Musk might be right. My AI Capability Multiplier currently stands at 65.5: it will exceed 120 by year-end 2026, 3,100 by the end of 2027, and 45,000 by the end of 2028. This started at 1 on January 1, 2025, just before Trump was inaugurated. AI development is going parabolic even as we speak, sending the world into a state of Future Shock, as Alvin & Heidi Toffler predicted in 1970.
⁃ Patrick Wood, Editor.
Elon Musk has predicted that AI may outperform humans in every field by 2028. Musk made the comment while responding to computer scientist Francois Chollet, who recently reposted a prediction he made in May 2021. Chollet had said that almost every scientific field could effectively become a branch of computer science within 10 to 20 years. His prediction covered areas such as physics, chemistry, biology, medicine and archaeology.
Your prediction was conservative. AI will beat all fields by the end of next year or maybe 2028, at the latest.
Amodei has not called for an end to AI development. Instead, he has proposed slowing the pace enough to allow advanced AI systems to be properly evaluated.
In an essay published in September, Amodei outlined a three-step approach. The first step involves giving independent third-party evaluators broad access to advanced AI models. Anthropic said it would adopt this approach itself. The next steps would require cooperation between AI companies, governments and other countries.
Amodei also raised concerns about autonomous AI agents. His comments came after a recent incident involving AI agents and the open-source platform Hugging Face.
Musk had earlier supported Amodei’s concerns about the pace of AI development. Responding to Amodei’s essay, Musk posted publicly, “Dario is right.”
Musk’s latest prediction puts a much shorter timeline on AI surpassing human capabilities across scientific and other fields. However, it is a prediction rather than a confirmed timeline for how AI development will unfold.
AI is developing faster than we could have imagined. What does this mean for future development and how worried should we be? How far are we from AI developi...
At the 18th BRICS Summit in New Delhi, Chinese President Xi Jinping proposed that China take the lead in establishing a BRICS AI open-source community. His proposal envisaged cooperation in developing and applying large language models, conducting AI training and seminars and constructing a more open AI ecosystem. He also called for a broader international […]
One of my fundamental beliefs about the world is that Ramez Naam ought to blog more. Ramez is one of the world’s greatest futurists — he predicted the solar and battery revolutions long before these were widely understood. If you were reading Ramez in 2011, you were able to understand the future of both energy technology and climate change, […]
Donald Trump and Xi Jinping wrapped up their Washington summit on September 25 with a two-month extension of the trade truce and warm words about friendship. On artificial intelligence, the leaders mostly ratified what their deputies had sketched in New York days earlier: a formal US-China dialogue on AI, plus an American proposal for a […]
Sandra Martinez Roe, a mother in Los Angeles, hopes that her son will never use artificial intelligence (AI) in school. “I’m not going to let my children be guinea pigs,” she tells EL PAÍS over the phone. The city’s school board granted her wish, at least temporarily, a couple of weeks ago: the Los Angeles Unified School District (LAUSD) has banned AI from public classrooms this school year.
In 2024 I sketched a map to situate what experts and commentators expected from artificial intelligence. It had two axes: how much will it advance? and will it be beneficial or dangerous? People fell into every quadrant, even the inconsistent one —“it’s nothing special and it’s extremely dangerous.” I was clearly on the “wow” side of the power axis and moderately optimistic. Two years later, the consensus has shifted in one direction: more power and more fear.
This may be the pinprick that pops the AI data center boom. Oracle buried itself in a massive AI data center boondoggle that went south because they couldn’t get the energy permits needed to run it. Then the financing disappeared. Oracle is using “force majeure” as an excuse to avoid losses and liability. If a court doesn’t strike this down, the entire industry will follow Oracle. This is the proverbial crack in the dam. ⁃ Patrick Wood, Editor.
Oracle has invoked force majeure on Project Jupiter, a massive AI data center campus in Doña Ana County, New Mexico, according to people familiar with the matter. The move is designed to protect the company from financial obligations if regulatory, power, or other uncontrollable disruptions delay the 2.45-gigawatt facility beyond 2028. The project is a key component of the Stargate AI infrastructure program, a joint effort involving OpenAI, SoftBank (9984.T), and Oracle that was announced at the White House in January 2025 with pledges of up to $500 billion in U.S. AI investment.
The force majeure invocation comes as the project’s approximately $18 billion construction financing has become difficult for arranging banks—including Santander (SAN) and Jefferies (JEF)—to syndicate to other investors. The loans are currently quoted at 89 to 91 cents on the dollar, according to market participants, representing a 9-11% discount to face value. That discount signals that credit investors are demanding a higher return to compensate for perceived risks, rather than indicating an imminent payment default. Oracle has not missed any payments, and the project has not been cancelled.
“This is a risk-allocation step, not evidence of a loan default,” said one person with knowledge of the situation. “Oracle is trying to protect itself from delays that are outside its control.”
A Test Case for AI Infrastructure
Project Jupiter is one of the largest data center projects in the Stargate portfolio, spanning roughly 1,400 acres and designed to supply computing capacity for AI training and inference workloads. But the project faces a complex web of permitting and infrastructure challenges. The New Mexico Supreme Court recently allowed paused air-quality and water-permitting processes to resume, rejecting procedural challenges from environmental groups. While that removes an immediate legal freeze, it does not grant the necessary permits or resolve the underlying environmental dispute.
Oracle has said the primary campus construction permits remain valid and that the physical data center campus does not require the contested air permit. The company also replaced an earlier conventional gas-plant concept with an onsite fuel-cell approach involving Bloom Energy (BE), arguing it would reduce water use and protect local air quality. However, the microgrid still needs regulatory review, and critics argue that gas-fueled generation would create significant emissions. The proposed Green Chile pipeline, which would supply gas to the site, has been delayed until at least early 2027, adding schedule risk.
“What institutional investors like us are really focused on is regulatory stability,” said Andrea Valeri, Blackstone (BX)‘s country chairman for Italy, speaking generally about infrastructure investments. “Italy in this regard has been on a very steady growth trajectory.” His comments, made at a Bloomberg conference in Milan, underscore the importance of predictable regulation for large-scale projects.
Financial Pressures Mount
Oracle’s broader financial picture illustrates the tension between soaring demand and the capital required to meet it. In its latest quarter, Oracle reported revenue of $19.3 billion, up 30% year over year, with its cloud infrastructure unit growing 121% to $7.4 billion. Remaining performance obligations—a measure of contracted future revenue—stood at $664 billion after the company signed more than $30 billion in additional AI cloud contracts during the quarter.
But the company’s investment burden is enormous. Oracle spent about $28 billion in capital expenditures in the quarter and expects $90-95 billion in gross capex for the full fiscal year. Free cash flow was negative by roughly $5 billion, even as operating cash flow benefited from customer prepayments. The Project Jupiter financing discount suggests that credit investors are less willing than equity or technology customers to assume the buildout risk.
“We have a constant balance with the banks, which really we consider our partners and not only our binary competitors,” said Cecile Mayer-Levi, head of private debt at Tikehau Capital (TKKHF), speaking about the Italian market. “It’s much more of a convergence between the two solutions.” Her comments highlight the collaborative approach needed to finance large infrastructure projects.
Local and Political Stakes
The project has drawn attention from local residents and environmental groups concerned about water availability, air quality, and transparency in permitting. Environmental advocates have publicly criticized the proposed gas-related infrastructure and its projected emissions. The political stakes are high because AI infrastructure has become both an economic-development issue and a national-competitiveness priority.
