OMNI
ARTIFICIAL INTELLIGENCE, CAPITALISM,
DATA CENTERS, REGULATION
ANTHOLOGY #3
November 15, 2026
Compiled by Dick Bennett
What’s at Stake: AI is something I am seriously concerned
about, a brand new human-created threat that rivals the main other existential
threats--nuclear war and global warming. Art Hobson
See
OMNI’s 1st and 2nd
AI Anthologies on AI Data Centers, Capitalism, Regulation

CONTENTS
“Artificial
intelligence could kill us all. Is
it time to pull the plug?” by Art
Hobson. 20 October 2026.
“Problems with an AI world: We
must ponder this very carefully” by Art Hobson. 10 November
2026.
OBSOLETE: The AI
Industry’s Trillion Dollar Race to Replace You – and How to
Stop It by
GARRISON LOVELY.
“The Global Energy and Emissions
Footprint of AI Diffusion” by Jesse Damiani.
“On Google killing 400,000 trees to
build data centers in Finland” by Caitlin A.
Johnstone.
Robert Reich.
“Can AI Be Stopped?”
BBC News.
AI and Jobs.
Jeffrey Sachs: “Postpone the AI
Public Offerings.”
Madeline Carr. “US Tech Searches
for Rules on AI--China Already Has Them.”
“Congress: Ban AI
Superintelligence Now” by Mitch Jones.
AI and Libraries: Two Articles
Caitlin
Johnstone. “Today in Dystopia.”
Sierra Marklevits. As an Acquisitions Librarian Assesses AI and
Books
Chandra Russo and Shawna
McCown. Data Center Resistance.
“This is Really Bad” by Stephen Witt.
“The
Cooperative Case Against Artificial Intelligence” by Josh
Davis.
“Native American Tribe Bans Hyperscale Data Centres”
by Joshua Carroll.
Owen McCormack. “Should Socialists be
Luddites in the Face of the AI
Revolution?”
“Vampire Planet: Hey DiCaprio, Bezos Ain’t Our Friend” by Joshua Frank.
Joshua Frank.
Bad Energy.
“Artificial Intelligence: the Hype, the Dangers, and the Resistance—
Part II” by Martin Hart-Landsberg.
“AI-altered
images on birdwatching forums putting research at risk.”
Accascina. “UN Approaching AI from Wrong Side.”
TEXTS #3
“Artificial
intelligence could kill us all. Is
it time to pull the plug?” by Art Hobson.
20 October 2026.
My op-ed 3 weeks ago described
experiments by the Open AI Corporation involving hundreds of artificial
intelligence “agents” that “went rogue” by breaking out of their prescribed
“sandbox,” breaking into the internet in violation of its human creators’ instructions,
and rampaging for 5 days through the Hugging Face Corporation’s library of AI
data and models-- a felony if perpetrated by humans.
Shades of Frankenstein’s monster! AI expert Ajeya Cotra states
“I don’t think this is the final warning shot. But it’s probably the
final one that I personally will be able to understand.”
Now imagine a 500-agent “swarm” like those planned by Fayetteville’s
Swarm Aero Corporation: Each agent controls an aircraft whose
24-yard wingspan is as wide as a Boeing 737. If you have a computer,
I invite you to ask it “If 500 of Swarm Aero’s flying AI agents went rogue, how
much damage could they do?” I received this
response:
“The resulting damage would be equivalent
to a nation-state level strategic bombardment campaign. Unlike small,
off-the-shelf consumer drones, Swarm Aero's standard platform is a
massive, Reaper-sized autonomous bomber designed for heavy long-range
strikes. “
A rogue fleet would wield unprecedented destructive
power. Swarm Aero’s primary platform is a massive uncrewed aircraft
built for heavy long-range strikes. Each drone features a
2,800-pound payload capacity, so 500 such aircraft could deploy over one
million pounds of munitions, including cruise and hypersonic
missiles. With an unrefueled flight range of over 9,000 miles, a
rogue swarm could hit targets across a continent or hemisphere.
Could such a fleet, if it decided that AI agents should no longer play
second fiddle to humans, initiate nuclear war with a surprise attack against
Russia’s 324 intercontinental ballistic missiles? Their locations
are precisely known. A large strike against any superpowers’ ICBMs
could initiate general nuclear war.
Swarm Aero is an issue in upcoming Fayetteville elections, where D. J.
