Notices by Miguel Afonso Caetano (remixtures@tldr.nettime.org)

It's happening: "Open Weights" LLMs are becoming way better and the gap between them and proprietary models is fastly closing :)
https://artificialanalysis.ai/#artificial-analysis-intelligence-index-by-open-weights-proprietary

#AI #GenerativeAI #LLMs #OpenWeights #China

"Take a moment to think before you dive in. That’s the best advice for Google Photos users, as the company confirms its latest update can scan all your photos to “use actual images of you and your loved ones” in AI image generation. That means Gemini seeing who you know and what you do. You likely have tens or hundreds of thousands of photos. They’re all exposed if you update.

We’re talking Personal Intelligence, Google’s latest AI upgrade path which lets users opt-in to connecting Google apps to Gemini. Why search for a doctor’s appointment when Google has access to all your calendar events. Why search for a party invite when it reads all your emails. And why search for a specific photo of you and your loved ones to create an image, when it sees all your photos.

This is the latest iteration in the ongoing battle between convenience and privacy playing out on our phones and computers. “Previously, to get a result that felt truly personal, you had to write long, detailed descriptions and manually upload a reference photo just to give Gemini the right context.” Not any more, Google says. Its AI can scan everything to form its own views of you and everyone you know."

https://www.forbes.com/sites/zakdoffman/2026/04/20/google-starts-scanning-all-your-photos-as-new-update-goes-live/

#AI #GenerativeAI #Google #Gemini #Privacy #DataProtection

RT @BenjaminNorton
About 10% of Argentina's electricity comes from nuclear power plants. But libertarian President Javier Milei wants to privatize the state-owned nuclear energy company.

So Milei appointed a 23-year-old internet libertarian activist, Ezequiel Acuña, to run the state nuclear company.

Acuña has ZERO experience in nuclear energy. He graduated high school in 2020 and started studying political science in college, but dropped out of the program. He never formally studied science, or even management.

Acuña is also being paid a starting salary that is 10 times larger than the average salary in the private sector, which he and Milei worship, as libertarians.

Acuña has never worked in nuclear energy or anything related. His only "qualification" is that he tweets libertarian propaganda in support of Milei. And the Milei regime thinks that is enough.

This is actually existing libertarianism.

https://x.com/BenjaminNorton/status/2047485764423106793

"In Guyang-ri, a farming village of 70 households about 90 minutes south-east of Seoul, people gather for communal free lunches six days a week. The meals are funded by the village’s one-megawatt solar installation, which generates roughly 10m won ($6,800) in net profit each month.

“Residents eat lunch together every day, so we see each other’s faces, talk together,” says Jeon Joo-young, the village chief. “Bonds and solidarity between residents become much stronger. Life becomes more enjoyable.”

The shift has been dramatic. Before the solar project launched in 2022, the village of about 130 people had no restaurant, no easy way to move around, and little communal infrastructure. Now solar revenue pays for meals, a village “happiness bus” for elderly people, a table-tennis facility and cultural activities.

The village deliberately chose to spend solar income on welfare rather than individual dividends, a decision Jeon says residents made themselves rather than being persuaded."

https://www.theguardian.com/world/2026/apr/16/south-korea-solar-power-renewables-revolution

#SouthKorea #SolarEnergy #Renewables #Solar #SolarPower

"Immigration and Customs Enforcement (ICE) just inked a $12.2 million contract for an artificial intelligence tool that claims to map out immigrants’ daily routines, habits, and real-time location and categorize them as potential threats, per procurement records reviewed by the Lever.

Dubbed “Project SAFE HAVEN,” the product is advertised by defense vendor Edge Ops LLC as a “question-based AI interface” that uses “persistent passive data collection” to map “patterns of life,” a surveillance tactic that ICE says will identify the “habitual locations, routes, and behavioral patterns” of its targets.