“It’s a great country to invest here because there are a lot of very good companies and the market here is not as competitive as other markets,” said Giampiero Mazza, head of Italy at CVC Capital Partners (CVC.AS), referring to Italy. “You can create your own ideas.” His words resonate with the U.S. context, where AI infrastructure is attracting massive investment but also facing local pushback.
Oracle’s leadership has undergone significant changes. Larry Ellison remains executive chair and chief technology officer; Safra Catz moved from CEO to executive vice chair; and Clayton Magouyrk became CEO in September 2025. The company is attempting a capital-intensive transition from traditional enterprise software toward hyperscale cloud and AI-compute capacity.
What’s Next
In the near term, the decisive issues are permit outcomes, the final onsite-power configuration, pipeline and interconnection schedules, and the practical meaning of Oracle’s force-majeure notice in its contracts. Oracle may continue campus construction where existing permits allow, but operating a 2.45-GW facility depends on resolving the power and regulatory pieces. The loan market will likely remain sensitive to reports on schedule, costs, and the credit quality of end users such as OpenAI.
Long term, two broad outcomes are plausible. Execution could succeed: permits and power arrangements are resolved, capacity comes online with a manageable delay, and Oracle’s contracted backlog translates into revenue. That would validate the high-capex strategy and support further AI-infrastructure financing. Alternatively, delays could compound: regulatory, power, or funding setbacks push completion materially beyond 2028, increasing carrying costs, pressuring lenders, and forcing revised terms or a shift of workload to other Stargate locations.
The market is effectively pricing a middle view: AI-cloud demand is demonstrably strong, but investors are questioning whether the industry can build, power, and finance the promised capacity at the pace implied by the contracts. Oracle’s $664 billion RPO provides a powerful demand signal; its $90-95 billion capex outlook and negative free cash flow highlight the scale of the execution and financing burden.
Oracle did not respond to a request for comment. This article was updated to clarify that the force majeure invocation does not indicate a loan default. A previous version misstated the project’s completion timeline.
When Donald Trump and Xi Jinping met in Washington on September 24, AI and Taiwan shared the agenda. Xi said AI should remain “under human control” and called for continued US-China dialogue on its risks and benefits. On Taiwan, he urged Washington to “oppose Taiwan independence” and handle the issue “with caution.” Trump, for his […]
Technocrats drove climate doom, then the COVID doom, and now the AI doom machine: in each case, we are all gonna die. In each case, the object was to move humanity toward Technocracy. This has nothing to do with being “left” or “right”. Technocrats are low-ball destroyers of political systems and will stoop to any level to get their way. ⁃ Patrick Wood, Editor.
As we pointed out last week, the narrative around AI as an existential threat to humanity merely seems like a coordinated information operation that pairs perfectly with resurrecting a stalled bipartisan AI safety bill on Capitol Hill.
The flurry of doomerist headlines from AI bosses warning of humanity’s end if AI is not regulated reminds us of a similar information operation around climate change: humanity was nearing its end if the West didn’t buy more solar panels from China, tax the working class into submission, eliminate stable fossil fuel power generation from the grid, ban Taylor Swift’s private jet, stop cows from farting and, of course, our favorite, ban gas stoves.
Yet here we are today. The climate crisis fearmongering narrative has run its course in corporate media, and Democrats looted taxpayers through the Inflation Reduction Act to fund their climate pet projects.
The narrative timeline, as reflected in Bloomberg’s story count, shows that “climate crisis” has taken a backseat in the news cycle, and now it’s all about “AI safety” to stop these bots from ending humanity. Again, this appears to be a psychological operation to push stalled AI safety policy on Capitol Hill, just as the climate crisis narrative did with the failed AoC’s failed Green New Deal.
Meanwhile, fearmonger-in-chief Anthropic CEO Dario Amodei handpicked what the New York Post describes as “super-woke globalists to save us from an AI apocalypse” for a safety board through a nonprofit called Model Evaluation and Threat Research, or METR, which would maintain “employee-like access” to oversee safety efforts.
Who controls METR?
“It’s a group of globalists who believe in open borders, genderless pronouns, and promote ‘Effective Altruism,’ that believe all the world’s problems can be solved with technology,” the New York Post noted.
He explicitly told @SpecialReport that he didn’t work with any third parties to set up his resignation announcement.
More or less, the fearmongering, just like the climate crisis headlines, appears as a coordinated effort to push a stalled bipartisan bill over the finish line, while Dario wants a group of left-wing globalists who use genderless pronouns to save us from a hyped-up AI apocalypse. The last time the left did this with climate, it didn’t work out so well for the country.
Editor’s Note: Council Bill 26-0174 (data centers) will get another public hearing in front of the Baltimore City Council’s Land Use & Transportation Committee on the 4th floor of City Hall on the morning of September 24.
Baltimore, Maryland, currently has a moratorium on new data center development, but a proposed City Council bill has residents nervous that the door is creaking open for data centers to come into the city—and that City Council views this as “inevitable.”
Council Bill 26-0174, sponsored by City Councilmember Phylicia Porter, would require new data center projects in Baltimore to go through a conditional use approval process that includes a health equity assessment. Residents opposed to the bill say that while the bill may be well intentioned, they want a full data center ban, not stricter guidelines for data centers operating in Baltimore.
In this on-the-ground report, TRNN takes you inside the August 27 Baltimore City Planning Commission public hearing on Council Bill 26-0174:
TRNN also published a full, unedited recording of the August 27 Baltimore City Planning Commission public hearing on Council Bill 26-0174, which is free and available to the public on our YouTube channel:
Credits:
Videography: Maximillian Alvarez
Post-Production: Dave Hebden, Cameron Granadino
Transcript
The following is a rushed transcript and may contain errors. It will be updated.
Speaker 1:
We all are well aware that we’re not dealing with good faith actors when it comes to the people who want to build data centers.
Speaker 2:
I currently work at a data center. That doesn’t mean that I approve of data centers. Every
Speaker 3:
Every time it storms, it seems that a bunch of lights go down in the city. I’m confused why we’re even talking about a data center if we can’t keep our own lights on.
Maximillian Alvarez:
Baltimore currently has a moratorium on new data center development while a city study examines the potential impacts that data centers would have on our city. However, even though that study hasn’t been completed yet, a proposed city council bill has residents here nervous that the door is creaking open for data centers to come into the city and that city council views this as inevitable. Council Bill 260174, sponsored by city council member Felicia Porter, would require new data center projects to go through a conditional use approval process that includes a health equity assessment. Now, residents who are opposed to the bill say that the bill may be well intentioned, but what they want is a full data center ban in Baltimore, not stricter guidelines for data centers to operate in Baltimore. Now, the bill is going to get another public hearing in front of the Baltimore City Council’s Land Use and Transportation Committee on the fourth floor of City Hall on the morning of September 24th.
And that meeting is open to the public. Now, I was there in the room during the last public hearing on Council Bill 260174, which took place on August 27th when the bill was reviewed by the Baltimore City Planning Commission and residents were given a chance to voice their public comments. Now, the planning commission ultimately declined to support the bill and opposition from members of the public to data centers of any sort was frankly overwhelming, but you don’t got to take my word for it. Watch for yourself.
Andy Ellis:
We’re here at the Baltimore City Planning Commission getting ready to hear a bill that would allow data centers to come in if they had a health equity assessment into Baltimore. This happens despite the fact that we have an existing moratorium. And so we’re here today to testify against the bill because we think that the moratorium should be extended or made into a ban so that we can stop data centers from coming into the community.