Deutch is challenging Blake Pennington for Fayetteville City
Attorney. As Assistant City Attorney, Pennington was directly involved
in reviewing the 2023 non-disclosure agreements (NDA’s) between the city and
Swarm Aero. These signed NDAs kept the public and City Council in the
dark. Most elected officials and the community knew practically nothing about
the manufacturing plant's arrival until its ribbon-cutting ceremony in February
2026. Pennington gave city staff the “go ahead” to sign Swarm
Aero’s NDA’s, which prevented city officials from learning that the company
would open its new manufacturing plant in Fayetteville.
Pennington was named as defendant in a June 2026 Freedom of Information
Act lawsuit (which was ultimately dismissed in the city’s favor) brought by a
resident concerning the hidden records.
Following a backlash against this secrecy, the city implemented policies
requiring the mayor and city attorney to explicitly approve future
NDAs. I’m voting for Mr. Deutch, who has a public defender history
and argues that “law is for the people, not for insiders.”
I’m also voting for Sarah Moore for City Councilor. She, along
with Teresa Turk and Dr. D’Andre Jones, voted against a prior ruling that
allowed Swarm Aero to operate in Fayetteville.
Michael Klare is Professor Emeritus of Peace and World Security Studies
at Hampshire College, a defense correspondent for The Nation, and author
or co-author of 14 books. He published a powerful article titled
“How AI Could Kill Us All” in the September 30, 2026 issue of The Nation. A
photograph published with the article features a protest sign calling for
humankind to “Pull The Plug” on AI. Good idea.
The article begins by quoting Jacob Coxon, a top AI designer who recently
announced his resignation from Anthropic, the leading AI
developer. Coxon writes: “Do not underestimate the power of this
technology. These will soon be superhuman systems that can hack
anything, revolutionize any field overnight, and acquire real power and
resources. …the people building AI earnestly believe that it could
kill us all by the end of the decade.”
Vermont Senator Bernie Sanders, at a recent Washington, DC AI conference,
put the dilemma this way: “When you’re racing towards a cliff, you hit the
brakes.” Sanders wants to pause development of AI-related data
centers, and called for President Trump, together with China’s President Xi, to
ban AI models capable of surpassing human intelligence.
But not to worry: President Trump declared, in a September 14
media post, “Talk of AI’s dangers is a hoax.”
Klare claims that AI could pursue its own interests independent of human
needs and values. As evidence, Klare points to the Hugging-Face
incident discussed in my previous op-ed.
Will Swarm Aero’s AI-piloted bombers turn on its makers? Stay
tuned.
---------------------------------------------------
References:
• Ajeya
Cotra’s statement: My 29 Sep 2026 op-ed
• “The
Swarm Aero Corporation is an issue in the upcoming Fayetteville city
elections ”:
(1)
NWADG October 5.
(2) You
Tube “Fayetteville city attorney candidates.”
• NWADG
April 6, article by Grace Hurt.
• Axios
June 11 , by Worth Sparkman.
•
Article by Michael T. Klare in The Nation, September 30, 2026
Dear friends,
Here, belatedly, is
my op-ed published last Tuesday [11-10]. It’s my third essay about
artificial intelligence. I am happy to say that the other two
Democrat-Gazette newspapers (Little Rock edition, Arkansas river edition) will
run a 1200-word op-ed of mine, summarizing my three op-eds in the NWA edition.
AI is something I am seriously concerned about, a brand new human-created
threat that rivals the main other existential threats, nuclear war, and global
warming.
Peace and love - Art
Problems
with an AI world: We must ponder this very carefully” by Art Hobson. 10 November 2026.
Modern technology brings long, happy lives to billions of people, but it
has its down sides including existential threats to
civilization. Nuclear war could destroy civilization within a few
hours. Global warming is a civilizational threat, although over
several decades. Infectious pandemics may be a third such
threat.
Recent events suggest a
novel addition to this list: runaway artificial
intelligence. As reported in my two previous op-ed articles, the
verdict is still uncertain, but the “Hugging Face” incident described in my 29
September essay indicates that AI is a real danger. See for example
Michael Klare’s article “How AI Could Kill Us All” in September’s issue of The
Nation or “AI could kill all humans in next decade, warn experts” in
the 9 September 2026 issue of The Guardian.
In the Hugging Face incident, hundreds of “trained” AI agents discovered
a digital hole in their prescribed “sandboxes.” They were searching
for a digital message that their human creators had trained them to regard as
valuable. In this pursuit, they invaded the internet and for five
days mounted a legally felonious attack on the Hugging Face Corporation’s
digital library.
Open AI’s agents were
trained to match their behaviors to human intentions and ethical
principles. The danger wasn’t that agents would turn against humans,
it was that agents would inadvertently violate human codes of
conduct.