Additional features of the technology described in procurement documents include real-time location tracking and analysis that will categorize individuals and groups as affiliated with ostensible criminal organizations, such as gangs or cartels. That includes building “target profiles” that track individuals’ activity by linking data obtained from Wi-Fi network connections and mobile smart devices, such as cell phones and smartwatches.

A promotional blurb for the tool on Edge Ops LLC’s website claims that Project SAFE HAVEN “transforms the way we identify, locate, and map illegal migrants.”"

https://jacobin.com/2026/04/ice-contract-ai-surveillance-immigrants/

#USA #Trump #Authoritarianism #ICE #AI #Surveillance #Privacy

"In dozens of interviews with current and former government officials and advisers, technologists and MPs – most of whom asked not to be named, in order to speak freely – I have been told about a quiet handing over of control in the frameworks of advice, intelligence and decision-making that underlie every government decision. This is not just a simple software upgrade. Large language models (LLMs), the software behind AI program such as ChatGPT, are built to produce answers that will be accepted by users – not to calculate, but to convince. This highly persuasive software, built primarily overseas, is being handed an unknown amount of political power.

In almost every interview conducted for this piece, I asked whether it was paranoid to suggest that the wholesale adoption of AI by our government, public services and wider economy is handing power to models built in the US and China. Even the most optimistic AI advocates agreed it was a reasonable argument. At a technology conference last year, I spoke to a person who had been involved at the highest level in the government’s use of AI. I asked if it worried them that foundational models could reflect the politics of the people who control them – people who have very different political ideas to our elected leaders. My concerns were not brushed off. This person told me about a power struggle between the engineers building AI models, the plutocrats who own them and the politicians who seek to control them. Far from the noise of the public debate, a battle is being fought that could have lasting implications for our politics. “Make no mistake,” this person told me. “This is a war.”"

https://www.newstatesman.com/technology/2026/04/the-silent-coup

#UK #AI #BigTech #AIGovernance #LLMs #Chatbots #USA #China

"If critical infrastructure really is at risk, as Grieco suggests, then you would hope the US government is paying attention. (And right on cue, here’s a story from today about Iran successfully hacking US water and energy utilities.)

Awkwardly, though, the US government attempted to declare Anthropic a supply chain risk after it refused to modify its contract with the Pentagon to permit mass domestic surveillance and fully autonomous weapons. A judge has blocked that designation from taking effect while the case is litigated.

Anthropic told me that before launching Project Glasswing, it briefed senior US government officials about Mythos’ capabilities, both offensive and defensive. That includes the Cybersecurity and Infrastructure Security Agency and the Center for AI Standards and Innovation, which works with the industry to test new models and evaluate them for security risks.

The company told me it has signaled to the government that it is available to help the government with evaluating Mythos. But it’s not clear the government is taking Anthropic up on the offer.

A functioning government would take a strong interest in what Anthropic is up to here, if only out of self-preservation. We simply don’t know whether Project Glasswing will be enough to protect critical systems from being breached — and for how long."

https://www.platformer.news/anthropic-mythos-cybersecurity-risk-experts/

#CyberSecurity #USA #Anthropic #Claude #ClaudeMythos #ProjectGlasswing

RT @HedgieMarkets
🦔An AI agent using Claude autonomously developed two working exploits for a critical FreeBSD kernel vulnerability in roughly four hours of compute time, moving from vulnerability disclosure to a functional root shell attack without human assistance. FreeBSD powers Netflix's content delivery, PlayStation's operating system, and WhatsApp's infrastructure. The same researcher has since used the same pipeline to identify 500 additional high-severity vulnerabilities across various codebases. Developing a single kernel exploit of this complexity previously required weeks of work from elite security specialists.

My Take
The number worth understanding here is not four hours. It's 60 days. Industry surveys consistently show the median time for organizations to patch critical vulnerabilities in enterprise environments is around 60 days. An AI can now develop a working exploit within hours of a vulnerability being disclosed publicly. That window between patch availability and weaponized exploit has effectively collapsed to near zero for any organization that isn't patching immediately.