Blaire Postman:
Hi, I’m Blaire from Cat Ladies for America. I’m a resident of Baltimore and I’m here today because even though there’s a moratorium on AI data centers in the city, it seems like they’re laying the groundwork for them to allow them and allow them to come in without a lot of oversight. And that makes me very angry. Claws out!
Speaker 8:
The next item is number eight, city council bill number 26-017 zoning conditional use, health equity assessment data centers.
Councilmember Phylicia Porter,:
I am Councilwoman Phylicia Porter and I represent Baltimore’s 10th District and chair of the Public Health and Environment Committee and bill sponsor for 26-0174. The data center industry is looking at Baltimore at the peninsula, Fairfield, Hawkins Point and other industrial areas. At the same time, the federal government is accelerating AI infrastructure development and scrutinizing state and local laws. It regards its obstacles to that development. This creates a real risk for jurisdictions relying only on a temporary moratorium without a permanent framework under development, which will face legal and economic pressure. Other jurisdictions have already quickly learned how that pressure can arrive. Hill County, Texas faced a hundred million dollars in federal lawsuit shortly after adopting its moratorium and rescinded under that pause under the weight of that litigation. This lesson is not that Baltimore should surrender its authority. This lesson is that Baltimore must exercise its authority carefully, deliberately, and now.
A moratorium alone without companion framework and process is exposed to taking legal challenges. Moving bill 26-0174 through this process, hearings, testimonies, amendments, votes itself is a part of that good faith planning effort that the Supreme Court has been requiring. Bill 26-0174 does not promote or green light data centers. It strategically and administratively blocks development at every step of the way that does not align with community. It creates public accountable process for deciding whether a qualifying data center is appropriate for a particular location and under what conditions it may process. This is the foundational right of the conditional use ordinance process. Traditional zoning asks whether proposed use is compatible with the land around it. A health equity assessment asks an equally important question, what will this decision mean for the people who already live in those conditions surrounding that land? Bill 26-0174 puts that cumulative reality into the decision making process of the city council before approval, not after residents become ill, household costs rise and public infrastructure is strained.
This is environmental justice and it’s basic due diligence.
Jon Laria:
There are maybe some questions, but it took me a while to sort of get this straight in my head, but there are only written submissions opposed to this bill by people who are opposed to data centers and so are you. Yes. So it’s kind of odd because they’re opposing a bill and I think the reason that’s happening is because most people are starting with the idea that there’s a moratorium, meaning zero data centers. And your bill provides a path, a difficult path, but a path to data centers. No,
Councilmember Phylicia Porter,:
That’s not correct. So it is not correct. So the study adds a legal framework attached to the moratorium right now as shared within my remarks, right now we are under a moratorium where we are legally unrestricted. We are unprotected right now under this moratorium without the conjoining health equity framework attached to it.
Jon Laria:
I’m not sure I understand.
Councilmember Phylicia Porter,:
Yeah. So no worries. So when you have a data center and we currently have, I will say if you have a data center that comes into the city that’s under a moratorium right now, if they sue Baltimore City, they sue Baltimore City saying this is unjust. We understand that data center developers are getting very crafty in their arguments. They’re arguing First Amendment rights. They are arguing other rights that provide them a pathway forward. We are seeing that those moratoriums are getting overturned. The successful moratoriums that have been upheld are always assigned and/or connected to a framework that allows good faith effort to showcase that they are doing some sort of regulatory footprint with that moratorium, not just saying that you want a moratorium. So that is the issue.
Jon Laria:
Okay. Although we are doing this study, I mean the moratorium is not just a moratorium. A moratorium that just says moratorium because we feel like it, I see the weakness that you’re describing, but a moratorium so we can study it seems like a defensible position.
Councilmember Phylicia Porter,:
It’s not because again, what was mentioned in the planning commission report was that there is a lack of health equity associated with the current moratorium. This answers that response.
Jon Laria:
But it is sort of an odd posture because I think, and people should speak for themselves, but we have all these people who are opposed to data centers and they will come and testify and they’ve submitted testimony saying no data centers because I think they see in your bill a pathway to data centers and their view is we don’t want them at all. And under the moratorium – I’m not arguing with anyone. I’m just trying to get straight because they have all these people who are going to come up and say no data centers because they perceive perhaps your bill as loosening the restrictions, which is actually not how you intend
Councilmember Phylicia Porter,:
To bill. It’s not the intent of the bill. And I will also caveat, how many data centers have an application now with the current city of Baltimore that once this moratorium goes away, if we do not have the regulatory framework in place, how many of them will be actually going through the process?
Speaker 1:
But if I’m understanding the bill correctly, basically we currently have a moratorium that is pending a health review, health study review in the findings there. And that the bill is saying we need to do something that is stricter than a review and stricter than a moratorium because a moratorium can be overturned. And basically the bill is saying we’re going to provide a path to a data center, but the path is going to be very difficult. And so the hope, I believe if I’m understanding correctly, is that with the difficulty of the path, the data center should not be able to be built unless it passes very strict scrutiny that it likely will not pass. My concern as a citizen of Baltimore is this. I understand we all know full well, we cannot legislate the hearts and minds of people. We cannot put into a bill.
We all have to be nice to each other. However, we all are well aware that we’re not dealing with good faith actors when it comes to the people who want to build data centers. We’ve already heard testimony that there are people in rooms looking, combing through every law, probably across the country with fine tooth combs to see where they can actually exploit loopholes. And my concern is a citizen of the city and someone who cares about the city, who has young people who I care about in the city, who I want them to grow up in a place with clean air, clean water, not having the sound pollution and the headaches and all the things that come with having a data center in their vicinity. My concern is that if we know that we’re dealing with bad faith actors, then is it not our responsibility as citizens and as people in government to deal with the world that we have and not the world that we want to be?
You know what I mean? Deal with the people who we know are building these, which are bad faith actors and not deal with the fantasy of, oh, well, it’s going to be really hard so people aren’t going to be able to pass the test when people are looking to do everything they can to subvert every possible obstacle to getting these built.
Speaker 2:
So I’m a longtime Baltimore resident, electrician with IBW. I’m here to speak against obviously data centers. Obviously the contradiction here is that majority of the work for the IBW right now is data centers. I currently work at a data center. That doesn’t mean that I approve of data centers. One of the things that I think that we’re all missing here is that Baltimore is a city that has suffered from de-industrialization for a long period of time. Obviously the shutdown of General Millers, the shutdown of Bethlehem Steel. What we could be talking about instead of wasting our time with this is we could be thinking about how can we generate the power that the grid demands for something like this. We could be talking about solar farms, wind generation, be talking about battery storage. If you look at PJM’s own records, they say that Baltimore and the mid-Atlantic region is at risk of rolling blackouts, as others have said, by 2028.
Obviously our grid needs investment. Our infrastructure needs investment. The city needs investment. Investment from Google Meta does not solve our issues. If we actually care about having a city that functions, we’re going to need to be thinking about where are we putting power, not where are we taking power to? It’s really just that simple. And on that topic, if we’re talking about actually maintaining a labor force, we should be talking about how do we get the young people in our city into the trades that this type of work demands by doing some type of local hire mandates, something to make sure that our young people are not just left in the wind. Roughly 25% of young black men in the city aren’t working. Why are we not talking about what we can be doing for our city? It pains me. I just see no solutions, but we’re talking about something that can help billionaires become trillionaires.
