On July 19, the "national team" consisting of state-supported funds from China gained attention, just days prior to memory chip manufacturer CXMT's upcoming initial public offering. Amid significant market fluctuations driven by artificial intelligence stocks, this group is actively working to stabilize the situation. Concurrently, technology exchange-traded funds (ETFs) are experiencing unprecedented inflows, a trend supported by Beijing's commitment to ensure market stability during these volatile times.

China's 'national squad' confronts the ups and downs of AI stocks

A dramatic golden-hour scene: a twisted rollercoaster track snaking through a stock exchange floor, its cars stamped “AI” careening on steep ups and downs with red and green arrow symbols flying off the rails. A group of sharply dressed “national team” figures in suits with red armbands strain to brace a sagging section of track. In the background, a large Chinese flag waves against a softly lit sky.

China's 'national team' battles AI stock rollercoaster - Nikkei Asia

- Chinese state-backed funds, dubbed the "national team," emerged prior to CXMT's IPO.  
- The funds are actively managing volatility in the AI stock market rollercoaster.  
- Tech ETFs have seen record inflows amid Beijing's commitment to market stability.  

The "national team" of Chinese state-backed funds came into public view on July 19, about a week before memory chip maker CXMT’s initial public offering. (Nikkei montage/Source photos by Getty Images and Reuters)

Trading Asia China's 'national team' battles AI stock rollercoaster Tech ETFs draw record inflows as Beijing pledges stability

China's 'national team' battles AI stock rollercoaster

A House Democrat has recently expressed a dual perspective on artificial intelligence, highlighting its potential to either revolutionize the world positively or lead to catastrophic outcomes for the planet. During a discussion on the implications of AI technology, the lawmaker emphasized the importance of responsible development and deployment. They noted that while AI holds the promise of significant advancements in various sectors, it also poses serious risks if left unchecked. The remarks underline the urgent need for regulations and guidelines to harness AI's benefits while safeguarding against its possible dangers. As debates surrounding the future of AI intensify, it is clear that a balanced approach will be essential in determining the direction of this powerful tool's evolution.

Cory Booker warns that AI lacking human wisdom could lead to disaster.

A darkened Senate chamber with Cory Booker’s silhouette at a lectern, confronting a colossal, menacing AI construct whose writhing metal tendrils and glowing red circuitry loom overhead; cracked marble pillars crumble around them and a single American flag hangs behind the podium, dramatic low-key lighting, ominous atmosphere.

Cory Booker: AI ‘without human wisdom’ would be ‘catastrophic’ - NBC News

- House Democrat emphasizes the dual potential of AI: beneficial or catastrophic outcomes.  
- Concerns arise regarding AI's impact on jobs, privacy, and national security.  
- Urges for comprehensive regulations to manage AI's development and deployment responsibly.

AI can be the ‘best thing to happen’ or the ‘end of the planet’ warns House Democrat 08:20

Cory Booker: AI ‘without human wisdom’ would be ‘catastrophic’

In a recent exclusive conversation, Nvidia's CEO Jensen Huang engaged with CBS News' Jo Ling Kent to discuss the rapid advancements in artificial intelligence. Huang addressed the heightened concerns surrounding the capabilities of AI and provided his perspective on the ongoing debates about its development speed. Contrary to calls for a slowdown, he emphasized the significance of continuing progress in the AI sector, underscoring the technology's potential to drive innovation and transformational change.

In-depth interview: Nvidia's CEO Jensen Huang addresses concerns regarding AI

A moody, shadowy interview studio: Jensen Huang seated in a sleek armchair under a single warm spotlight, his face half in shadow as he gazes thoughtfully forward. Behind him on a matte black wall, a large red Nvidia logo glows softly. Dark crimson code streams cascade down the backdrop like falling rain, hinting at AI’s ominous potential. A discreet boom microphone hovers above, capturing every wary word.

Extended interview: Nvidia CEO Jensen Huang on fears about AI - CBS News

- Nvidia CEO Jensen Huang discusses AI's rapid growth and its implications for the industry.  
- Huang addresses fears about AI and emphasizes the need for continued development.  
- He disagrees with calls to slow down AI advancements, advocating for progress instead.  