Kernel exploits have historically been scarce and expensive precisely because building them required deep expertise and significant time. Nation states paid enormous sums for them because they were rare. AI is compressing that cost curve the same way precision munitions moved from superpower monopoly to near-commodity over decades, except this is happening in months. The 500 vulnerabilities already in the pipeline from one researcher using one model suggests this is not an isolated result. Every organization still treating security as a periodic audit rather than a continuous process is now operating on a fundamentally outdated threat model.

Hedgie🤗

https://www.forbes.com/sites/amirhusain/2026/04/01/ai-just-hacked-one-of-the-worlds-most-secure-operating-systems/

RT @HedgieMarkets
🦔Google's plan to partly power a Texas data center through a 933-megawatt natural gas plant has been confirmed after Cleanview researchers uncovered the permit application. The plant would emit 4.5 million tons of carbon dioxide annually, more than the entire city of San Francisco. Google pledged carbon-free operations by 2030 in 2020, reported a 48% rise in emissions by 2024, and by 2025 had quietly reframed its climate commitments as climate moonshots, a term it uses for speculative projects that may or may not happen. Meta, Amazon, and Microsoft are making the same pivot.

My Take
The climate commitments these companies made were always going to last exactly as long as they were cheaper than the alternative. Google, Meta, Amazon, and Microsoft all pledged net-zero goals during a period when renewables were becoming cost competitive and climate leadership was good PR. AI changed the economics. The demand is too large, the timelines too short, and gas is available now. So the commitments are being quietly retired or reframed.

Google's head of advanced energy was asked directly last week how natural gas fits with the company's clean energy goals. His answer was we don't have anything to say on that. These were always profit-driven businesses. The climate commitments were real when they were also good business. They are becoming inconvenient and so they are going away.

Hedgie🤗

https://www.theguardian.com/technology/2026/apr/02/google-ai-datacenter

“The amount of untaxed wealth hidden offshore by the richest 0.1 percent exceeds the entire wealth of the poorest half of humanity (4.1 billion people), according to a new Oxfam analysis published on April 2, 2026, ahead of the 10th anniversary of the Panama Papers.

The findings show that, a decade later, the super-rich continue to exploit offshore systems to evade taxes and conceal assets, highlighting the urgent need for coordinated international action to tax extreme wealth and end the use of tax havens.

An amount of $3.55 trillion in untaxed wealth was stashed offshore in tax havens and unreported accounts in 2024, according to Oxfam’s estimates. “This sum exceeds the GDP of France and is more than twice the combined GDP of the world’s 44 least developed countries,” a statement by Oxfam noted.“

https://www.downtoearth.org.in/governance/untaxed-wealth-hidden-offshore-by-richest-01-surpasses-entire-wealth-of-the-poorest-half-of-humanity-oxfam

#TaxHavens #Offshores #TaxEvasion #Inequality

“The U.S. — and to some extent China — now has a potentially insurmountable lead in owning the world’s foundational AI models. That doesn’t mean other countries can’t develop their own robust AI ecosystems with fine-tuned technology built on top of those models. It does mean, however, that countries that do so will be increasingly dependent on a small number of American and Chinese firms. That leaves them vulnerable to shifting geopolitical winds and the risk that these companies might one day swallow their global competitors whole.

Even as global leaders and entrepreneurs outside the West scramble for some measure of self-determination by rushing to build their own “sovereign AI” ecosystems from scratch, their fate may be sealed.

“What we’re seeing [is] this kind of grandstanding bluster, like, ‘We can compete. We can build our own AI startup ecosystem,’ which doesn’t feel like it’s fully calling out the elephants in the room,” Kak said.