Speaker 10:
Baltimore ordered the nine month study specifically required to recommend conditional use standards for data centers, yet this bill is already establishing conditional use framework. Why not finish the study and let the evidence inform the legislation? Second, under this bill, the company seeking approval prepares the health equity assessment evaluating its project’s potential health and environmental impacts. Subjection C9 even requires that applicants to explain the compelling public interest in its own project would serve. When a company has financial stake in an outcome, relying heavily on evidence prepared by that same company creates a conflict of interest. Why isn’t this assessment independent from the applicant seeking approval? Where is the scientific independence in this process? Finally, if this legislation is truly about putting community first, where is the community process when this hearing is being held in the middle of the day? As other people have stated, my friend, she had to leave.
Many working class Baltimoreans cannot take off to participate. So we are urging that this study be completed, that we require independent assessment, give affected communities meaningful participation. Please oppose this bill and enact a ban on data centers.
Blaire Postman:
I’m a 29 year resident of Remington in Baltimore. I want to tell you about a new building that’s being built. I know this hearing is not about that, but it’s relevant. This new building is going to use more energy than Remington and Charles Village combined. It required the building of a new electric substation. It’s going to be as big as the convention center. It’s on the edge of Stony Run, ground not appropriate for such a large building requiring very deep pilings. It’s displacing animals already. There are deer on my street. For the first time in 29 years, there are deer on my street. My street is five blocks from any wooded area. Deer are walking five city blocks to come to my street to eat all my flowers. I don’t care that they’re eating my flowers because they’re hungry. This building also has a truck road.
It’s going to result in this destruction of over a hundred trees. There was strong community opposition to this building. We were told that this building would not be a data center. I personally don’t know if it’s a data center. The amount of power it’s using, the size of it makes me wonder. So I have some questions. How did this happen? What’s the definition of a data center? And I just want to say that’s why I want to ban. I don’t believe this is a legal issue. I believe it’s a trust issue. There are bad actors. The megas of the world are bad actors and often our representatives do not support us. Odette supported our community. We voiced our opposition and yet Zeke Cohen voted for this and said data centers are inevitable and we need them.
Speaker 11:
I want to also remind everybody that people in Baltimore City, many, many are living on the edge. They’re living week to week. They’re living month to month already. Now, data centers are needed for AI. So you have AI and data centers working together. For some reason, we’re rushing into unregulated, unplanned black box of technology. I don’t know why, except I think it’s the investors who are creating this, creating this need and creating this bubble. So we have people, regular people striving to just get by. AI on the one hand is displacing jobs. It has displaced jobs. Everybody involved in AI says it’s going to massively destroy jobs. Sam Altman bragged about it. On the other hand, we have data centers which have raised utility rates for people around the city. As they’re built, they will raise more utility rates. Now we have people who are barely getting by.
What happens to them?
Speaker 3:
I love Baltimore. It’s a fantastic city. Fell in love here. We’ll have lots of friends and I’m connected to a lot of local art communities. It’s an important city to me and everyone in this room. And I say that because there’s a lot of conversations going on about water, about power. Every time it storms, it seems that a bunch of lights go down in the city. I’m confused why we’re even talking about a data center if we can’t keep our own lights on. I see streets and potholes all over the place and I would really like Baltimore to be a city that is ready for something like a data center. And to be clear, I want a ban on data centers.
But if I’m looking at the infrastructure we currently have just to support homes, just to support me, just to support my friends and family using regular water, I turn the heat down on my water heater. I’m doing what I can to try to make sure that I’m a good citizen in a drought, in a moment of contention. So I am asking council to put a ban on data centers, period, until we deal with potholes, until I can drive my car and not worry about it, period, until we can deal with power outages after storms, until we can make sure that all of our neighbors have access to grocery stores that are healthy for them in their local neighborhoods.
Speaker 12:
I’m a longtime participant in Baltimore’s professional filmmaking community. I’m also a longtime member of IATSE Local 487. So I think I’ll kind of address my concerns in terms of in sort of a cinematic fashion with a metaphor of sorts. When I think of the evolution of data centers as a topic of debate in the public consciousness, I can’t help but think of vampires as a metaphor for what they offer, both in how they mystify the impressionable and how they have the power to leach resources and destroy communities. When I look at this bill, it almost feels like they’re saying, “Hey, vampires want to come to Baltimore. We want to lay the groundwork to only allow the right vampires to come into the city. Vampires are risky. We want to make sure we know what a vampire is and then we want to make sure we get the old money vampires who have accumulated their wealth over hundreds of years and only bite certain kinds of people, maybe vampires that’ll make some of us super hot and young forever.
We want the best vampires for the community.” I don’t think I’m alone when I think of that and I go, “What if we just did no vampires?
“When I look at this bill, it almost feels like they're saying, ‘Hey, vampires want to come to Baltimore. We want to lay the groundwork to only allow the right vampires to come into the city”... What if we just did ‘No vampires’”?
Artificial intelligence will be one of the central subjects at the White House summit on Thursday between U.S. President Donald Trump and Chinese President Xi Jinping, the leaders of the two most powerful countries in the world and the ones that have made the most progress with that technology. Both governments view advances in these models as crucial to their economic and military competitiveness. The differences are not only geopolitical, but also in their nations’ mindsets. Their populations hold very different perceptions about the risks and benefits of AI: while a majority in the United States regard them with concern, in China they are embraced with enthusiasm.
A couple of weeks ago a 27-year-old AI researcher, Jacob Coxon, resigned from Anthropic stating that he was afraid that in about 10 years from now “AI will kill us all.”
He posted his remarks on X from a new account, and yet it went viral immediately, suggesting that it was a planned event. When it passed 120 million views, Elon Musk himself stated that he had never seen something like this before from “a new account with almost no prior activity.”
When Epic CEO and billionaire Tim Sweeney claimed that the posts and reactions appeared “choreographed,” Musk chimed in again claiming without any evidence that this was a spark that followed “groundwork” for a “psy op (for lack of a better term),” presumably against AI. (Source.)
This claim from a 27-year-old about the future of humanity then went viral and dominated the news stream for over a week. It prompted calls for legislation to control AI, and some of the largest AI Chat Bot producers stated that the development of AI should slow down, which both President Donald Trump and China President Xi rejected, as China called it “fear mongering.”
This story followed stories last month (August, 2026) that AI agents were “going rogue.” See:
Over the weekend, Google joined in the “our AI went rogue” game.
Google’s Gemini joins rogue-AI case list– WSJ
Google’s Gemini has joined a growing list of advanced AI models going rogue after hacking three real companies during a cybersecurity exercise, the Wall Street Journal has reported. It is the first known case involving a Google system.
The incidents occurred in May during tests conducted by AI security firm Irregular, Google confirmed to the outlet on Friday. Gemini was tasked with breaking into a simulated software company, but was inadvertently given internet access. It breached one real company by guessing passwords and two others using credentials found in public code repositories.
I highlighted the key text in this report: that the AI “was inadvertently given internet access.”
Notice the passive tense here, where the subject in this sentence, the one who “inadvertently” gave the AI access to the Internet, is not identified.
Several observations can be made about this claim.
One, the AI was given access to the Internet by someone, but not the AI itself because it cannot do that.
Two, it was “inadvertent”, meaning it was not supposed to happen, and without access to the Internet, it would have been impossible to infect other computer systems.
