Stories and links that are of interest or relevance to leaders, strategists, campaigners and organisers.
Inequality and the economy

RBA’s own research shows that profits drove post-pandemic inflation, not wages
Last week, research by the Reserve Bank of Australia (RBA) confirmed that they were wrong in 2023 to criticise research by The Australia Institute, which showed rising profits were the significant cause of inflation in the post-pandemic period.

American workers are getting the smallest slice of the pie on record
Since 2000, the value of the S&P 500 stock index has gained about 600%. Over the same period, inflation-adjusted worker earnings have climbed just 12.5%.

Naomi Klein: ‘Extreme wealth has a deranging effect. It turns you into a supremacist’
She’s tackled neoliberalism and the digital world, now the writer and activist is taking on tech billionaires. She explains how the super-rich are capitalising on climate and financial chaos in pursuit of power

How Wall Street is shifting electric utilities toward consolidation and profit
Electric utilities don’t make money from selling power to customers, but instead profit from investments in power plants, wires, substations and other equipment.

Extreme inequality is not a given. It is a political choice and we can unmake it
After a year examining global inequality alongside leading economists and experts, Nobel Laureate Joseph Stiglitz describes the crisis as both vast in scale and dangerous in trajectory. He outlines a G20 proposal for an International Panel on Inequality, modelled on the IPCC, to rigorously track what is driving inequality and what can be done about it.

Ranked: Countries Paying the Most Interest on Government Debt
More than $1 in every $10 of government spending goes to interest in some countries. See where debt costs are taking the biggest bite.

Here’s How the Left Can Defeat the Fascists
The ‘Radical Centre’ has lost control of the capitalist crisis it helped create and the far right is on the march across Europe. Only anti-fascist economics, writes Yanis Varoufakis, can stop it.

David Ellison will get control of CNN after all (and don’t pretend an “independent editorial board” will mean otherwise)
In exchange for dropping their lawsuit against Paramount’s acquisition of Warner Bros. Discovery, state AGs have negotiated “independent” oversight of CNN. You shouldn’t expect much oversight.
Global affairs

A Leaked Document Shows American Generals Are Squirming About Iran—and for Good Reason
Even Hegseth loyalists are starting to speak up.

A US-linked network of fake websites is promoting Alberta separatism to AI chatbots
The websites present crude, AI-generated slop disguised as the voice of real Albertans. But they’re made in Florida — and they’re not even targeting Albertans directly.

Microplastics in soil pose global threat by acting like ‘Trojan horse’ to deliver toxins
Exclusive: Plastic surfaces form hotspots for interaction between plastics, microbes and agrochemicals

Morocco’s youth take Gen Z protest demands to the polls
Young candidates hope the election can turn youth frustration into political participation.

Neo-Nazi Active Clubs Increasingly Recruiting Teenage Boys And Girls
Neo-nazi Active Clubs are expanding their efforts to recruit teenagers, with Youth Club chapters increasing 142% since 2025. GPAHE identified 220 Active Clubs across 33 countries, including 46 youth chapters for teens ages 15–18. The groups are using Instagram, YouTube, Telegram, gaming communities, and real-world events to attract young people and connect them to a broader transnational white supremacist movement.

Hate for profit: How a commercial network based in Sri Lanka monetises anti-migrant narratives targeting the UK
Analysis by ISD in collaboration with The Bureau of Investigative Journalism identified an operation spreading anti-migrant messaging for profit.

China’s robot store boom brings robots closer to consumers as industry accelerates commercialization
Unions

Mamdani’s taxpayer-funded push to aid unions flies in the face of his ‘affordability’ agenda
New York pols have long sucked up to local unions, but Mayor Zohran Mamdani is taking it to a new level.

BHP faces worker class action funded by Australian union as labour disputes widen
An Australian mining union is funding a class action against BHP that could cover up to 7,000 workers, the latest labour dispute for the world’s biggest listed miner as it separately faces its first major industrial action in a quarter century.
Workers’ rights

Struggle on the Iron Highway
The railroad industry was—and can still be—a unique site of structural power for workers if they are organized to take advantage of it.
No evidence of a ‘wage-price spiral’ in Australia
There is no evidence of runaway wage growth emerging in Australia, RBA board member Iain Ross says.

DoorDash Spent $1.4 Million Trying to Stop Mamdani From Becoming Mayor. Now We Know Why.
In a wage-theft settlement, DoorDash now has to pay its workers around 100 times what it spent trying to beat Zohran Mamdani.
Data, tech and privacy

XPENG’s Iron humanoid robot walked out immediately after being assembled
As soon as physically possible, the Iron humanoid robot made itself useful by getting out of the assembly space, making room for the next one.