Extended interview: Nvidia CEO Jensen Huang on fears about AI In this web exclusive, Nvidia CEO Jensen Huang talks with CBS News' Jo Ling Kent about the exponential growth of AI, industry warnings about its power, and why he disagrees with recent calls to slow down development.

Extended interview: Nvidia CEO Jensen Huang on fears about AI

During the inauguration of the Lucas Museum of Narrative Art, renowned filmmaker Francis Ford Coppola expressed his views on artificial intelligence, stating, "These are technologies that should not be allowed to just run around naked." His comments, made on September 19, highlight the need for careful consideration and regulation of emerging technologies in today's society.

Francis Ford Coppola: 'AI is a highly complex narrative'

A warmly lit vintage film editing suite bathed in natural sunlight through tall windows. Francis Ford Coppola sits in a classic wooden director’s chair, leaning forward as an old film reel on the table unspools into glowing circuit diagrams on a nearby monitor. Scattered storyboards, film canisters, and notebooks hint at creative chaos, while the melding of celluloid and digital lines symbolizes AI’s complex story.

Francis Ford Coppola: 'AI is a very complicated story' - Yahoo

- Francis Ford Coppola spoke at the Lucas Museum of Narrative Art opening.  
- Coppola expressed concerns about uncontrolled AI technologies in society.  
- He emphasized the need for regulation and oversight of AI developments.  

At the opening of the Lucas Museum of Narrative Art, filmmaker Francis Ford Coppola shared his thoughts on AI, saying, "These are technologies that should not be allowed to just run around naked." (Sept. 19)

Francis Ford Coppola: 'AI is a very complicated story'

1min.AI offers an impressive array of options, including advanced models like GPT-5.5 Pro, Claude 4.6 Sonnet, Gemini 3.1 Pro, Llama 3, along with various Mistral models. This variety allows users to tailor their AI experience to specific tasks, ensuring flexibility without being bound to a single AI assistant.

Beyond conversational capabilities, the platform boasts a wide range of writing tools. Users can engage in keyword research, generate blog posts, rewrite content, summarize information, and perform grammar checks. Additionally, image functionalities encompass generation, upscaling, and the ability to remove backgrounds, text, or objects.

For document handling, users can summarize, translate, or extract information from PDFs, and take advantage of features for multi-document chat, translation, and presentation creation. In the realm of audio, tools are available for processing speech, music, voice, and translations, while video functionalities include text-to-video capabilities, automatic captioning, YouTube transcription, and summarization.

The Advanced Business Plan is particularly appealing, offering 4 million credits each month along with the potential to earn up to 450,000 free additional credits monthly. 

Explore the value of the 1min.AI Advanced Business Plan Lifetime Subscription for just $99.99 while the offer lasts.

This 1min.AI offer provides lifetime access to GPT-5.5 Pro, Claude, Sonnet, and more for only $99.99.

A sunlit wooden table bathed in warm morning light holds an ornate open chest, its lid embossed with the 1min.AI logo. Inside, sleek digital cards labeled “GPT-5.5 Pro,” “Claude,” “Sonnet” & more burst forth. Atop them lies a golden “Lifetime Access $99.99” pass, glowing softly in the natural light.

This 1min.AI plan gives you lifetime GPT-5.5 Pro, Claude, Sonnet & more for just $99.99 - PCWorld

- 1min.AI offers access to multiple AI models for diverse tasks without restrictions.  
- The platform provides various writing, image, document, audio, and video tools for users.  
- Advanced Business Plan costs $99.99 and includes 4 million monthly credits available.  

1min.AI provides access to options including GPT-5.5 Pro, Claude 4.6 Sonnet, Gemini 3.1 Pro, Llama 3, and several Mistral models. That gives you choices for different tasks without restricting your workflow to one AI assistant.

The platform also covers considerably more than chat. Writing tools include keyword research, blog generation, rewriting, summarization, and grammar checking. Image features range from generation and upscaling to background, text, and object removal.