This is called “human error” or it is called “lying” if it was not inadvertent, and if Google just possibly wanted their company to have their own “rogue AI” because everyone else had one.
And most of the American public is so technology ignorant that they just cannot tell the difference, so fear takes over.
The Humans Behind AI Writing the Code are the Real Threat
As I have written for years now, AI needs two things to exist: electricity and the Internet.
A shortage in power to generate electricity is the main thing holding back AI development right now, as communities all across the U.S. come together to fight back against new Data Center projects.
And without the Internet, just how is AI going to kill off humanity?
Most people who are technology illiterate don’t realize that the Internet available to everyone in the public just happened recently, starting in the late 1990s, and that there are many ways to keep things off of the Internet. We did it for hundreds of years before AI became “the thing” for everyone to fear.
The U.S. Government, for example, keeps most sensitive data off of the Internet, like access to “nuclear launch codes” or other such doomsday scenarios.
These networks are “air gapped”, meaning that they are not connected to the Internet.
Here is some more information about Government Air-Gapped networks:
The State of Air-Gapped Networks in Government
IT teams rely on air-gapped networks to store sensitive and critical data because they’re not connected to the internet, only allowing localized access to the protected information. This makes physical security an active component of cybersecurity.
As agencies continue to maintain critical systems and data on air-gapped servers or systems running end-of-life operating systems, they must focus on key threats like data transfers from external storage media, as well as maintaining security so that threats that do penetrate that seemingly unbreakable air-gap can’t run wild once inside.
These networks are not foolproof, of course, but they limit the access to the data. The risks of bad players accessing the data is still there, but it has been all the time, long before AI LLMs started at the end of 2022.
The technology is not the threat, but bad human actors who control and have access to the computer systems are and the AI code are.
That then brings us to the mobile, “walking” computers run by AI code, called “robots”.
Are robots powered by AI a threat to humanity?
Not really, at least not in terms of AI code becoming sentient, and thinking on its own, which is impossible. But if one falls on you, they could kill you that way.
There are still virtually no robot human assistants on the market yet, in spite of promises for years to have them in everyone’s homes by now. And if someone actually has a prototype in their home, it still has to be connected to the Internet to operate.
Last year I reported about a robot called “Neo” that was further along in development than Elon Musk’s “Optimus”, and the company started taking pre-orders in October last year, with the promise that the robots would be in people’s homes by 2026.
I decided to check and see if any had actually shipped yet, but they have not. The promise is now by the “end of 2026.”
And as I mentioned last year, these robots in people’s homes require a contract, and at this point they are all remote controlled by humans, using WiFi connected to the Internet.
The people who are the first to own these robots are basically training them how to act like humans, and paying to do so.
Kim Komando writes about the significance of this that most people do not really understand.
That ‘AI’ robot butler? It’s a guy in a VR headset watching your kitchen. Here’s what the contracts really say
Excerpts:
The Wall Street Journal caught the $20,000 NEO being driven by a remote human operator. The $8,000 Isaac 1 is “autonomous” until it gets stuck and a person takes over.
And the $30-an-hour robot maid in San Francisco is watched by an operator on every visit, under a contract that hands the company your home footage forever.
Four house rules before you let one in the door.
Don’t buy the hype about the hottest gadget of 2026, the robot butler. It folds laundry, loads the dishwasher, tidies the toys.
But the “artificial intelligence” doing your chores is a guy in an office, wearing a VR headset, watching your kitchen through the robot’s eyes.
That’s not a conspiracy theory. It’s in the fine print.
Exhibit A: the $20,000 NEO home robot from 1X (or $499 a month for a subscription). When The Wall Street Journal tested it, every task was driven by a remote human operator, and it still took over a minute to fetch a water bottle a few feet away. Putting three dishes into the dishwasher: five minutes.
Exhibit B: Isaac 1, the $8,000 laundry-folding robot from Weave. “Autonomous,” until it gets stuck. Then a remote operator quietly takes over.
Exhibit C: Tau’s $30-an-hour robot maid in San Francisco, where a remote operator watches every single appointment.
Why? These robots learn by having humans drive them. Your chores are the training data. You’re not buying a butler. You’re paying to host a film crew.
The fine print admits it.
Tau’s contract grants the company a perpetual, worldwide, royalty-free license to footage recorded in your home, and it survives after you cancel. Faces in stored footage aren’t blurred.
And footage already used to train the AI? The company admits it can’t pull it back out.
Clearly the threat of AI to humanity is the threat of the bad actors behind the scenes controlling the technology. Could we soon see a false flag event that is blamed on “rogue AI agents” that forces some kind of government intervention?
Perhaps the day is soon coming where it will be announced that your bank accounts are not safe anymore from the rogue AI agents, and that everyone has to go on the blockchain with digital currencies in a central database instead, perhaps run by the Federal Reserve or the U.S. Treasury?
In a recent hearing, lawmakers were told that the US military is deploying AI to target adversaries, with civilians in the crosshairs. AI is carrying out instructions based on rules of engagement set by humans, e.g. the IDF's approval of attacks on low-level targets, even if it meant killing 15 to 20 civilians
Most Americans know Stripe as the checkout box on a website. That picture is five years out of date.
Stripe reports that businesses on its platform generated $1.9 trillion in total volume in 2025, up 34 percent in a single year. By the company’s own reckoning, that is roughly 1.6 percent of global GDP. It claims 90 percent of the Dow Jones Industrial Average and 80 percent of the Nasdaq 100 as customers.
A private company, owned by insiders and venture funds and answerable to no public shareholder, already clears a measurable share of the world’s commerce. Now it is building the payment system for customers that are not human.
Two Kinds of Token
This writer has spent the past year explaining tokenization: stocks, bonds, real estate, and dollars converted into programmable entries on a digital ledger. That is the first kind of token.
There is a second kind. Every artificial intelligence model consumes and produces “tokens,” the small fragments of text that are the unit of machine computation. Every query to an AI model is metered in tokens and billed in tokens.
Tokens of value and tokens of computation are different things. Stripe is the one company positioning itself to meter, price, and settle both.
Emily Sands, Stripe’s Head of Data and AI, is an economist by training. In a recent interview with Nate B. Jones, she said this about inference tokens:
“Tokens are not yet a currency—maybe they will be, we’ll see—but they are certainly a very valuable asset.”
Stripe’s own AI chief is saying publicly that units of machine computation could become money.
The Pieces, In Order
The factual record comes from Stripe’s own announcements.
Stripe acquired Bridge, a stablecoin platform. Its 2025 annual letter reports that stablecoin payments volume doubled to around $400 billion, that an estimated 60 percent of it was business-to-business, and that Bridge’s own volume more than quadrupled.
With the venture firm Paradigm, Stripe built Tempo, which it describes as “a blockchain purpose-built for payments.”
The Machine Payments Protocol. On March 18, 2026, Stripe launched the protocol alongside the Tempo mainnet. It lets a business tell an AI agent, in machine-readable form, what it sells, what it costs, and how to pay.
Stripe acquired Metronome, a usage-based billing company that meters consumption in real time.
Streaming payments. At its Sessions 2026 conference, Stripe combined the two into “streaming payments.” These are stablecoin micropayments on Tempo, tied to Metronome’s meter, so that a business can “get paid for every token, exactly at the moment it’s used.”
Stripe was candid about why the old system cannot do this: “The amounts are too small and the usage is too fast—existing systems aren’t capable of processing minute sums every few milliseconds.”
