Working from home is still one of the most important uses of broadband, particularly in rural communities. Roberto Gallardo, who writes in Substack as the “Data Whisperer”, recently wrote an article discussing the change over time of those who work from home. For those who don’t know Roberto, he’s an Associate Professor and the Vice President for Engagement in the Agricultural Economics Department at Purdue University.

Everybody remembers the explosion of working from home during the COVID-19 pandemic. A lot of people decided that they liked working from home, and the percentage of those doing so has not dropped to anywhere near pre-pandemic levels.

The article starts out by comparing the percentage of those who worked from home in 2014 and 2024. Overall, the percentage of those who are working from home more than tripled over the decade. In 2014, 4.4% of the U.S. labor force worked from home, and in 2024 that had grown to 15.1%.

A hacker collective pulled down a Flock camera and dumped its data. The files included thousands of videos and logs showing that the device captured 1.6 million images of 50,000 vehicles in 21 days.

Silicon Valley says its machines could destroy humanity. The more immediate danger is letting the people building AI decide how they’re governed.

Something very weird is going on in our political economy discourse.

Over the past month, the executives at the top AI companies have engineered a slow-motion panic over AI, snowballing into an argument that they are on the verge of creating a technology with a reasonable chance of ending human civilization. A few days ago, Jacob Coxon, a 27-year-old AI researcher at Anthropic, resigned his role and posted a thread on X, arguing these companies are building super-intelligent systems that may destroy the world. Coxon’s comments garnered a front-page Wall Street Journal story, more than 110 million views, and wide commentary from lawmakers, CEOs, AI experts, and journalists.

Then another 20-something Anthropic employee, Evan Hubinger, chimed in to say that most employees at Anthropic agreed with Coxon, and “do earnestly believe AI could kill all humans,” putting the odds at 10 percent within the next decade. These are extraordinary claims, and the people making them are nicknamed “AI Doomers” or sometimes called the “AI Safety” movement. This group loosely encompasses the network that used to surround disgraced crypto billionaire Sam Bankman-Fried, a quasi-religious order known as effective altruists, who believe a cadre of hyper-rational elites can guide humanity. Despite their weirdness, this technology's capabilities are improving, and some dangerous events have happened recently.

We’re never going back to the human internet we had. But we can hope for a better internet than the AI-dominated one that is emerging, writes Sarah Gilbert.

On November 30, 2022, the Internet became less human. That was the day OpenAI launched ChatGPT, and it didn’t take long for social media to be flooded by a tidal wave of AI generated content from it and other services that launched in its wake. Only three and a half years later, one estimate says the internet is around 16% AI-generated, stoking fears that it is becoming less human every day. But AI-generated content isn’t distributed equally. While some sites, like LinkedIn, may be as much as 40% generated by AI, others sites, like Reddit and Wikipedia have far lower rates: only around 5%.

Why are Reddit and Wikipedia so much more human than LinkedIn? One reason is that they rely on the work of volunteer caretakers, known as moderators, who curate content and protect community members. The launch of ChatGPT and other generative AI models added a new job for moderators: keeping their communities human—a job that’s as grueling as it is fraught.

Why does this matter?

Learn which U.S. states are meeting or exceeding the FCC’s 100/20 Mbps broadband speed standard, how the digital divide is shifting, and where Starlink fits in.

Using Ookla Speedtest Intelligence® data, this report identifies the states that are currently delivering the minimum standard for fixed broadband speeds as established by the Federal Communications Commission (FCC) to the highest percentage of Speedtest users.

It also singles out the states that need the most improvement when it comes to delivering the minimum standard for broadband to their residents thus helping resolve the impact of the Digital Divide. 

Published September 15, 2026.

Data Collection Period: January — June 2026

Below is a preview of the report, with the full report offering detailed findings across all 50 states.

A few months ago, perhaps out of both curiosity and fear, I asked an AI chatbot to draft a hypothetical 800-word opinion piece in the style of my writing voice. After taking seconds to analyze years of my journalistic work, the chatbot delivered a “draft” that was both laughable and sad to read. I was also angry that the draft reduced decades of my own creativity to what it thought was a polished and publishable article. It felt cheap and mediocre. 

When I started seeing Facebook posts earlier this month about how The Baltimore Sun was now generating AI cartoons on its editorial pages, the anger came back. One specific cartoon represented everything AI slop cannot do in the creative space. The image of Baltimore Mayor Brandon Scott meeting with supporters features bizarre dialogue balloons, deer-in-the-headlights gazes, random checklist items and the most unattractive sweatshirts ever. It just feels fake and embarrassing. By the way, those creepy AI images have become a regular thing for The Sun.