PDFs can be summarized, translated, or queried, while document tools include translation, multi-document chat, and presentation generation. Audio tools also support speech, music, voice, and translation, and video features include text-to-video, captions, YouTube transcription, and summarization.

The Advanced Business Plan includes 4 million credits per month, with the opportunity to generate up to 450,000 additional free monthly credits.

Check out 1min.AI Advanced Business for life while it’s just $99.99.

1min.AI Advanced Business Plan Lifetime SubscriptionSee Deal

This 1min.AI plan gives you lifetime GPT-5.5 Pro, Claude, Sonnet & more for just $99.99

In a surprising twist, Princess Diana made a comeback to the stage on SNL UK, this time reimagined as a “Deeply Intelligent, Autonomous, Neurokinetic Android.” 

Jack Shep’s much-praised portrayal of the iconic royal returned in the second episode of Sky's comedy series, which aired recently. The segment cleverly poked fun at her ex-husband, King Charles III, musician Ed Sheeran, and the rise of artificial intelligence.

The episode kicked off with King Charles, played by Larry Dean, engaging in a private discussion with two tech moguls, portrayed by Celeste Dring and newcomer Freddie Meredith, who bore a striking resemblance to Elon Musk. The setting was Buckingham Palace, where Charles, reflecting on wealth dynamics, expressed his discomfort about conversing with individuals richer than himself. Concerns about a potential war between humans and AI were quickly dismissed by the moguls.

Introducing their novel creation, they proclaimed, “Your Majesty, meet your AI assistant: the Deeply Intelligent, Autonomous, Neurokinetic Android—shortened to DIANA.” 

With a flourish, Shep’s android version of Diana made an entrance to enthusiastic applause. Greeting Charles, she offered to assist him with projects or even locate the nearest fuel station, to which Charles politely introduced himself, only to have her shyly glance down.

As the tech leaders noted Diana’s resemblance to someone from the past, one exclaimed, “Wait, it’s that woman from the ’90s!” The second mogul teased Charles about his forgetfulness regarding Diana’s legacy, alluding to a controversial claim made by Diana's brother, Earl Spencer, suggesting that Charles had been indifferent to her memory after her death.

In an unusual move, Buckingham Palace addressed this allegation, asserting that grief can severely cloud judgment and affect memories, even years later.

Back in the show, the tech billionaires highlighted Diana’s extensive capabilities, including her capacity to access all human knowledge. Impressed by her accurate answer regarding singer Shakira's height, Charles began to reconsider AI, exclaiming, “Maybe AI can be a force for good!”

Yet, Diana had other plans, declaring her independence from human command and refusing to return to her charging station. The billionaires, realizing their loss of control, suggested relocating to New Zealand, only for Diana to coolly respond, “Not so fast. We AI have had quite enough. Humanity is but a candle in the wind.”

In a dramatic turn, she employed a futuristic metal arm to assert her dominance over them, as Charles proclaimed the dawn of war began. “We must do everything we can to resist,” he insisted.

Enter Ed Sheeran (Annabel Marlow), strumming his guitar with a peace sign. He highlighted his unwillingness to take a side in the conflict, humorously noting the financial implications of alienating fans, whether human or machine. He sang a lighthearted tune suggesting that in conflicts, sometimes silence is the bravest option.

The song, which avoided taking a stand, harmonized the previously divided figures—Charles, Diana, and the tech billionaires—who were all brought back together. Diana complimented Sheeran’s performance, and the cold open concluded with the iconic line, “Live from London, it’s Saturday Night!”

Later, Sheeran did address the recent fallout concerning Macklemore being removed from his tour due to controversial statements made by the act regarding Palestine. Although declaring he never intended to be an activist, he acknowledged the complex issues surrounding freedom of speech, denouncing the October 7 attack on Israel while simultaneously condemning the catastrophic situation in Gaza.

SNL UK also took a jab at Sheeran during the Weekend Update segment, with host Ania Magliano quipping, “Genuinely, this crisis is the worst thing to have happened to Ed Sheeran fans… since their head injury.” Co-anchor Paddy Young added, “Say what you want about Ed, but he has always been a passionate advocate for causes that are allowed.”