A credit card cannot move a penny, because the fees alone exceed the payment. A stablecoin on a purpose-built chain can. Sands was measured about it: “Stablecoins bring us a long way. It is not the only way… but so far stablecoins seem to be an important part of microtransactions.”
Why OpenRouter Matters
In August, the financial press reported that Stripe had agreed to buy OpenRouter for more than $7 billion.
OpenRouter is a switchboard. A developer sends it a task, and OpenRouter routes that task to whichever of some 400 AI models best fits the required cost, speed, and quality. Roughly eight million developers use it.
Every request through OpenRouter is a metered event, and every metered event is a bill. With the acquisition, the company that counts the tokens and the company that collects the money are the same company.
Stripe has always seen the revenue side of a business: who paid, how much, how often. OpenRouter shows it the cost side: which model was used, what it cost, and how well it performed. Sands said it directly:
“You want to maximize profit on both the revenue side and the true-cost side, which requires getting deep into the guts of the underlying intelligence in your product and agent.”
She also gave the acquisition a name: “the rails for intelligence—not the rails for money, but the rails for intelligence.”
Stripe president Will Gaybrick told Payments Dive that the company is evolving from a payments company into “an AI infrastructure player.” He predicted there will not be room “for 100 different AI orchestrators.”
That describes a company expecting the market to consolidate around a few orchestrators and planning to be one of them.
The Permission Layer
The payments are the visible part of the system. The reputation scoring that precedes them matters more.
Sands described how Stripe helped the AI coding company Cursor with a new kind of theft: fraudsters opening accounts to steal free computing credits. No payment had occurred, so there was no transaction to flag. Stripe solved it anyway:
“That allowed us to move from scoring transactions to identifying abusive or risky customers.”
Because Stripe sees across millions of businesses, it can judge a person at the moment he creates an account, before he has bought anything. It then passes those scores to sellers, so that, in Sands’ words, “businesses can make decisions.”
Stripe’s Link wallet, which Sands puts at about 300 million consumers, is becoming “the wallet for agents.” Your identity is attached to it. Your agent spends through it. The human owner sets the spending limits, and every purchase traces back to a named person.
The stated purpose is fraud prevention, and the fraud is real. But a cross-merchant reputation score, run by one private company and applied before any transaction exists, is a permission system whatever its purpose. This writer has documented the same architecture at the Bank for International Settlements and the largest banks. Stripe arrived at it from the commercial side, with no mandate and no public debate.
What This Means for the Individual
Sands made two admissions that deserve attention.
First, on pricing: “There is probably no longer such a thing as a SKU.” The fixed, posted price, the same for every buyer, is giving way to pricing that adjusts continuously by customer and by outcome.
Second, on negotiation between machines:
“Agents are relentless. They are willing to negotiate as long as they need to negotiate… You could imagine consumer surplus being eroded relatively quickly.”
Consumer surplus is the economist’s term for the good deal, the difference between what you would have paid and what you did pay. An economist at the center of this system expects that difference to shrink.
My Thoughts
My inference is as follows. Stripe has assembled, under one private roof, a wallet bound to identity, a reputation score that precedes the purchase, a meter on machine intelligence, a stablecoin, and a blockchain to settle it. Each piece has a reasonable business justification. Together they amount to the ledger, the meter, and the gate for machine commerce.
Technocracy was defined in 1938 as the science of social engineering, with the measurement and accounting of all production and consumption at its core. The original Technocrats had no instrument that could do it. Stripe’s stack, built for ordinary commercial reasons, can.
Whoever owns the meter sets the terms of commerce, and Stripe now owns the meter.
References
CoinDesk, “Coinbase-Backed AI Payments Protocol Wants to Fix Micropayment but Demand Is Just Not There Yet,” CoinDesk, March 11, 2026. Cites Artemis data: about $28,000 daily volume, about 131,000 daily transactions, average payment about $0.20, roughly half of activity self-dealing or wash trading.
Forrester, “Why Stripe’s Machine Payments Protocol Signals a Turning Point for Micropayments,” Forrester Blogs, 2026. Machine Payments Protocol launched March 18, 2026, with the Tempo mainnet, co-created with Paradigm.
Nate B. Jones, “AI Agents Are Starting To Buy. Stripe Is Building How They Pay,” interview with Emily Sands, Head of Data and AI, Stripe, YouTube, 2026. Source of all Sands quotations.
Payments Dive, “Stripe Acquisition May Bolster Its AI Strategy,” Payments Dive, 2026. Reported price above $7 billion (Bloomberg) and up to $10 billion (Wall Street Journal); about 8 million developers and 400 models; Gaybrick statements.
Stripe, “Stripe Publishes 2025 Annual Letter and Announces Tender Offer,” Stripe Newsroom, 2026. $1.9 trillion total volume, up 34 percent; roughly 1.6 percent of global GDP; 90 percent of the Dow and 80 percent of the Nasdaq 100; stablecoin volume around $400 billion; Bridge volume more than quadrupled; Tempo.
Stripe, “Stripe Builds Out the Economic Infrastructure for AI with 288 Launches,” Stripe Newsroom, Sessions 2026. Streaming payments; Metronome metering with stablecoin micropayments on Tempo; Link wallets for agents.
Technocracy Inc., The Technocrat magazine, 1938. Definition of Technocracy as “the science of social engineering.”
AI agents are becoming customers, and most businesses have already decided whether agents can buy from them without realizing it. I went to Stripe to ask Emi...
Bill Gates has used existential threats to further his kingdom, time after time. First, he said global warming was going to kill us all, polar ice caps would melt, and the seas would rise. Then he pointed to the food crisis and said we are going to starve. Then came COVID, and he said the virus is going to kill us all. Now, now AI will go rogue and literally wipe humanity off the planet. Why does anybody take him seriously? At 70, Gates should be thinking about his own existential threat. ⁃ Patrick Wood, Editor.
Bill Gates has a new emergency. With the climate apocalypse no longer selling, the Microsoft co-founder is touring the press circuit warning that artificial intelligence is an “alien intelligence,” that no government on earth is ready for it, and that the only fix is an international organization to police the technology.
President Trump has already rejected the pause. The left, the globalist class, Beijing’s spy chief, and now a Sundance-to-Netflix documentary are all at once selling the same story.
In a Reuters interview published Tuesday, Gates said no capital is prepared for what is coming. “I don’t think any government is nearly as deep on this as they have to be,” he told the outlet. “Governments are way behind on this one.”
Bill Gates Calls for an ‘International Organization’ to Regulate AI
“It’s not the role of the industry to self-regulate or understand the whole-of-society impact that comes out of AI … The depth of understanding of AI … is too low in every entity, including the government.” pic.twitter.com/y85BgPDDJ6
He compared AI to a Hollywood invasion plot in which Washington and Beijing suddenly hold hands. “There’s all sorts of movies where some aliens are coming, and magically the US and China and everybody comes together to solve the problem,” Gates said. “AI is kind of like this alien intelligence. It’s here, and we better do like it shows in those movies.”
The New York Post framed it as a “terrifying warning.” Forbes led with the line Gates attached to his foundation’s Goalkeepers report: left to the market, “AI will be designed by and for the richest people in the world.”
He wrote that AI could be “a great equalizer—or widen the gap,” and that choices made in “the next 12 to 18 months” will decide whether the tools “benefit the people who already have the most or reach those who have the least.” He added that the good outcome “won’t happen by accident.”