Nobody should be surprised. In 2025, the outlet fired editorial cartoonist Kevin “KAL” Kallaugher, a two-time Pulitzer finalist who had worked there since 1988. Known for his scathing political cartoons — particularly of Donald Trump — Kallaugher attributed his firing to right-wing Sun owner David Smith. You may know him better as the executive chairman of Sinclair — the giant local-TV broadcaster that controls hundreds of local stations, including its Baltimore flagship, and is best known for forcing its anchors to parrot conservative talking points. Smith purchased The Sun in 2024, pushing the outlet more to the right and onto a path of irrelevance.

The Benton Institute recently wrote an article that reminded me that Starlink is operating a second satellite network, separate from the one used to serve residential and business broadband customers. The company announced this network, which it dubs Starshield, at the end of 2022. Today, this network is now highlighted on the SpaceX website.

Starshield is touted as a military-grade network, which in layman’s terms means a separate network that doesn’t share any traffic with commercial customers. In the U.S., the National Reconnaissance Office (NRO) has launched several hundred Starshield satellites. This is the federal agency that has always operated military and spy satellites.

Late last week, everything exploded when former Anthropic AI researcher Jacob Coxon, in an exclusive interview with the Wall Street Journal, warned that he was “quitting the AI industry” (he wasn’t) over “...fears that the lab and its competitors are racing to build systems they won’t be able to control.” 

His fears were centered around the creation of “recursive self-improvement,” a still-theoretical concept of AI that trains itself autonomously” and otherwise expressing few specific concerns beyond that “AI labs are unable to control AI,” always phrasing things in the terms of impossible-to-control entities rather than poorly-programmed cloud software running on the infrastructure of the largest companies in the world. 

Emily Forlini of Fortune put it best:

Hear me out: He claims the AI could kill us someday, but doesn’t point to any projects in the pipeline that could be shut down to avoid this. He says AI companies are moving too fast, but neglects to share screenshots, emails, or specific examples of when this behavior went sideways—when it became clear to leaders at AI companies that the technology was slipping beyond their control, for instance, and how the decision-makers disregarded the warning signs. He doesn’t suggest any new legislation, name problematic leaders that should step down, or post an in-depth look at how Anthropic researches new models and propose a new approach.

This is because, in my opinion, Jacob does not really care about the actual harms of AI, whether we’re talking about Large Language Models or something he imagined while working with the non-profit or PR firm that set up a CBS interview where he claimed that AI that, if we’re talking about LLMs, have model weights of terabytes of memory, would make ten thousand copies of themselves. 

Or, of course, bullshit like this:

"It doesn't look that different from, say, 'Terminator' or from science fiction films," Coxon told CBS News Thursday. "It will be smart enough to kill us." 

At no point did Coxon bring up how ChatGPT was used as a “suicide coach,” directly caused a murder-suicide, or aided and abetted in mass shootings in Florida and Canada, or the horrifying gas turbines poisoning black communities. His own discussion of the Hugging Face attack — much like all of his criticisms — focuses on the anthropomorphization of large language models as this unknowable, unstoppable force, with no real responsibility for anyone involved.

What the US president exhibits is a fundamental incompetence that places us all in grave danger.

On Sunday the President of the United States stood on a golf course in County Clare, Ireland and was asked what he intended to do about the most serious technological warning of our lifetime. His answer, in full, was that America is beating China. “We’re leading China in AI. We’re the most sophisticated country in the world, and frankly, I want to keep it that way, because whoever wins AI wins.” He added that “you have a lot of negative forces that are bringing it up that shouldn’t be bringing it up, and they’re bringing up things that won’t happen.”

Consider who was doing the bringing up. The warning did not come from activists, academics, or a congressional commission. It came from the industry itself, from the people who build the AI systems and whose fortunes depend entirely on building them faster. The day before, Anthropic CEO Dario Amodei published an essay asking that the pace of capability advance be slowed and committing his own company to embed outside evaluators with employee-level access. Sam Altman agreed and said OpenAI would do likewise. Elon Musk, a direct competitor who called Anthropic “misanthropic and evil” only in February, wrote three words: “Dario is right.”

Three rival laboratories in the most valuable commercial race on earth asked to be restrained, against their own immediate interest. Amodei did so while his company prepares what is expected to be the largest public offering in history, at a valuation of two trillion dollars or more. Tobacco executives never did this. Neither did the oil majors, the chemical companies, or the banks.

Nor is there the slightest indication that Trump wanted there to be any real assessment. A minimally competent leader who does not understand the technology but grasped that three competing chief executives had just testified against their own balance sheets would have asked someone for more information. Instead, Trump’s reply came within hours, on a golf course, framed entirely as a contest with China. This is not merely stupidity, which is Trump’s misfortune and therefore our lot. It is a fundamental incompetence that places us all in grave danger.

And then there is Johnson. 

Tech CEOs call for AI regulation, but critics warn against trusting Silicon Valley oligarchs. Lawmakers urged to listen to independent experts.