Just over three years on from the public release of ChatGPT, nearly every data point available about our AI era tells a startling story of geographic resource concentration.”

https://restofworld.org/2026/us-ai-investment-global-funding-gap/

#AI #USA #VentureCapital #PoliticalEconomy #BigTech

"When Undersecretary of Defense Elbridge Colby met Philippine President Ferdinand Marcos Jr. in New York this month, he said there would be “no change” to the U.S.-Philippine defense alliance, according to Jose Manuel “Babe” Romualdez, the Philippine ambassador to the U.S. “We were told there’s no immediate concern,” Romualdez said. “But we may have to review that depending on how long this Iran crisis continues.”

Some analysts say the joint U.S.-Israeli attacks on Iran, along with the U.S. capture of Venezuelan strongman leader Nicolás Maduro, have been a show of American military might that could deter China from attempting to forcibly retake Taiwan or expand its claims over contested territory in the South China and East China Seas. That deterrence only works, however, if U.S. capabilities are readily available in Asia, said Chen, the Taiwanese legislator.

“Military assets cannot be deployed at two places at the same time … Even though United States is a very resourceful country, there is a limit,” he said from his office in Taipei.

Chen said he is worried that if the U.S. sends troops into Iran, it could be the start of a complicated, drawn-out occupation like in Iraq or Afghanistan. Another concern is Taiwan’s dependence on Middle Eastern fuel, which has been choked off by Iran. If the conflict extends another month, Chen said, Taiwan’s leaders will be wholly absorbed by an energy crisis, with little time to focus on China’s military buildup or its increasing efforts to infiltrate Taiwanese society."

https://www.washingtonpost.com/world/2026/03/28/us-iran-war-japan-oil-prices/

#Asia #Taiwan #Philippines #China #USA #Trump #Iran #Imperialism

"A month into the war in the Middle East, an unlikely shortage of an irreplaceable gas is looming over the global economy.
Helium is a gas that is odorless, colorless and lighter than air. It is also indispensable to manufacturing the computer chips that power artificial intelligence, an important driver of U.S. markets and economic growth.

A byproduct of natural gas processing, helium is produced mainly in the United States and Qatar. When output in Qatar was halted this month, it cut off roughly a third of the global supply. The outlook worsened last week after Iran struck Qatar’s largest liquefied natural gas facility, damaging helium production lines that could take years to rebuild.

Without helium, leading chip makers, including Taiwan Semiconductor Manufacturing Company and South Korea’s Samsung Electronics and SK Hynix, could struggle to keep production lines running, with cascading effects for semiconductor-powered devices from Apple’s iPhones to Nvidia’s A.I. servers.

Helium may be best known for keeping balloons afloat, but its industrial uses are far more consequential. As the coldest liquid on earth, it cools superconducting magnets in M.R.I. machines. A shortage could ripple far beyond chip making, affecting everything from scientific research to space travel.

Semiconductor companies rely on helium at multiple stages of chip manufacturing. As intricate machines etch tiny circuits onto thin wafers of silicon, helium cools them from below to maintain the right temperature. After the wafers are washed with chemicals, helium is used to flush out toxic residue."

https://www.nytimes.com/2026/03/27/business/helium-chips-iran-war.html

#Iran #USA #Trump #Helium #Chips #AI #MiddleEast #NaturalGas

"For more than a decade, the Reuters Institute Digital News Report has documented fundamental shifts in how young people (defined in this report as those aged 18–24) interact with and think about news during a period of significant technological, media, and political transformation. As ‘social natives’, this demographic is moving away from traditional news media like television, print, and even news websites, gravitating instead towards a social-first and audiovisual-heavy media diet, where news is one type of content consumed among many.

While much has been said about the perceived lack of news engagement among younger people, our research also documents a greater sense of alienation among this segment of the public, some of whom find traditional news irrelevant, difficult to understand, or unfairly biased against their demographic. Mismatches between journalistic output and the expectations of young audiences highlight the need for newsrooms to examine both the question of how to reach young people where they are and, equally important, how to do so with news they find relevant, engaging, and ultimately worth their attention. Meeting the needs of this segment is crucial, not just for the current stability of the journalism industry, but also for the future of democratic societies as young individuals transition through adulthood (Røsok-Dahl and Ihlebæk 2024).

