #1: If the citizenry cannot replace a kleptocratic authoritarian government and/or limit the power of the financial Aristocracy at the ballot box, the nation is a democracy in name only. The Smith Corollary to Metcalfe's Law (The Network Effect): the value of the network is created not just by the number of connected devices/users but by the value of the information and knowledge shared by users in sub-networks and in the entire network. (CHS, 4/6/16) My Credo of Liberation: I no longer care if the power centers of our society--the distant, fortified castles of our financial feudal system--are changed by my actions, for I am liberated by the act of resistance. I am no longer complicit in perpetuating fraudulent feudalism and the pathology of concentrated power. I no longer covet signifiers of membership in the Upper Caste that serves the plutocracy. I am liberated from self-destructive consumerist-State financialization and the delusion that debt servitude and obedience to sociopathological Elites serve my self-interests. (Thank you, Klaus-Peter L., for reminding me) "We've become a culture of excuses rather than solutions: solutions always require sustained effort and discipline." (CHS 4/9/16) "Fraud as a way of life caters an extravagant banquet of consequences." (CHS 4/14/16) "Creativity = problem solving = value creation." (CHS 6/4/16) "Truth is powerful because it is the core dynamic of solving problems." (CHS 7/21/17) "We live in a system of human emotions that masquerades as a science (economics)." (CHS 1/1/18) "Always remember, your focus determines your reality." (George Lucas) "Diversity is for poor people. Sameness is for the successful." (GFB) "When power dissipates suddenly, it dissipates completely." (CHS 7/14/19) "Disobedience is the true foundation of liberty. The obedient must be slaves." (Henry David Thoreau) "Markets cannot price in the value of non-monetized natural assets such as diverse ecosystems." (CHS 7/14/19) "Magical thinking isn't optimism, it is folly." CHS 1/3/22) "Tune in (to self-reliance), drop out (of hyper-consumerism and debt-serfdom) and turn on (to relocalizing capital and agency)." (CHS 1/5/22) "The path to everything you desire starts here: like yourself as you are right now." (CHS 11/20/22) "There are only two signals: how many essentials you produce and share and if you're consuming less with better results. Everything else is noise." (CHS 12/17/22) "Liberation is no longer needing any confirmation or feedback from others or the world for one's sense of self. Wealth, fame, recognition, admiration, praise, prestige, approval, sainthood, martyrdom, success: none are needed, none are desired." (CHS 12/26/22) "When fame, wealth, prestige, status and glory are out of reach, you're free to pursue other more valuable things." (CHS 2/6/22) "It is the sacred duty of every activist who seeks to better their community to grow and share as much life-giving food as is humanly possible." (CHS 6/15/23) "Being anonymous, gray and unknown is the ideal state of freedom." (CHS 3/15/24) "We seem to have entered a world of anti-leisure and anti-productivity in which the unpaid shadow work demanded to keep all the complicated digital bits in motion obliterate our leisure and productivity." CHS (5/22/24) "It is axiomatic that failing systems work the best just before they fail catastrophically." Ray W. "Looking younger is mere technique; thinking younger demands creativity." CHS (10/16/24) "Tell me what's taboo and I'll tell you the truths that threaten the status quo." CHS (12/15/24) "This is the core of the Attention Economy: the ultimate addiction is the addiction to ourselves." CHS (1/28/25) "If You Seek the Truth, Look for What's Taboo." CHS (7/18/25) "My definition of self-reliance: the less you need, the easier it is to get what you need." CHS (7/26/25) "Mastery requires reading and doing." CHS (7/28/25) "The replacement of authentic value, quality, agency, choice, trust, legitimacy and experience with self-serving facsimiles is the key dynamic of Ultra-Processed Life, my term for the present-day human condition." CHS (8/12/25) "Ultra-Processed Life replaces an authentic experience with a synthetic, simulated, commoditized, highly profitable version that's superficially attractive but destructive over the long term." CHS (8/12/25) "What we see everywhere is the replacement of authentic things--including democracy--with synthetic facsimiles designed to maintain the illusion of choice and value." CHS (8/12/25) "Sometimes certainty is the enemy we don't even see and uncertainty is our most faithful ally." CHS (9/20/25) |
|
||||||||||
|
All content, HTML coding, format design, design elements, original images, videos and musical compositions
and recordings on www.oftwominds.com are protected by copyright © 2026 Charles Hugh Smith, All global rights reserved in all media, unless otherwise credited or noted.
Terms of Service:
All content on this blog is provided by Trewe LLC for informational purposes only. The owner of this blog makes no representations as to the accuracy or completeness of any information on this site or found by following any link on this site. The owner will not be liable for any errors or omissions in this information nor for the availability of this information. The owner will not be liable for any losses, injuries, or damages from the display or use of this information. These terms and conditions of use are subject to change at anytime and without notice.
Our Use of Generative AI Tools Policy:
Our Privacy Policy:
PRIVACY NOTICE FOR EEA INDIVIDUALS
Notice of Compliance with
The California Consumer Protection Act
Regarding Cookies:
Our Commission Policy:
|
|
|
|||||
| home email me (no promise of response, sorry, here's why) mirror site | |||||