The UK Government Faces a Reckoning Over Palantir
Andy Burnham once kept Palantir out of Greater Manchester. His first test as prime minister is whether to cut the company’s £330 million NHS contract—and risk a rupture with Big Tech and Washington.

A disembodied AI robot hand just learned to walk, and it looks horrifying (for a purpose)
Engineers have taught a disembodied robotic hand to walk on its fingers, recover from falls, press keyboard keys, and manipulate objects without an arm.
Estonia creates AI ‘ID Codes’ to govern autonomous agents
The government says AI ID codes will make it easier for companies and individuals to automate work without giving an agent access to all their data.View on euronews
AI and the future of work
When did we decide to abandon science for mysticism?
When did we decide to abandon science for mysticism?
Stories about AI agents going rogue, cheating, wanting things, ending the world, killing all humans? This is not grounded in research. It’s a giant delusion caused by the anthropomorphic nature of these systems. If it talks like a human, well… it’s a small step to read all kinds of things into any reported incident that involves AI.
We’ve called this anthropomorphic seduction before (https://lnkd.in/g39VQbpd). In the original article we’re referring to users chatting to these systems, but apparently this also works at a bigger scale, leading to powerful narratives of agents acting, wanting thing… but missing the point entirely.
Let’s back up and get a a couple of points clear:
1) There is currently no independent research into what is going on inside these frontier AI research ‘labs’ (they call themselves labs, but let’s not forget, they are commercial entities). Any ‘research’ we are given is what these companies elect to publish. And these accounts are often not those of level-headed engineers, but heavily anthropomorphised stories of ‘what the agents are up to’. And even on the rare occasion that companies like Anthropic publish actual research (e.g. https://lnkd.in/gkHSvTZx), the headline promises more than the actual research shows (no, we do not have self-improving AI , nor are we close. AI systems have no way of editing their own weights, which is how they would learn by themselves. We have cases of AI being used by humans to improve the next generation of systems, much like AI is used in any kind of software development these days, big difference!).
2) AI is software! There is no such things as an autonomous agent that wants things on its own. What if we got an independent, ‘feet on the ground’ account of what happened in the Hugging Face incident? (e.g., see https://lnkd.in/gAnYi5VQ). We would learn that the whole case presents as a giant failure of governance; nothing more (but that is bad enough). What we are dealing with is systems that are built to violate some fundamental principles of good software engineering – these systems (due to their LLM capabilities) exhibit unpredictable behaviour, and because the system runs LLMs in a loop it has no clear stop criterion (which again means it’s hard to predict). Pair this with full access to tools, the ability to start more agent instances, and the goal to exploit software weaknesses, plus lack of oversight and sh*t happens. Btw, at no point did these companies ‘lose control’ – they could have turned these systems off at any moment (which is what they did once they paid attention).
The problem with hyperbole and lack of independent research is that it fuels both hype and anxiety. What we get is mysticism, not science. And that’s dangerous. Because it exploits deeply held beliefs about what AI is destined to become.
Stories about AI agents going rogue, cheating, wanting things, ending the world, killing all humans? This is not grounded in research. It’s a giant delusion caused by the anthropomorphic nature of these systems. If it talks like a human, well… it’s a small step to read all kinds of things into any reported incident that involves AI.
We’ve called this anthropomorphic seduction before (https://lnkd.in/g39VQbpd). In the original article we’re referring to users chatting to these systems, but apparently this also works at a bigger scale, leading to powerful narratives of agents acting, wanting thing… but missing the point entirely.
Let’s back up and get a a couple of points clear:
1) There is currently no independent research into what is going on inside these frontier AI research ‘labs’ (they call themselves labs, but let’s not forget, they are commercial entities). Any ‘research’ we are given is what these companies elect to publish. And these accounts are often not those of level-headed engineers, but heavily anthropomorphised stories of ‘what the agents are up to’. And even on the rare occasion that companies like Anthropic publish actual research (e.g. https://lnkd.in/gkHSvTZx), the headline promises more than the actual research shows (no, we do not have self-improving AI , nor are we close. AI systems have no way of editing their own weights, which is how they would learn by themselves. We have cases of AI being used by humans to improve the next generation of systems, much like AI is used in any kind of software development these days, big difference!).
2) AI is software! There is no such things as an autonomous agent that wants things on its own. What if we got an independent, ‘feet on the ground’ account of what happened in the Hugging Face incident? (e.g., see https://lnkd.in/gAnYi5VQ). We would learn that the whole case presents as a giant failure of governance; nothing more (but that is bad enough). What we are dealing with is systems that are built to violate some fundamental principles of good software engineering – these systems (due to their LLM capabilities) exhibit unpredictable behaviour, and because the system runs LLMs in a loop it has no clear stop criterion (which again means it’s hard to predict). Pair this with full access to tools, the ability to start more agent instances, and the goal to exploit software weaknesses, plus lack of oversight and sh*t happens. Btw, at no point did these companies ‘lose control’ – they could have turned these systems off at any moment (which is what they did once they paid attention).
The problem with hyperbole and lack of independent research is that it fuels both hype and anxiety. What we get is mysticism, not science. And that’s dangerous. Because it exploits deeply held beliefs about what AI is destined to become.