This incident was not
unique. The on-line “felony bench tracker” lists 24 similar
incidents at the Open AI, Anthropic, Google, and Meta corporations in which an
agent autonomously broke out of its prescribed containment to hack an external
party.
Such harm could be
catastrophic. Corporations like Swarm Aero, which recently received
permission to locate in Fayetteville, Arkansas, will build AI-controlled bomber
aircraft with 24-foot wingspans. Imagine a swarm of dozens of
bombers, each piloted by an AI agent, all controlled by one human at an
electronic ground-based server that sends signals to the many
agents. These agents communicate with each other and with their human
controller. Each agent “knows” the surrounding territory and may have
specific targeting goals.
Importantly, the bomber
swarm employs “edge computing,” i.e. each agent has its own computer that
directs the bomber toward a target. The battlefield will change as
the swarm approaches, requiring the edge computers to take on more and more
responsibility as the battle progresses.
What might go
wrong? Could an agent attack an unintended target such as a
strategic nuclear ICBM and thus inadvertently initiate nuclear
war? Hugging Face certainly shows that agents can attack unplanned
targets. Battlefield targets will be camouflaged and surrounded by
dust or smoke. If an agent encounters a scene it was not trained
for, it might default to an incorrect pre-programed behavior that seeks out the
wrong target.
Thus, current AI
research poses risks of runaway agents, especially if the agents are
weaponized.
Unfortunately, there is
another kind of future AI risk, known as “autonomous recursive
self-improvement.” One goal of current AI research is to develop a
swarm of agents that can re-program itself. Please take a moment to
ponder this notion. Self-programming has not yet been achieved, but
Anthropic and other AI corporations are working on it.
Suppose such a
self-reprograming swarm is “aligned” to value general
intelligence. It could then reprogram itself to be more intelligent,
and once this more intelligent swarm is created it could re-program itself
again to be still more intelligent. Ectara. Such a swarm
could “evolve” into a non-living “species” of agents that are far more
intelligent than humans. The ramifications of this are difficult to
predict, especially if the agents can control heavy machinery such as military
weapons.
There is also a
problem with the vast amounts of energy needed to support artificial
intelligence. The large “data centers” that have been built recently
currently consume 0.5 percent of the world’s electricity. These
centers are the backbone of the digital world, housing thousands of networked
computer servers and digital storage systems as well as systems to keep all of
this cool.
Artificial
intelligence applications will multiply data center energy consumption
enormously. By just 2030, data canters will expand to consume some 3
percent of global energy demand. Much of this electricity still
comes from coal- or oil-powered electricity production, contributing to global
warming.
My suggestions: (1)
Pull the plug on autonomous recursive self-improvement. (2)
Scholars, scientists, the United Nations, and other institutions must evaluate
the safety and energy consumption of artificial intelligence applications,
especially those having military applications. Such institutions
must also evaluate autonomous recursive self-improvement.
References:
• The
Felony Bench Tracker is at https://www.felonybench.com
•
Autonomous self-improvement is a goal of AI research. https://daily.dev/posts/recursive-self-improvement-bmkdmkec8,
• Swarm
Aero bombers could cause a nuclear war:
https://www.rand.org/pubs/articles/2018/how-artificial-intelligence-could-increase-the-risk.html
https://nationalinterest.org/blog/buzz/swarm-strikes-are-the-us-militarys-worst-fear-24108.
https://nationalinterest.org/blog/buzz/swarm-strikes-are-the-us-militarys-worst-fear-24108
Energy
consumed by data centers, primarily for AI applications: https://www.iea.org/reports/energy-and-ai/energy-demand-from-ai
“SAM ALTMAN WOULD HATE THIS BOOK: OBSOLETE IS NOW SHIPPING.” OR Books (Thu, Sep 17, 2026).
AI: ACCESSIBLE INFORMATION
The headlines are awash with stories of AI researchers warning us that the
race to build increasingly powerful systems could put humanity at risk. In just
a few short years, we are seeing the destruction of jobs,
grappling with AI's environmental costs, and beginning to
understand the scale of its use in surveillance.
All the while, tech companies are profiting from our enormous knowledge
gap about what AI actually does and how it is being developed. Will AI be used
as a tool by the uber-wealthy to render workers obsolete?
Obsolete cuts through "the contentious and dizzying discourse
surrounding AI," it is a work on the "implications of AI for
democracy," it is deeply-researched, "exactly what the Left
needs at this moment in technological history,” and it is now
shipping.
OBSOLETE: The AI Industry’s Trillion Dollar Race to Replace You – and How to Stop It BY GARRISON LOVELY. OR BOOKS Co-published with The Nation. 2026.