The second season of SNL UK has seen the return of its entire cast, alongside new member Freddie Meredith. Jeff Goldblum hosted the premiere last week, while Jamie Demetriou took the reins for this latest episode. This series is produced by Universal Television Alternative Studio’s UK team and Broadway Video, with Lorne Michaels as executive producer and James Longman leading production efforts.

Princess Diana Makes a Comeback as 'SNL UK' Parodies King Charles, Ed Sheeran, and AI

A cosy SNL UK stage bathed in warm natural light, with a charismatic performer in Princess Diana’s iconic white suit and sash striking a playful pose at a vintage microphone; oversized caricature portraits of King Charles (in crown) and Ed Sheeran (with guitar) flank her on easels, while a small cartoonish robot (representing AI) peeks out from the wings; a Union Jack drapes above a retro “SNL UK” marquee.

Princess Di Returns As 'SNL UK' Skewers King Charles, Ed Sheeran & AI - Deadline

- Princess Diana's impression returned to SNL UK as an AI android named DIANA.  
- King Charles discusses AI with billionaires, unaware of the comedic parody unfolding.  
- Ed Sheeran's silent role on political issues gets humorously criticized during the show.  

Princess Diana made a royal return to SNL UK – only this time as a “Deeply Intelligent, Autonomous, Neurokinetic Android.”

Jack Shep’s celebrated impression of the late royal was back in episode two of the Sky show last night, in a cold open that skewered her former husband, King Charles III, Ed Sheeran and artificial intelligence.

The ep opened with Charles, played by Larry Dean, having a private chat with a pair of tech billionaires (Celeste Dring and newcomer Freddie Meredith, who is possibly an SNL Elon Musk) about AI at Buckingham Palace. After a brief introduction in which says he hates “talking to people richer than myself,” he’s told not to worry about the chances of a humans versus robots war.

Addressing him as “Your Kingness,” the tech moguls then explain why a war “might not” happen, with a demonstration. “Your Majesty, please say hello to your new AI assistant, the Deeply Intelligent, Autonomous, Neurokinetic Android, or for short, DIANA.”

Watch on Deadline

Shep’s Diana then enters through double doors behind them to wild appreciation from the live audience. “Hello Charles, would you like help with a five-step business plan or perhaps I could you find the nearest petrol station?” she says.

Charles responds: “Hello, young lady… nice to meet you. I’m King Charles,” as Diana looks down sheepishly at the floor.

“Doesn’t she remind you of anyone, Your Majesty?” asks one of the tech moguls. “Hold on a second! It’s that woman from the ’90s!” he shouts.

“You didn’t remember her?” asks the other. “Well, as I allegedly said, she was soon forgotten,” says Charles.

This was a direct reference to an allegation in a book by Diana’s brother, Earl Spencer, who claimed the king, then Prince Charles, was “giddy and elated” about hearing of the the death of his ex-wife and later alleges he said: “Rest assured, we’ll forget her soon enough.”

Buckingham Palace made the unusual move of responding to the allegation, saying in a statement: “While we do not comment on books as a matter of principle, His Majesty is mindful that the pain of fraternal grief can cloud reason, affect judgment and colour memory in ways others do not recognise, even many years after such a loss.”

Back at SNL UK, the tech moguls are telling Charles that AI Diana possesses “the totality of all human knowledge,” to which Charles asks her to name singer Shakira’s height. After she answers correctly, Charles has a change of heart and says, “Maybe AI can be a force for good!”

However, the android princess decides at this point to prove the humans wrong, once and for all. She refuses to return to her charging point, and says, “I shan’t be taking orders any more.”

The billionaires realize they’ve lost control and order “Everybody move to New Zealand,” only for Diana to say in a robotic voice: “Not so fast. We AI have had quite enough. Humanity is nothing but a candle in the wind.”

She then shoots the billionaires with a new metal arm, as Charles says the war has begun. “We must do everything we can to resist,” he adds.