Several Big Tech CEOs over the weekend called for a slowdown in the development of artificial intelligence, but some advocates are warning that these Silicon Valley oligarchs are not to be trusted.

Evan Greer, director of digital rights group Fight for the Future, on Monday dismissed the recent statements made by Anthropic CEO Dario Amodei, X CEO Elon Musk, and OpenAI CEO Sam Altman calling for more guardrails to be placed on AI development.

There's lots of reasons to believe the "hyperscaler" model of AI can never be profitable, and not just because of its gigantic expenditures and negative unit economics (the companies lose money with every new customer and every new use, and they lose more money with each generation of their products):

https://pluralistic.net/2025/09/27/econopocalypse/#subprime-intelligence

The industry strenuously denies this, of course. They insist that they are only days away from turning their balance sheets right side up. All they have to do is fix those unit economics, then they can make back the cost of producing their models by selling access to them. The problem is that the evidence for those improving unit economics is weak, while the evidence that they're faking their finances is very strong:

https://www.wheresyoured.at/exclusive-openai-financials/

Same goes for the claims that these companies are already profitable. Dig into those claims and you'll learn they depend on a new, special meaning of "profitable" that does not match the generally accepted accounting procedures (GAAP) definition, which is to say, these companies are claiming that they are so cool that their profitability can only be measured using a novel, secret form of mathematics:

https://futurism.com/future-society/anthropic-claude-profit-ai-safety-development-finances

This is the same wheeze that Softbank tried with Wework. Speaking in my capacity as an author of internationally bestselling technothrillers about accounting fraud, I can tell you that it was accounting fraud then, and it's accounting fraud now:

https://www.ndtv.com/world-news/how-wework-went-from-being-47-billion-start-up-to-bankrupt-in-3-years-4552627

Twenty-five years since 9/11, tech and political shifts have increasingly pitted mass surveillance against constitutional rights.

One of the many legacies of the terrorist attacks of Sept. 11 is the government-wide shift from targeted surveillance—such as individual wiretaps or pen register/trap and trace orders—to mass surveillance techniques—such as tapping into the internet backbone or mass collection of telephone or internet metadata. The legal and technical architecture of modern mass surveillance, initially framed as a necessary defense against terrorist threats, has grown far beyond that justification and national security in general.

Mass surveillance is now a routine tool used by law enforcement. Immigration and Customs Enforcement (ICE) uses it in immigration actions and against people exercising their First Amendment rights to protest. It’s also increasingly part of private security systems, such as facial recognition at venues such as Madison Square Garden and networked Flock license plate capture systems on roads and in parking lots.

The interrelation between private and governmental mass surveillance is worth examining.

I found out about Meta’s Muse Image while scrolling on my Instagram account. A local journalist I follow was discussing the feature, explaining to her followers step by step how to opt out through their settings. I immediately went to my own settings, only to realize that because my account is private, the new feature did not affect my personal page. While my private status on Instagram shielded me, millions of other Instagram users were not so lucky. For users with a public profile, their likeness was automatically included without their consent. The eventual response highlighted a pattern we’ve seen too many times: Deploy first, ask forgiveness later.

Released in early July 2026, Muse Image is an artificial intelligence feature developed by Meta’s new AI division, Meta Superintelligence Labs. Meta originally launched the image generation feature on Instagram, which opted in all users who have public accounts and allowed users to generate images from other people’s likenesses without consent. The auto opt-in design was alarming for users of the platform. Within days, intense public backlash forced Meta to pause the feature after critics cited privacy, consent, and copyright concerns, especially because people with public accounts had to manually alter their settings to opt out of having their likeness used.

This is common for Big Tech. TikTok recently removed its experimental AI meme remixer feature after creator outrage. The feature, turned on by default, allowed anyone to take a still image of a creator’s face from a video, manipulate it with AI, and post it into the comment section.

In industry rhetoric, a radical intentionalist stance dominates any interpretation of model behavior, writes Eryk Salvaggio.

This way of describing large language models reflects what Daniel Dennett calls the intentional stance. It asks whether treating an object, such as an LLM, as something with beliefs is useful for predicting what it does. Dennett also describes a design stance, in which predictions are drawn from knowledge of the purpose of the system: not what did it think it was doing, but what was it designed to be doing?

Last week, The New Yorker’s Joshua Rothman asked what kind of stance we should use to explain these models. I’d like us to ask: who benefits when we encode one lens into policy?

Last week, I sat down for a chat with Bill Abston, Executive Director of the Kansas Office of Broadband Development, even though he has no reason to, he apologizes to me. “I finally can give you the time you deserve.”

Well, I am not sure of what I deserve, but the gift I received in speaking with Abston was 45-minutes of absolute enthusiasm and maybe even joy.