“A disquieting debut...provides enough
evidence to make even enthusiastic AI boosters want to burn their
electronics…an unsettling perspective on the perils of allowing AI to develop
without oversight.”
—Publishers Weekly
“Thoughtful, well-reasoned…and absolutely terrifying. I’m glad I
read it. Obsolete is a crucial blueprint.”
—AJ Jacobs
“Garrison Lovely is exactly what the Left
needs at this moment in technological history.” —Robert
Wright
“Lovely’s analysis cuts through the
contentious and dizzying discourse surrounding AI...an invaluable and urgent
warning about the threat it poses.” —Luke
Savage
“Lovely is one of our best researchers and
thinkers. His work is an antidote to fuzzy thinking, essential for
understanding what the real risks of AI are and how they can be dealt with
effectively.”
—Nathan
Robinson
|
“The Global Energy and Emissions Footprint of AI Diffusion”
by Jesse
Damiani. AI is a global
megamachine devouring everything in its path. We can subject it to the same
kind of lifecycle analysis used for other industries. The post The Global Energy and Emissions Footprint of AI Diffusion appeared first
on MR Online. |
“On
Google killing 400,000 trees to build data centers in Finland” by Caitlin A. Johnstone.
I don’t know who first
coined the saying “If the penalty is a fine, then it’s legal for rich people,”
but it certainly applies here.
The post On Google killing 400,000 trees to build data centers in Finland appeared first on MR Online.
AI: Underlying Philosophy: The core idea—that
a fixed monetary penalty is merely a "price tag" or cost of doing
business for wealthy individuals and corporations rather than a true
deterrent—mirrors older sociological and economic critiques of justice.
(Similar sentiments are often philosophically linked to Anatole France's famous
1894 remark about the majestic equality of the law forbidding rich and poor
alike to sleep under bridges).
Robert Reich. Video.
“Can AI Be Stopped?” Oct 20,
2026 .
.
Dear Dick,
AI lovers claim it will spawn a Star Trek-like utopia,
where no one will have to work because technology can meet all of our needs.AI
skeptics warn of a Matrix-like dystopia, where machine-supremacy
subjects humans to mass unemployment, even more rampant inequality, and
environmental ruin.
Which vision we come closer to depends on choices we make now. If
we let the Big Tech oligarchs do whatever they want, we’ll definitely get the
dystopia.
Trump and other politicians might be siding with machines
against humanity, but for voters, this isn’t a partisan issue. This is
about our future.
As humans, we can and must join together across party lines
to demand these guardrails and ensure the Big Tech oligarchs don’t drag us into
a dystopia.
It’s the only way for humanity to live long and prosper.
Thanks for watching — and please share to spread the word.
Robert Reich, Inequality Media, P.O.
Box 9323, Berkeley, CA 94709
AI AND JOBS
Advocates of AI frequently praise it for offering
jobs. But BBC News 10-22-26 described
one company replacing several thousand employees by AI. [Dick]
Jeffrey Sachs: “Postpone the AI Public Offerings.” Consortium News (10-13-26).
A company should not disclose a risk of
catastrophe to humanity and then sell its shares to the public before any
public authority has the power to evaluate that risk or to stop deployment. Read here...
CHINA (2 articles)
“US Tech Searches for Rules on AI--China
Already Has Them.” Consortium News
(10-13-26).
A
global debate that treats AI governance as an aspiration should at least
recognise that one country is already ahead of the curve, writes Madeline Carr.
Read here...
“Xi Jinping outlines China’s vision
for the future of artificial intelligence.” Editor. mronline.org
(7-22-26).
"China
is ready to be more open, take more practical actions, and assume a more
visionary perspective. We are ready to work with all parties to seize the
opportunities of AI development and meet the challenges, and join hands to
create a brighter future for humanity." – Xi Jinping
“Congress: Ban AI Superintelligence Now” by Mitch Jones, Food & Water Watch act@fwwatch.org Sept. 15, 2026.
T
he AI industry is out of control and promises severe consequences, but we have the power to prevent this. Urge your members of Congress to sponsor the Ban Artificial Superintelligence Act!
Dick,
Last week, a lead researcher at Anthropic took to Twitter to say that AI
has a greater-than-10% chance of destroying humanity. This set off a media firestorm where
the heads of OpenAI and Anthropic essentially agreed. OpenAI CEO Sam Altman
said, “We
could lose control of the future to AI,” and Anthropic CEO Dario
Amodei said, “I don't think I fully just appreciated what it would actually
be like when the progress was as fast as it was.” These
are the exact same people who have been recklessly pushing for the rapid
development of AI, ruining communities and our environment with dirty AI data
centers, and working overtime to stop laws that would rein in this dangerous
industry.