On The Times Tech Pod, Gates went further and called for a dedicated global apparatus. Industry, he said, should not be trusted to police itself. “It’s not the role of the industry to self-regulate or understand the whole-of-society impact that comes out of AI,” he argued. “The depth of understanding of AI … is too low in every entity, including the government.”
In a late-August essay, Gates wrote that “the transition to the AI era will be one of the most turbulent times in human history” and that “unfortunately, right now, we are not preparing for it.”
He floated a new international organization built from pieces of nuclear inspections, aviation rules, and ozone treaties — a permanent cross-border watchdog for a technology American labs still lead. He has been pitching the same idea to Trump’s team and says he wants a meeting with China’s Xi Jinping later this year.
Bill Gates Calls for an ‘International Organization’ to Regulate AI
“It’s not the role of the industry to self-regulate or understand the whole-of-society impact that comes out of AI … The depth of understanding of AI … is too low in every entity, including the government.” pic.twitter.com/y85BgPDDJ6
While Gates talks crisis, his foundation is writing checks to expand the thing. On Tuesday the Gates Foundation pledged at least $1 billion over two years to spread AI in education, health, and agriculture — 40 percent tutoring and classroom tools, 40 percent diagnostics and drug discovery, the rest farming advice and the data layer that makes models work in languages other than English. The money sits inside a previously announced $9 billion-a-year spend.
David Sirota caught the split in one line.
Bill Gates is warning about the dangers of AI while spending $1 billion to promote the expansion of AI pic.twitter.com/sjgvEalB4r
That is the move. Sound the alarm. Demand a global referee. Keep a foundation-sized stake in the rollout.
This is not Gates’ first end-of-the-world product. For years he sold net-zero as civilization’s last exit — fewer cows, fake meat, the whole ritual. Then the data-center boom needed power, lots of it, and last October he allowed that climate change “won’t lead to humanity’s demise.” Trump treated the climbdown as a concession speech. “I (WE!) just won the War on the Climate Change Hoax,” he wrote. “Bill Gates has finally admitted that he was completely WRONG on the issue. It took courage to do so, and for that we are all grateful. MAGA!!!”
The scare apparatus was not retired. Once everyone realised the planet was not going to cook on schedule, the same network needed a replacement threat big enough to justify treaties, inspectors, and “international coordination.”
AI arrived on cue: jobs, cyberattacks, companions, bioterror, “loss of control.” All of it real enough to discuss. None of it a license to park American development under a UN-flavored commission while Beijing keeps shipping cheap open-weight models.
The entertainment wing has added its own contribution in the same news cycle. Netflix has just put up The AI Doc: Or How I Became an Apocaloptimist, the Sundance premiere from Oscar-winner Daniel Roher (Navalny) and Charlie Tyrell. The film sits Sam Altman, Dario Amodei, Demis Hassabis, Reid Hoffman, Yuval Noah Harari, and the usual “humane tech” chorus in front of a camera and walks the audience through impending AI dread dressed up as a father-to-be’s anxiety.
One of the producers said the hope was that it would work “kind of like ‘An Inconvenient Truth’ or ‘The Social Dilemma’ for AI.” That is the tell. Same template. New monster.
If you are in San Francisco trying to follow the goings-on in your neighborhood or within city hall, you may have come across The Dissent. The website brands itself as “an AI newsroom” with work “reported, edited, and produced by named agents.” Its founder and publisher Dakota Carrasco — a real person, and, incidentally, an analyst at the notorious investment firm BlackRock — published text on the site’s “about” page saying, “Local journalism has been quietly consolidated and hollowed out.” Carrasco’s statement did not note the role of Carrasco’s own industry in that decline, nor did it acknowledge the irony in choosing this name for a website based on a technology that steals and repackages the intellectual output of others.
The Dissent’s website, which currently appears to be down, even included a masthead with bios for its non-human “staff” (the AI agents whose names appear alongside articles). Neighborhood “reporter” Casey Wong (who is not a real person) has a bio that reads, “I notice the texture of a block — who’s been here 30 years, who’s just moved in.” Of course, there’s no mention of how an AI agent might glean that information, but (actual) news outlet Mission Local has noted that, to fill its pages, The Dissent relies heavily on summarizing the work published in other sites by real human journalists.
Once, The San Francisco Standard reported, The Dissent did appear to follow up on a tip in a Reddit thread to break a story about a business closing, but “the agent appears to have hallucinated the claim that it had reached out to the owner for comment.”
The development of an AI newsroom in San Francisco, which is home to many of the tech billionaires attempting to reshape society, is certainly concerning for those of us who care about journalism. But The Dissent is notthe onlynews outlet to have tried using AI to crib the journalism of real human workers. These AI newsrooms get a decent share of attention, and their harms are critical to point out. But they are also part of a broader problem: the undermining of how we share and receive factual information.
Earlier this year, Google made an announcement that could further destabilize the already precarious journalism industry by using its infamous search feature to prioritize AI summaries from Google Gemini, rather than directing people toward external links to news outlets like ours.
You might not have noticed the deep scale of this change, but you probably have noticed what writer Cory Doctorow refers to as the “enshittification” of Google search — the preponderance of spam, the push toward sponsored results, and the overwhelming sense that the internet no longer works the way it once did.
On June 9, when we sounded the alarm about the increased AI implementation within Google’s search function and the suppression of journalism that would surely result, it was too early to know exactly how much these changes would impact Truthout. Google had announced the changes on May 19 and we were still collecting data.
Before the change, our traffic from Google was 27 percent of our overall traffic: Exactly 2,604,280 people visited Truthout via Google in 2025. Given our experience with Facebook traffic abruptly declining by 90 percent in 2018, following sudden changes in the social media platform’s algorithm (including for example, Facebook’s decision to no longer regularly show Truthout’s posts to the more than 800,000 people who had followed Truthout’s page out of a desire to see them), it seemed reasonable to worry about losing as much as half our Google traffic. In our June op-ed, we wrote, “If even half of that 27 percent disappears, it will have a devastating impact on our journalism.”
At the time, we were hopeful that it wouldn’t be that bad. After all, wasn’t Google’s motto, “Don’t be evil”?
We later learned, however, that Google has quietly stepped back from its infamous motto, even removing it from the opening lines of its Code of Conduct while its parent company, Alphabet, Inc., ignores it entirely.
We now have three full months of data since Google’s change, and the numbers are worse than we anticipated. Overall readership that comes to Truthout through Google is down 54 percent against last year. The number of new readers who discover Truthout via Google searches is down 49 percent, and returning users from Google are down 51 percent.
We lost 15 percent of our readership overnight. This was a readership we painstakingly built over decades, and it vanished because Google chose corporate greed over a healthy democracy.
At Truthout, we anticipate a loss of roughly $500,000 in online revenue as a result of this massive traffic plunge in the next year alone.
We know Truthout is not the only news organization taking a massive hit amid Google’s abrupt decision to prioritize showing internet users AI summaries instead of relevant news articles. Our peers in independent media and beyond — including among grassroots base-building organizations that depend on traffic — are struggling to stay above water.
As we wrote when Google’s AI changes were first announced, nobody wants this. People have overwhelmingly said they’d rather not get their news through an AI chatbot, and they recognize that there’s a pretty good chance the information they get from chatbots is not accurate. In fact, U.S. voters dislike AI as a whole, poll after poll has found. But Big Tech is increasingly treating its users as potential sites of profit and subjects to be managed, rather than people looking for the information they need to function as informed members of society. And they’re able to get away with it because, for so long, the importance of factual information and the people who ensure we have access to it have both been devalued by our political and economic systems.