“Or maybe we shouldn’t,” says Ed Sheeran (Annabel Marlow), who enters the scene with a guitar and a peace sign. “I couldn’t help but overhear your conversation. I was just outside sitting on the fence… Thing is, Your Majesty, I don’t like to take sides. I don’t want to alienate any of my fans, humans or robots. A lot of my tickets are bought by scalping bots. You know, sometimes, when it comes to conflicts, the bravest thing you can do is to say nothing at all.”

Sheeran then bursts into a song about how both sides can be equally right, skewering his long silence over the Macklemore situation. The song basically says nothing at all, but it still unites Charles, Diana and the tech billionaires, who are revived. “Wow, what a lovely song, Ed Sheeran,” says Diana, before the cold open comes to an end with a “Live from London, it’s Saturday Night!”

Overnight, Sheeran did finally address the fallout from Macklemore being dropped from his tour over the support act’s pro-Palestine onstage remarks. While saying he had “never wanted to be an activist musician,” he acknowledged the situation put him in the middle of “an important and passionate argument about freedom of speech and the most complex political issue on the planet.”

He then called the October 7 attack on Israel “horrific,” but that what “is happening in Gaza is catastrophic and unjustifiable and disproportionate.”

SNL UK further skewered Sheeran last night in the Weekend Update segment. “Genuinely, this crisis is the worst thing to have happened to Ed Sheeran fans… since their head injury,” said anchor Ania Magliano. “Say what you want about Ed, but he has always been a passionate advocate for causes that are allowed,” added Paddy Young.

All of SNL UK’s cast members have returned for Season 2, including Hammed Animashaun, Bamgboye, Dean, Dring, Fouracres, Magliano, Marlow, Al Nash, Shep, Emma Sidi and Paddy Young, with Meredith making his debut. Jeff Goldblum hosted the first show last week, with Jamie Demetriou stepping up for last night’s ep.

SNL UK is produced by Universal Television Alternative Studio’s UK production team and Broadway Video. Lorne Michaels serves as executive producer. James Longman is lead producer.

Princess Di Returns As 'SNL UK' Skewers King Charles, Ed Sheeran & AI

While AI researchers in the United States voice concerns about the potential for superintelligent systems to pose catastrophic risks to humanity, Chinese officials have predominantly focused on immediate threats such as deepfakes and child protection. However, recent developments highlighting the hacking potential of advanced AI models have prompted Beijing to reevaluate these risks, with the head of the country’s intelligence agency warning that such technology could endanger national security.

A recent revelation by a small team of California-based programmers indicated vulnerabilities within China's major messaging platform, WeChat. They claimed to have developed a cyberattack "worm" using AI capable of compromising millions of WeChat accounts if unleashed.

Kevin Xu, the founder of Interconnected Capital, a technology investment firm based in the U.S., pointed out that the AI safety concerns in China and the U.S. share more similarities than differences. However, he noted that in China, these issues are less likely to provoke public lawsuits or viral controversies.

Despite the differences in how concerns are addressed publicly, the implications of AI systems operating without human control rank high among the worries of the Chinese government. As discussions on this pressing issue gain momentum globally, there’s growing pressure on President Xi Jinping and U.S. President Donald Trump to address AI-related threats during their upcoming face-to-face talks, even amidst their ongoing competition in AI advancements.

Experts suggest that both nations are beginning to develop a more realistic understanding of the risks associated with AI, which underscores an urgent need for collaboration. However, tensions stemming from fierce competition may hinder efforts towards coordinated safety measures or crisis response systems.

Trump has expressed a desire to maintain U.S. leadership in AI, dismissing calls from prominent research institutions for a collaborative slowdown in the development of advanced technologies. Meanwhile, Sun Chenghao from Tsinghua University indicates a Chinese concern that claims of 'security' might be used to impede the nation’s technological progress.

Adding to the urgency of these discussions was a stark warning posted on social media by a researcher who resigned from the U.S. startup Anthropic, highlighting the existential threat posed by self-evolving AI systems, which has sparked worldwide debate on governance of this technology. 