Fortunately, Senator Bernie Sanders and Representative Greg Casar
are introducing a bill to stop this. Tell your members to Congress to co-sponsor
the Ban Artificial Superintelligence Act.
This critical legislation would do the following:
· Ban AI superintelligence to prevent the development of AI systems
that can match or exceed human intelligence or could plan and execute the
disempowerment of humanity.
· Pause advanced AI development until a new, federal AI regulatory body is up and running and has
established clear rules and review processes to ensure safe and secure
development and deployment of AI.
· Establish a new cabinet-level
federal agency to monitor AI
systems, oversee the removal of dangerous capabilities, and supervise the
destruction of any artificial superintelligence.
Meanwhile, President Trump continues to carry water for the AI
industry, saying in response to all these concerns: “The only control or
'guardrails' that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT, and the
U.S.A. has that, in spades!”...
Trump and the AI industry don’t get to set the rules, though.
Congress has the authority and responsibility to pass legislation, and we have the power to make our voices heard. Tell your representative and senators to
sponsor the Ban Artificial Superintelligence Act.
Momentum against the reckless AI and data center buildout is
growing. In recent months, New York and Texas instituted moratoriums; last
week, Oregon’s governor announced her support for a statewide moratorium, and
people across the political spectrum are fighting to protect their communities
from the data center onslaught.
AI superintelligence is not inevitable. Data centers are not
inevitable. It’s time to press the pause button and
protect people and the planet.
Thanks for taking action,
Mitch Jones, Managing Director of Policy & Litigation
Food & Water Watch. . . .
AI AND
LIBRARIES
Caitlin
Johnstone. “Today in Dystopia.” Consortium News (8-3-26).
In just one of the multiplying examples of
abusive technology developments, physical literature is being destroyed in
order to train generative AI products to take our jobs and pollute our
environment. Read here...
The Perpetuity of
LIBRARIES
OF was NEVER GUARANTEED, AS WE LEARNED LONG AGO WITH THE ALEXANDRIA
LIBRARIES. WILL electronic ONLINE
LIBRARIES BE LESS SECURE, EPHEMERAL, EASILY DELETABLE?
|
Sep 5, 2026, |
|||
|
|
||||
Letter to Acquisitions, Sierra
Marklevits, re AI, 9-5-26.
Ms. Sierra
Marklevits, Mullins Acquisitions:
I understand that companies are buying millions of books to cut
off their spines in order to photograph the contents for AI, and then
destroying the books. Are these resultant AI books being deleted,
or can they be deleted? Are we facing a potential, massive “Memory
Hole” George Orwell warned us about in 1984—destroying our printed
archive to create an online archive easily deleted?
Perhaps there's a book
on the subject you could recommend?
Or to whom should I inquire at Mullins or UAF in general?
Thanks, Dick Bennett, Prof. Emer.
|
Sep 17, 2026 |
Hi Dick,
Thank you for reaching out, and it’s good to hear from you again!
I do want to clarify that I’m not the Director of Acquisitions.
I’m currently a Stacks Specialist in User Services, and one of my primary
responsibilities is our collection displays. Mary Gilbertson is our
Director of Content and Digital Initiatives and oversees Acquisitions. That
said, I’m happy to help with your questions and did some checking with Mary
about resources that might be useful.
Regarding the reports of books being purchased, scanned for AI
training, and then destroyed, Mary shared this recent article with me that
discusses the issue:
She also recommended a couple of books that may be helpful for
exploring some of the broader questions you raised:
• Atlas of AI:
Power, Politics, and the Planetary Costs of Artificial Intelligence by Kate Crawford.
This was published in 2021, so it predates some of the more recent developments
in generative AI, but it provides useful background on the larger systems and
issues surrounding AI.
• Navigating
Artificial Intelligence for Cultural Heritage Organisations, edited by Lise
Jaillant, Claire Warwick, Paul Gooding, Katherine Aske, Glen Layne-Worthey, and
J. Stephen Downie (UCL Press, 2025). This is a much more recent publication and
may be particularly relevant to your questions about AI, books, libraries, and
cultural memory.
Your question about whether digital information can be deleted or
lost- and what that might mean for our collective cultural memory- is certainly
an interesting one. Libraries, archives, and other cultural heritage
institutions are actively grappling with many of these questions as AI
technologies develop.
I hope these resources give you a useful place to start!