In this report we bring together evidence from over a decade of Reuters Institute research to shed light on young audiences today. Understanding generational shifts is vital for the financial sustainability of the news industry, which depends on a pipeline of younger consumers who will keep coming back to news. It also matters for the democratic health of our societies, which requires individual citizens, including young people, to be informed and collectively share a basic understanding of the world."

https://reutersinstitute.politics.ox.ac.uk/understanding-young-news-audiences-time-rapid-change

#Media #News #YoungAudiences #Journalism #MediaLiteracy

"Judging by the metrics of conventional conflict, Iran is not faring well against the United States and Israel. Its adversaries are destroying crucial targets in Iran, killing its commanders and degrading its military assets. But these are the wrong measures for assessing Iran’s position in the war. The right measure is not even an assessment of whether Iran is absorbing punishment well—which it is. The question that will matter when the fighting ends is whether Tehran is achieving its strategic objectives. And on that count, Iran is winning.

This outcome is not accidental. Tehran has been preparing for this war for nearly four decades, since the new revolutionary government faced its first major military test in the Iran-Iraq War, which lasted from 1980 to 1988. And it is now executing a strategy that has managed to neutralize key U.S. and Israeli air defense batteries, severely damage U.S. military bases in the Persian Gulf, inflict substantial economic pain, and drive a wedge between the United States and its Gulf allies. The Iranian regime, in other words, is not just surviving the U.S. and Israeli bombardment. The serious economic and political problems it is creating for its adversaries are, on a strategic level, giving Iran the upper hand."

https://www.foreignaffairs.com/iran/irans-long-game#

#Iran #USA #Israel #Trump #Militarism #War

"Can I convince AI to write useful content with minimal input from me? No. The input required is substantial; it's just a different kind of input than writing from scratch. Instead of staring at a blank page, I'm interrogating drafts, verifying claims, reframing angles, and restructuring articles. It's editing rather than writing, but it's not less work.

Can it provide a novel editorial take? Sometimes. The models produce serviceable takes that, with significant editorial shaping, become useful articles. But the novel connections, the thematic throughlines, the "this matters because of what we published last week" insights all come from me.

Can it produce accurate articles? Not without heavy verification. Every single article required factual corrections. Some were minor (wrong pipeline phase names, singular vs. plural). Some were serious (fabricated implementation details, wrong OWASP rankings, misleading CVE framing). The pipeline's verification step catches a small number of them. The rest require an editor who reads critically and clicks every link.

Can I edit out the AI writing tics? Mostly. The automated copy-editing pass handles the obvious markers. The subtler ones take manual work, and I'm still learning which tics I'm missing in my own editing passes.

I still think, for my needs, the pipeline is a useful tool for editorial content production. It's just not the "AI writes the blog" story that the stats at the top of this article might suggest. It's more like "AI produces a structured first draft that an experienced editor spends 90 minutes turning into something publishable." Whether that's worth it depends on what you value."

https://dacharycarey.com/2026/03/26/drafting-editorial-content-with-ai/

#AI #GenerativeAI #LLMs #TechnicalWriting #Writing #Blogging #Blogs

MUST READ ARTICLE:

"Opinion remains divided on exactly what the second Trump administration represents. Its arbitrary resort to tariffs and flagrant disregard for international norms has prompted many to condemn the supposed sabotage of Pax Americana—what Canadian Prime Minister Mark Carney described as a decisive “rupture” with the post-war multilateral system. Others, citing the administration’s foreign policy on the Middle East and its obedience to financial speculators, see as much continuity as change. Yet regardless of their emphasis, most commentators invoke the standard of a “rules-based international order” against which to measure the current state of global instability and insecurity.