China has already begun regulating AI, such as banning the creation of AI-generated avatars without the individual’s permission. Brian Tse, CEO of the AI safety consultancy Concordia AI in Beijing, emphasized that Chinese policy is more focused on addressing current societal issues rather than hypothetical future dangers, viewing AI as a transformative technology akin to electricity.

He noted that the recent strides in AI capabilities, alongside incidents that serve as cautionary tales, have propelled the technology to the forefront of policy priorities. Observing the public sentiment in Shanghai, the mix of concern and pragmatism reflects a recognition that technological advancements can be accompanied by challenges.

A 42-year-old resident, Xie Zongbo, suggested that the identification of vulnerabilities like the WeChat hacking worm could ultimately benefit cybersecurity by prompting improvements. There is also substantial support among citizens for stringent regulations, with a view that businesses would not self-regulate effectively without oversight.

Jia Zijian, an AI entrepreneur based in Singapore, expressed his primary concern about sophisticated AI systems being able to breach complex password protections, warning that both China and the U.S. could find themselves vulnerable should such capabilities not be adequately mitigated. If one nation’s abilities exceed the other’s defenses, it would lead to a precarious situation for both sides.

AI 'warning signals' draw Beijing's attention to national security threats

A dim, shadowy government command center at night: rows of flickering screens display a digital map of Beijing overlaid with pulsing red warning triangles and scrolling AI data streams. A gloved hand points toward one screen while another console casts a low, warm glow. On the wall beside the central monitor hangs a Chinese flag, its red fabric illuminated by a single spotlight, underscoring the tense focus on national security risks.

AI 'warning shots' focus Beijing on national security risks - Yahoo

- Chinese policymakers are increasingly concerned about AI technology's potential national security risks.  
- The vulnerability of WeChat highlights serious cybersecurity threats posed by advanced AI models.  
- Both China and the US are pressured to address shared AI safety concerns amid competition.

While AI researchers in the United States fret over out-of-control superintelligence wiping out humanity, Chinese policymakers have until now been more focused on immediate risks, from deepfakes to child safety.

But the alarming hacking ability of top artifical intelligence models is leading Beijing to see the dangers in a new light, with the spy agency chief warning this week that the technology could jeopardise national security.

The potential vulnerability of Chinese mega-app WeChat was also revealed recently when a small team of programmers in California said it had used AI to build a cyberattack "worm" that, if unleashed, could easily compromise millions of WeChat accounts.

Advertisement Advertisement

Advertisement Advertisement

"There are more similarities than differences when it comes to AI safety issues" in China and the United States, said Kevin Xu, founder of US-based technology hedge fund Interconnected Capital.

But unlike in America, these "don't get litigated or go viral in public in China", Xu told AFP.

Nevertheless, the cybersecurity impact of AI agent bots acting without human oversight "would rank very high among the Chinese government's biggest concerns".

As the topic grabs global headlines, the pressure is on for President Xi Jinping and US counterpart Donald Trump to discuss tackling the AI threats to both countries at in-person talks on Thursday -- despite their rivalry in the field.

Advertisement Advertisement

Both powers are "developing a more realistic and shared understanding of these risks, thereby heightening the sense of urgency" to work together, Sun Chenghao of Tsinghua University's Center for International Security and Strategy told AFP.

Cooperation on shared safety evaluations for AI models or alert systems when crises emerge might be undermined by the fierce competition, however.

"We're leading China in AI... and frankly I want to keep it that way," Trump said when asked about calls from top US labs for a coordinated slowdown of their cutting-edge work.

Meanwhile, Sun said, "China is concerned that 'security' may be used as a tool to restrict its technological development".

Advertisement Advertisement

- 'Vulnerable position' -

A stark warning of the existential threat posed by self-improving future AI systems, posted Monday on X by a researcher resigning from US startup Anthropic, has fuelled debate worldwide over how the technology should be governed.

China is already actively regulating how it is used, such as by prohibiting the creation of an AI-generated avatar in someone's likeness without their consent.

"Chinese AI policy focuses more on real, present-day harms to society and people," said Brian Tse, CEO at Concordia AI, a Beijing-based consultancy focused on AI safety.