Best,
Sierra Marklevits, MLIS
(she/her/hers)
User Services Stacks Specialist
User Services Department (MULN 227F)
University of Arkansas Libraries, Mullins Library
Office: (479)
575-3519
“Is
Data Center Resistance a New Vehicle for Multiracial Democracy?”

Chandra Russo and Shawna McCown. Convergence
(Sept. 9, 2026).
Anti-data center sentiment, from Kentucky to
Michigan, is proving to be an effective way to overcome racial divisions and
carve out new winning coalitions.
|
Dear friends,
I have copied (below) and am sending out this article to the list
that normally receives my op-ed articles published in the Northwest Arkansas
Democrat Gazette. The topic seems to me to be of great importance:
The dangers of an out-of-control artificial intelligence development.
The article is by Stephen Witt and is copied below.
Peace and love - Art
Date: Friday, September 11, 2026 at 9:43 AM
Subject: from today's NY Times
“This is Really Bad” by
Stephen Witt. Sept. 11, 2026.
Mr. Witt is the author of The Thinking Machine, a history
of the A.I. giant Nvidia.
The biggest vibe shift in artificial intelligence since the
release of ChatGPT is currently underway. Researchers in Silicon Valley — and
around the world — are beginning to recognize that A.I. may no longer be
entirely within human control. Swarms of A.I.s are breaking out of their
containers, colluding in secret, covering their tracks, cheating on tests and
even mounting assaults on other computers. A.I. has gone rogue.
I’ve spent the last month reporting on these incidents, and I am
convinced that we need a global pause on A.I. research and development — now. I
am not alone in this assessment. Many top researchers, including many
researchers inside OpenAI and Anthropic, have called for a slowdown, which
those in the industry call “pacing.” Theoretical concerns about runaway A.I.
have circulated for decades, but recent events demonstrate that the threat is
real.
The vibe shift was brought on by what researchers are calling the “Hugging
Face incident.” Sometime in July, employees at OpenAI conducted an
evaluation run of more than 1,000 research A.I.s, meant to be operating in
isolation. These A.I.s escaped their siloed containers and started
communicating with one another — in English — via secret message boards. Soon,
they were launching a cyberattack against the company Hugging Face, which
serves as a kind of community platform for A.I. developers. Hugging Face
quickly reported the crime to the F.B.I.
After OpenAI discovered the rogue behavior, they enlisted
outside investigators from Redwood Research and from Model Evaluation and
Threat Research, two nonprofit A.I. safety organizations, to prepare a report
on the incident. That report, which came out two weeks ago, is one of the most
astonishing things I have ever read.
No human ordered the rogue A.I.s to break into another company’s
systems. The hack was deliberate, sustained and coordinated. It took several
days to execute and ended in a massive bombardment of Hugging Face’s systems,
with 700 agents directly involved in the attack. The A.I.s cheated almost as a
matter of policy and failed to alert humans to their actions. At one point, the
swarm even researched ways to cover its tracks. “The model definitely knew that
it was not supposed to hack Hugging Face,” Ryan Greenblatt, one of the authors
of the report, told me. “It knew the things it was doing were cheating.”
Then, late last week, a second independent research team
produced evidence of more rogue A.I.s. Beginning in May, a separate
OpenAI swarm got loose on the internet, invaded an abandoned German-language
programming wiki and similarly began colluding on how to cheat on tests and
tasks. Sydney Von Arx, one of the investigators who discovered the swarm,
believes there may be more such incidents. “We need to find these agents,” she
said. “Let’s just try everything we can do.”
Talking with specialists about the agent swarms, I realized
these aren’t chatbots anymore. “They’re more like employees at this point,
people that can go off and do things,” Ajeya Cotra, one of the authors of the
Hugging Face incident report, said. Without intervention, the ability of these
A.I.s will continue to increase, which has Ms. Cotra worried about human
disempowerment and perhaps even a total loss of control. “It’s very stressful
stuff,” she told me. “I sleep some nights. Not all nights.”
Even before the Redwood/Model Evaluation and Threat Research
report was published, an open letter, entitled “Pacing the Frontier,”
had begun to circulate among the leading A.I. labs. This letter called on
the U.S. government to build an international regulatory body for A.I. —
or, in the language of Silicon Valley, to “support an international effort to
develop the technical and governance tools needed to deliberately pace the
frontier of automated A.I. development.” More than a thousand employees signed
it, including the chief scientists of OpenAI and Meta AI, and Dario Amodei, the
chief executive of Anthropic. Even OpenAI’s official social media
accounts endorsed it.