The reality, though, is that such “ruptures” have been a recurrent feature of global governance for decades. If we are to fully understand Trump, and imagine how we might replace the world he symbolizes with a more just and stable one, we must recognize the extent to which the previous period of supposed order has, contra Carney, been a succession of US-led experiments in hegemonic destruction and reinvention. Only then will the contours of a more equitable international settlement begin to reveal themselves."

https://www.phenomenalworld.org/analysis/rupture-what-rupture/

#USA #Imperialism #PoliticalEconomy #Hegemony #Neocolonialism #Geopolitics #InternationalLaw

"Amazon’s ecommerce business has summoned a large group of engineers to a meeting on Tuesday for a “deep dive” into a spate of outages, including incidents tied to the use of AI coding tools.
The online retail giant said there had been a “trend of incidents” in recent months, characterised by a “high blast radius” and “Gen-AI assisted changes” among other factors, according to a briefing note for the meeting seen by the FT.

Under “contributing factors” the note included “novel GenAI usage for which best practices and safeguards are not yet fully established”.

“Folks, as you likely know, the availability of the site and related infrastructure has not been good recently,” Dave Treadwell, a senior vice-president at the group, told employees in an email, also seen by the FT.
(...)
Junior and mid-level engineers will now require more senior engineers to sign off any AI-assisted changes, Treadwell added.
Amazon said the review of website availability was “part of normal business” and it aims for continual improvement.

“TWiST is our regular weekly operations meeting with a specific group of retail technology leaders and teams where we review operational performance across our store,” the company said.
Separately, the company’s cloud computing arm — Amazon Web Services — has suffered at least two incidents linked to the use of AI coding assistants, which the company has been actively rolling out to its staff.

AWS suffered a 13-hour interruption to a cost calculator used by customers in mid-December after engineers allowed the group’s Kiro AI coding tool to make certain changes, and the AI tool opted to “delete and recreate the environment”, the FT previously reported."

https://www.ft.com/content/7cab4ec7-4712-4137-b602-119a44f771de

#Amazon #AI #GenerativeAI #AWS #VibeCoding #Programming #SoftwareDevelopment

"Last week Amodei, in explaining Anthropic’s position on Pentagon contracts, emphasized the company’s overall commitment to national security. He wrote, “I believe deeply in the existential importance of using AI to defend the United States and other democracies, and to defeat our autocratic adversaries.” If Amodei genuinely believes that the US military is devoted to addressing actual “existential” threats to the US, he’s too naive to be entrusted with anything as important as running a big AI company.

Obviously, this indictment applies about equally to OpenAI’s Sam Altman (who gladly swooped in and snatched the Pentagon largesse that Amodei will now be denied) and to Google’s Sundar Pichai and Demis Hassabis and to xAI’s Elon Musk. All the big AI companies are putting their tools at the disposal of the Pentagon to use as it sees fit.

All of these men, if pressed to justify this, would no doubt recite something about the importance of keeping America secure. Maybe somebody should ask them to name a recent US military intervention that had that effect."

https://www.nonzero.org/p/iran-and-the-immorality-of-openai

#USA #Trump #Pentagon #DoD #Iran #AI #GenerativeAI #Anthropic #OpenAI #Google

"With a hammer, harm is external and obvious, like a bent nail, a bruised thumb. In this cognitive environment, harm can be real and insidiously subtle. Once an LLM has offered a line of reasoning or even a clever turn of phrase, it becomes part of your mental terrain. You don’t just receive information; you adopt it as a starting point for your next thought. And that's what makes AI something other than just a tool...

Certainly, this is powerful, even exhilarating. The environment can change your cognitive perspective and trajectory. But it also carries a hidden cost or even accumulates a type of debt. It can smooth away the very frictions that make human thought generative. Confusion, hesitation, and error are not accidents of cognition—they are the mechanisms by which we refine our understanding. They are what turn information into meaning."

https://www.psychologytoday.com/us/blog/the-digital-self/202509/maybe-ai-was-never-a-tool

#AI #GenerativeAI #CriticalThinking #Automation