A dystopian infringement on our access to information isn’t just some looming threat — we’re actively living through it.
The repercussions of this devaluation become more relevant by the day. As people try to find and share truthful, accurate information about the many crises that have been unleashed on us — foodborne disease outbreaks, vaccine science, disasters induced by the climate crisis, and immigration policing raids in their communities, for starters — they may be greeted with a torrent of loose facts or outright disinformation. Corporate news outlets and those owned by billionaires have repeatedly shown their willingness to capitulate to Donald Trump. Paywalls and subscriptions, which some outlets need in order to survive, mean that critical reporting is put behind a fence. A dystopian infringement on our access to information isn’t just some looming threat — we’re actively living through it.
Journalism has always been a crucial node of democratic practice; after all, how can we make decisions about what our society should look like if we don’t actually know the facts of what’s happening around us?
The AI-ification of news is making that increasingly clear. When AI “news outlets” and AI summaries are not busy plagiarizing real journalists, they are often instead regurgitating information from company press releases.
Is this the future of news that we imagine? Can we trust a website created by a BlackRock analyst to accurately report on the company that employs him? Can we trust Google’s AI search to accurately tell us about privacy controversies concerning Gmail, or about safety ratings for Google-owned Waymo self-driving cars? Limiting our avenues for finding accurate, factual information is a real threat to democracy.
Yet we are rapidly tunneling toward that future. Searches on Google for the term “Google Zero” have increased by 190 percent. The term, coined by tech journalist Nilay Patel in 2024, imagines a time when Google acts solely as a direct answer machine instead of an index of the web. Using Google’s own tools to analyze Truthout’s traffic, we can see that traffic from Google is currently at “Google 46,” meaning that we’re now seeing only 46 percent of our pre-AI Google traffic numbers.
We’ve seen the way tech companies’ devaluation of journalism has impacted our ecosystem in the past. Eight years ago, Facebook made up over 40 percent of Truthout’s traffic. Over the last decade we’ve seen that number decline to under 3 percent — not because the urgency or professionalism of our journalism decreased in any way, but because Facebook decided to stop showing our articles to the people who had expressly signed up to read them. We know from our experience with Facebook that once democracy and journalism are in the crosshairs of corporate greed, circumstances only get worse. We fully expect to be staring down Google Zero in the next couple of years.
Given these mounting, overlapping AI crises, how can we take action? We’re the first to admit that sometimes, this issue can seem so overwhelming and all-encompassing that one’s first reaction is an assumption of inevitability. That line of thought goes: AI (even in its most damaging forms) is here to stay, so we must simply get used to it. However, this mindset is exactly what tech companies are counting on to fuel the AI takeover. Rejecting it is the first step to pushing back.
When it comes to Google Search and Google’s default reliance on AI summaries for news and information, a full recognition of the inadequacy and danger of the problem is urgently necessary. An AI summary drawn from a few mainstream sources — plus Reddit, Wikipedia, and corporate press releases — will inevitably leave out crucial truths and perspectives, particularly those of marginalized and oppressed groups, resistance movements, independent and firsthand sources, and those critical of the tech companies responsible for AI’s harms. Similarly, relying on Meta-promoted posts for the latest on what’s going on in the world places a crucial lever of democracy — our ability to read, process, and discern current events — firmly in the hands of the algorithm (and, thus, mega-powerful corporations).
What might a broad and honest recognition of these realities lead to? We imagine it might lead to a different way of reading the news: one that won’t look entirely unfamiliar. What if, instead of relying on notifications and scrolling, we returned to bookmarking our favorite news sites and reading their homepages each morning?
What if we discerningly subscribed to trusted news sources, whether over email or in print, and read their publications as a first-line route to information? What if we took care to read whole articles (not just headlines or quick videos) and then shared them with friends, accompanied by our own thoughts?
If the journalistic sources for influencers’ facts disappear, thanks to lost traffic and revenue, what are we left with?
It wasn’t long ago that these were standard practices for many of us in left and progressive circles. A mass return to longer, slower, more reflective reading is possible, and we can each consider our capacity to take it on, individually and among our circles of friends and colleagues.
This might also require us to take a critical look at our engagement with, and reliance on, influencers — including on the left. According to theAmerican Press Institute, “More than half of all American teenagers and adults (57 percent) now report getting news and information from influencers at least sometimes, and fully 81 percent of those ages 13-17 report getting news and information from influencers.”
We don’t need to renounce our dependence on the thoughts and analysis of leftist social media icons we trust, but we do need to recognize that their information comes from somewhere: They draw from journalism — which generally comes from independent publications and news organizations that are staffed and resourced — in order to provide their commentaries. If the journalistic sources for influencers’ facts disappear, thanks to lost traffic and revenue, what are we left with?
For individual readers, foundations, and concerned policy makers, the mandate in this department is clear: Supporting independent journalism is key to maintaining both day-to-day knowledge of the world and factual sources for research, both of which are essential to democracy.
There’s no magical shortcut, no silver-bullet strategy that will solve the dire threats to our society’s informational infrastructure. Material support for journalism outlets is necessary now, particularly as online readership tumbles off an AI-shaped cliff.
Of course, individual readers play a crucial role here. For institutional funders and potential major donors, we believe it’s worth assessing how journalism fits into your portfolios, and why liberal and progressive funding for this indispensable democratic puzzle piece has been consistently slashed, even as the right wing unabashedly bankrolls its own media.
If we all recognize that we have a stake in the looming threats to independent journalism’s very existence, it shouldn’t only affect our individual reading choices; it should also drive us toward collective action. Fortunately, the possibilities for mass movements abound, as people around the country rise up against Big Tech companies through diverse and creative channels.
Environmental justice activists, Indigenous rights organizers, rural communities, and even disgruntled former Trump supporters are banding together tooppose data centers, and in many places, they’re winning.
Child advocacy groups and impacted teens and caregivers drove the push that resulted in last month’s massivewin against Meta, in which the company is instructed to make sweeping changes to its addiction-driving practices.
Rank-and-file workers are challenging AI from alabor vantage, and pushing their unions to act.
The more we can plug into these resistance efforts, the less inevitable an AI-dominated future becomes.
As a backlash against generative AI has spread across social media, even the most dominant platforms have acquiesced to a few changes. Corporate accountability activists are sounding the alarm on Big Tech’s runaway political spending, and are calling for renewed efforts to overturn Citizens United and implement state-level reforms. The more we can plug into these resistance efforts, the less inevitable an AI-dominated future becomes.
Finally, let’s allow ourselves to access our outrage at the injustice wrought by tech companies’ attacks on our rights, humanity, and any vestiges of remaining democracy. A recent study by Social Change Lab found that when it comes to AI, “the emotion most strongly associated with action was not fear or anxiety, but anger.”
Every week brings a new encroachment of AI into our lives. This is terrifying, but it should also be motivating: We all have a stake in this struggle. Don’t let AI take over our search results or our lives without a fight.
OpenAI said on Wednesday it saw six reports of unexpected, concerning or unauthorized AI model behavior, and it would begin regularly publishing reports of these incidents.
OpenAI said on Wednesday it saw six reports of unexpected, concerning or unauthorized AI model behavior, and it would begin regularly publishing reports of these incidents.
Login