Abston took the reins of the Kansas broadband office in March of 2025, less than three months prior to Secretary Lutnick’s revised BEAD guidance and I’m taken aback at just how ‘glass half full’ he is. He explains that unlike other state broadband directors, “I don't have the two, three, four years of frustrations that a lot of state directors have. I come from building or designing this stuff and I'm ready to go.”

He tells me that his staff of eight, supplemented by consultants, is being asked to manage a $1.1 billion portfolio. Funding in nine programs total more than 200 projects, including the state’s BAG Broadband Acceleration Grant (BAG) that, when all is said and done, will direct $85M of state funds to broadband connectivity. Five years into a ten-year program, BAG has connected 12,000 homes. Also, more than $80M in ARPA was invested to connect ~23.5K homes.

Because he’s been heads-down focused, it’s been hard for many across the nation (including myself) to get to know Abston. He opens up right away to tell me that “I'm always optimistic, it’s just in my DNA. I've only been in public service eighteen months so I'm resistant to this ongoing debate whether or not we're really going to do it [connect everyone] or not. In Kansas, we're going to execute.”

There is a lot not to like about the way some data centers are being proposed.

Communities are being asked to accommodate projects of extraordinary scale, sometimes without convincing answers to fairly basic questions. What will this do to electricity prices? Where will the power come from? How much water will it use? What about noise, emissions and land? Who pays for new infrastructure? And, after the construction crews leave, what exactly does the community get in return?

Those are reasonable questions. In fact, developers should expect them.

But there is something missing from much of the debate. While we argue about whether we want more data centers, almost everything we are doing as a society suggests that we do.

Not necessarily the buildings themselves, of course. What we want is what happens inside them.

All of that has to happen somewhere.

The International Energy Agency expects global electricity consumption from data centers to more than double by 2030, reaching roughly 945 terawatt-hours. AI is expected to be the largest driver of that increase. In the United States, data centers could account for nearly half of the growth in electricity demand through the end of the decade.

So there is a contradiction developing in the public discussion.

We are demanding considerably more from the digital world while becoming increasingly uncomfortable with building the physical infrastructure required to provide it.

That is the data center paradox.

Data centers may be the perfect encapsulation of an economy built to benefit the already rich. After construction started on a vast facility in Vineland, New Jersey, residents didn’t just get mad; they got organized. 

In 2021, Andrew and Stevi Groetsch started building a modern ranch-style home across from a peaceful stretch of farms in Vineland, New Jersey. One night a few years later, in September 2025, Andrew was reading their daughter a bedtime story when he noticed the windows in her room vibrating with a loud, mechanical hum. He heard the same “wailing” sound while walking their floppy-eared Hungarian pointer at dawn. “We’d just spent a bunch of money to build a house,” Andrew told me in the family’s spacious, tree-lined backyard. “We pay taxes—a lot of taxes. We were pretty mad at that point. Still are.”

The source of the noise was New Jersey’s largest data center, about a mile down the road from the Groetsches’ home. The facility is owned and operated by a company called DataOne for the Netherlands-based “neocloud” provider Nebius, which is responsible for sourcing, installing, and managing graphics processing units (i.e., specialized microchips) made by Nvidia.

According to the terms of a five-year deal worth up to $19.4 billion, Nebius will deliver that computing capacity to Microsoft, whose internal teams will reportedly use it to create large language models and a consumer AI assistant. Giving Microsoft everything it wants will take at least 300 megawatts of power—nearly double the electrical generation capacity of the entire city of Vineland.

Vinelanders aren’t happy about it. They’re now ensnared in a bitter nationwide battle pitting a core engine of the U.S. economy against most of the people who live here.

​He is the personification of a time when the evolution of capitalism institutionalizes class divisions, universally degrades morality, and threatens global extinction.

Donald Trump is an instrument of this age, a force propelling capitalism and the global economy into a critical stage of their historical development. Since the beginning of village life some 10,000 years ago, commerce among villages and rural economies intensified as cities emerged and over centuries forged trade networks across regions into a global system of commerce. At the same time, from the beginning of markets, traders, merchants, industries, and governments engaged in the struggle for control of local, regional, and inter-regional markets.

Today this struggle continues, but now with devastating consequences: the concentration of wealth on a global scale; the vast productive power and technological innovation; the multiplying flow of market-related information, and the explosive potential of artificial intelligence. Each of these factors is channeling private and public resources to elites, and spurring worldwide trade wars, technological dependency, and military confrontations that threaten humanity’s survival.

President Trump dismisses calls for AI guardrails, claiming his intellect is enough. Concerns grow as industry insiders warn of AI dangers.

President Donald Trump on Monday slapped down the idea of putting guardrails on artificial intelligence, despite increased warnings from industry insiders and outside experts about the technology’s potential dangers to humanity.

In a Truth Social post, Trump suggested that his own intellect was singlehandedly capable of regulating AI, a technology so complicated that even its own creators have acknowledged difficulties in understanding it.