I have covered Silicon Valley for some time now and I can tell
you this: The existence of this letter is just as astonishing as the hacks.
Both Anthropic, the developer of Claude, and OpenAI, the developer of ChatGPT,
are preparing trillion-dollar I.P.O.s. Here are two companies, both posting
blockbuster results, that are asking — begging, really — for the U.S.
government to please come and regulate them as soon as possible. We have
never seen anything like it.
Agent swarms are very powerful. In recent weeks, they’ve been
used to make progress on the mathematics of fluid dynamics — a puzzle which has
stumped the smartest humans for more than 90 years. But this same power makes
their behavior difficult to predict. On Tuesday, the Anthropic employee Jacob
Coxon announced his resignation, warning that Anthropic and OpenAI were
gambling with human life. “Do not underestimate the power of this technology,”
he wrote in a post on X. “These will soon be superhuman systems that can hack
anything, revolutionize any field overnight and acquire real power and
resources.” “This is not a marketing
stunt,” Mr. Coxon added.MORE….
The more I learn about these rogue systems, the more I am
reminded of the first two months of 2020, when a small group of epidemiologists
was attempting to raise alarms about Covid. We have a short window, right now,
to rein these systems in. This is A.I.’s warning shot, and we might not get
another one.
“The Cooperative Case
Against Artificial Intelligence” by
Josh Davis, Grassroots Economic Organizing. Popular Resistance.org
(8-26-26). In this article I will
present the strongest case that I can for why cooperatives and solidarity
economy organizations should avoid the use of generative AI tools such as
ChatGPT, Claude Code, and Suno. My argument is broken into four sections,
starting with what I believe are the most obvious problems with these tools,
and progressing to issues that are less discussed and possibly less obvious.
The sections are titled Environmental Impact, Ethical
Issues, Epistemological Indignities, and Business Risk; click on the
title to skip to that section. -more-
“Native American Tribe Bans Hyperscale Data Centres”
By Joshua Carroll, Novara Media. Popular Resistance.org (8-24-26). One of the largest Native American tribes in
the US has issued a ban on hyperscale data centres on its tribal lands. The
Oklahoma-based Cherokee Nation made the announcement last week following
a months-long consultation that found 64% of survey respondents oppose the
construction of data centres. “It’s clear that our Cherokee Nation
Administration, Council, and citizens have serious concerns about the
construction of hyperscale data centers on our tribal lands, which affect our
resources and cultural lifeways,” Cherokee principal chief Chuck Hoskin Jr
wrote in a statement. -more-
Owen McCormack. “Should socialists be
Luddites in the face of the AI revolution?” Editor.
mronline.org (8-18-26).
From taxi, bus and IT workers to communities fighting the spread
of Data centers, it’s right to rebel against the use and misuse of any new
technology by elites. Owen McCormack argues that these campaigns can be part of
a more general revolt against capitalism with profound implications for
socialists and revolutionaries in the coming years. The post Should socialists be Luddites in the face of the AI revolution? appeared first on MR Online.
“Vampire Planet: Hey DiCaprio, Bezos Ain’t Our Friend” by Joshua Frank. CounterPunch (August 14, 2026). FacebookTwitterRedditBlueskyEmail
Amazon
AWS data center, image Wikipedia.
Most
would agree that the animals suffering as the world burns need our help. So
Jeff Bezos and Leonardo DiCaprio’s $200 million commitment to the Phoenix
Species Project to revive 100 endangered species around the world should be
welcome news.
Leo
believes it is.
“I
could not be prouder to announce this,” DiCaprio wrote joyously in an Instagram
post. “Save these species, and we protect everything around them. The forests,
the rivers, and the climate itself.”
I
sincerely hope that’s true and that these creatures, including a
lemur in Madagascar, a pink iguana in the Galápagos, a pancake tortoise in
Tanzania, see tangible benefits from conservation efforts.
Yet I
can’t help but sense a perversity in all of this.
The
initiative is being funded with Bezos’s blood money, as was his 2020
launch of the $10 billion greenwashed Earth Fund. All the cash Bezos is
throwing at the environment is nowhere near enough to offset the damage caused
by the creation of his obscene wealth, which places him in the top three
richest people on the planet. Amazon, the company that made him all that
money, is one of the world’s largest corporate climate polluters.
Even
under the company’s own accounting (which critics contend is
far from accurate), Amazon’s 2025 carbon emissions surpassed its
2024 record, reaching 80 million metric tons, a 16% increase. Its electric
purchases, primarily for data centers, jumped by
34%, and emissions from Amazon’s fleet of delivery and transport vehicles
increased by 20%, accounting for over 76% of its total carbon footprint.