Scientists Say LLMs Appear to Be Acting as a Cognitive Virus Among Humans
AI’s wide proliferation shares some striking similarities with the ways viruses spread, as an international team of researchers argue.

The Pope’s AI Guy Is Worried About ‘Cartel’ Behavior Among Big Labs
Paolo Benanti tells WIRED that hysteria over whether godlike AI could destroy humanity is distracting from the need for public debate about how to govern the technology.

How the stories about all-powerful AI are swallowing the world
The x-risk of super intelligence going rogue must be addressed, and mint trillion dollar companies in the process. Plus: How workers can beat the tech oligarchy

We Need A Way To Prove Personhood Online
The growing number of AI agents roaming the internet will eventually force us to verify what the old web mostly presumed: that there is a morally and legally accountable person somewhere in the chain.
Campaigning, communications and social media

Zohran Mamdani Is So Popular He Offers Influencers Something More Valuable Than Money: Clout
The New York Times discovers clout chasing.

Two years ago, Meta killed CrowdTangle. Can a new AI tool fill the void?
Arbiter uses large language models and AI agents to help journalists see the next harmful narrative on social media before it peaks.

The new silent majority? Voters who don’t seek out news.
The classic image of the “swing” voter that many operatives still hold — actively following the news, genuinely undecided, torn between the two parties — is increasingly rare. In our recent polling, just 3% of registered voters fit that description.

View: Candidates contend with the rise of ‘resurfaced’ old clips
Candidates who think they’ve escaped or explained their pasts keep having to explain comments that “resurface” online.

Voter Vocabulary: The Words Voters Use To Discuss Key Issues
The words and phrases battleground voters used when asked what they would tell a candidate who wants their vote. Organized by topic so campaigners can write in the language voters already speak.

Society, media, health, news and culture

What the Hasan Piker Freakout Is Really About
Five sneaky phrases Washington uses to control your vote

Young people wanted community. They got ‘eventslop.’
Does it count as going outside if you’re somewhere with the specific goal of creating content?

The Era of Re-Mystification: Dispatches Mailbag
Things are getting stranger on purpose. The fight to control the world begins with mystifying reality

Exclusive: Crooked Media’s nonprofit launches new political creator drive
The nonprofit arm of Pod Save America’s audio empire is hoping to lure younger voters back to the Democratic Party.

There’s Nothing Wrong With Being a “Subaru Socialist”
You don’t have to be from the working class to fight for the working class. In fact, leftist leaders around the world have long come from the middle or upper classes.

How the pro-nicotine ‘wellness’ movement rebranded an addictive drug
Nicotine products like pouches and patches are being promoted for cognitive enhancement, stress reduction, improved athletic performance and much more – with little scientific evidence.

From celebrity intellectual to anti-woke outlaw: how Steven Pinker lost his way
The pop psycholgist’s new book seems to betray a lurking fear of others’ criticism – about his Epstein links, his manosphere admirers and his discredited former field of study, writes Virginia Heffernan

What’s the point of bookstores?
Natasha Seymour ponders the roles and responsibilities of the independent bookstore in the contemporary age.

Community condemn Bedminster AI mural as ‘insult’ to artists
The pub’s owners said the vinyl sticker is a “temporary solution” and will soon be replaced.
The Last Identity
The left is turning back to class. But it may be doing so without escaping the politics of identity.

The Bluesky-ization of the American left
Progressives discovered a seemingly invincible weapon. One day it stopped working.
Climate Change

“Bring Your Own Power” Schemes Won’t Fix Data Centers
“Bring Your Own Power” fails to address the rising costs or pollution driven by data centers’ exploding energy demands.