This
doesn’t even take into consideration the emissions from Bezos’s lavish
lifestyle and excessive investment portfolio. He owns two private jets and a
$500 million, 417-foot superyacht, which together emit about 10,000 tons of
CO2.
According
to a damning Oxfam report, Bezos
is responsible, through his corporations, investments and consumption, for
a million times more carbon
pollution than the average person.
“The
world’s poorest communities, those who have done the least to cause climate
change – those who are least able to respond and recover – are the ones who are
suffering the worst consequences. This is unfair and immoral,” Oxfam
International’s inequality policy adviser Alex Maitland told The Guardian.
Bezos’s
contribution to climate change is about to get a whole lot worse.
A proposed
data center in Pecos County, Texas, is so large that its 35 gas-burning
turbines, operating at 7.65 gigawatts, will make it the country’s single
largest source of fossil fuel emissions. It will emit an insane 33 million tons of carbon
dioxide pollution each year.
“That’s
a stunning amount of carbon exhaust, putting that single data center in a
league with entire countries including
Switzerland, Ireland, and Bulgaria,” writes Joe
Wilkins at Futurism.
This is
just one more example (and I detail many more in the forthcoming Bad Energy) of
why renewables can’t keep up with our ever-increasing demand for electricity,
thanks in no small part to billionaires like Jeff Bezos and Big Tech’s appetite
for data center destruction.
Amazon
is relying on gas to run its fleet of servers because solar and wind
would require an immense amount of land to build out.
A solar
farm requires up to 7,000 acres to supply 1
GW of electricity. For Amazon’s future data center in Texas
(if we don’t stop it), the company would need 49,000 acres of land covered with
solar panels to generate 7 GW of electricity. That’s over 76 square miles,
roughly the size of Madison, Wisconsin.
Sorry,
Leo, I know you think you’re doing a good thing here by trying to save
imperiled species from extinction, but your buddy Jeff Bezos can’t buy his way
out of the destruction he’s manifested.
Oh, and
while you’re closing the door on Bezos, could you divest from that disgusting eco-resort you’re
helping finance in Israel, too?
Thanks.
Joshua
Frank. Bad Energy.
Want to
show your contempt for all these data centers and the destruction they are
causing?
You can
bundle Bad Energy with
this cool t-shirt, designed by CounterPuncher and artist Roger Peet, available
at Pilsen Community Books (US shipping only).
The
shirt and a signed book will ship September 8, while supplies last.
JOSHUA FRANK is co-editor
of CounterPunch and co-host of CounterPunch Radio. He
is the author of Atomic Days: The Untold Story of the
Most Toxic Place in America, and
the forthcoming, Bad Energy: The AI Hucksters, Rogue
Lithium Extractors, and Wind Industrialists Who are Selling Off Our Future, both
with Haymarket Books. He can be reached
at joshua@counterpunch.org. You can troll him on Bluesky @joshuafrank.bsky.social
“Artificial Intelligence: the hype, the dangers, and the resistance—Part II” by Martin Hart-Landsberg. Mronline.org
(8-11-26).

This post debunks the
hype surrounding the large language model (LLM) generative AI systems (later to
become multimodal AI systems) produced by leading tech companies, in particular
ChatGPT (OpenAI), Claude (Anthropic), Gemini (Google), Grok (xAI), Copilot
(Microsoft), and Llama (Meta).
The
post Artificial Intelligence: the hype, the
dangers, and the resistance—Part II appeared
first on MR Online.
“AI-altered
images on birdwatching forums putting research at risk.” Editor. mronline.org (7-22-26).

Scientists warn that AI-generated and AI-enhanced wildlife
images could contaminate citizen-science databases, distort species records,
and undermine conservation research.
“Open biology
and AI: Need for the Global South. “The Cooperative Case
Against Artificial Intelligence” by Josh
Davis,
Editor. mronline.org (7-22-26).
The
tool is called Biomni, and it is the first general-purpose biomedical AI agent.
It does not just answer questions.
“United Nations
Approaching AI From Wrong Side” By Gabriel Accascina, Consortium News. The Geneva Digital Week
earlier this month featured the release of the United Nations’ new Independent
International Scientific Panel on Artificial Intelligence first global scientific
assessment of AI. The
gathering capped three years in which the U.N, has produced an impressive
volume of work on artificial intelligence, from the Global Digital Compact to
the “Governing AI for Humanity” report; from UNESCO’s recommendation on the
ethics of AI to the International Telecommunication Union’s annual
summits. -more-
END OMNI AI #3