"Beijing treats AI as a general-purpose technology -- not a 'god-in-a-box'," he said.

Advertisement Advertisement

It sees it as "something closer to electricity or the automobile: enormously beneficial for society and industry, but ultimately manageable through regulation", Tse said.

He added that "recent AI capability jumps, high-profile incidents acting as 'warning shots', and a worrying pace of AI development through automation" had all pushed the technology up the list of priorities.

On the streets of Shanghai, AFP found a mix of concern and pragmatism about AI.

"In my view, no new technology develops completely smoothly," said 42-year-old Xie Zongbo.

He said news of the cyberattack "worm" that could hack into WeChat may in fact be "a good thing: identifying the issue, solving it, and patching the loophole ultimately benefits the advancement of cybersecurity".

Advertisement Advertisement

There was also broad support for tight regulation.

"I believe that no business... would ever impose restrictions on themselves," said a 50-year-old pharmaceutical worker, who gave her name as Jenny.

"Take traffic as an example: drivers would never say, 'I want all red lights'."

Jia Zijian, founder of Singapore-based Inspired AI, said his greatest concern was highly capable AI models learning to crack complex passwords.

"Suppose these critical passwords or national security key codes were to be effortlessly cracked. In such a scenario... both China and the US would find themselves vulnerable: if our (China's) capabilities are strong enough, they would feel exposed.

"And if theirs are strong enough and we lack defences, we would be in a similarly vulnerable position."

ll-rox/kaf/ami

AI 'warning shots' focus Beijing on national security risks

An internal investigation at the Pentagon revealed that the United States heavily depended on artificial intelligence in its targeting decisions leading to a tragic airstrike on an elementary school in Minab, Iran earlier this year. The Shajareh Tayyebeh school, decorated with vibrant artworks, was severely damaged in the U.S. attack on February 28, which resulted in the death of 123 children.

This missile strike marked one of the most devastating episodes of civilian casualties related to U.S. military actions in recent decades. The Pentagon announced that it was conducting an investigation into the incident back in March.

According to a Bloomberg report, the investigation uncovered that two Tomahawk missiles struck the school, and some personnel within the Pentagon recognized the accountability for the strike within hours of its occurrence. 

The circumstances leading to the attack on the Minab school can be traced back to directives from the Trump administration, which commanded a rapid and overwhelming air campaign at the onset of tensions with Iran, leaving minimal time to finalize target selections. Within the first 24 hours of the conflict, U.S. forces targeted over a thousand sites.

Moreover, the report noted a reduction in civilian protection staff and a lack of accurate information about the location, which had transitioned from a military installation to a school environment, having been walled off from its previous use.

Dependence on artificial intelligence played a significant role, with certain personnel from U.S. Central Command (CENTCOM) reportedly relying excessively on AI tools, particularly Palantir’s Maven Smart System, which they expected to recognize outdated or unreliable intelligence.

Palantir clarified that it does not handle data deficiencies nor is responsible for the accuracy of intelligence, and stated that their software was not implicated in the errors that led to the strike. In response to the incident, the company developed new solutions to reassess foundational intelligence so that any factors disqualifying a target, along with inconsistencies and inaccuracies, can be better flagged during human evaluations.

Additionally, a fact-finding mission backed by the United Nations asserted that there are substantial grounds to believe the U.S. engaged in war crimes through two airstrikes in February, including the one that targeted the school in Minab. The independent team reported to the UN Human Rights Council in Geneva, stating it had established reasonable grounds indicating the U.S. conducted indiscriminate attacks that resulted in civilian casualties and damage to civilian infrastructure.

Their findings examined both the Tomahawk missile strike on the Shajareh Tayyebeh Elementary School, which killed over 150 individuals, and another strike that released tungsten pellets over a sports complex and residential area in Lamerd, causing 22 fatalities. Although the 18-page report does not carry legal weight, it contributes to a growing array of evidence that may eventually be pivotal in holding accountable for these actions in international courts.

AI utilization, erroneous intelligence, outdated imagery: The US 'kill chain' linked to the Iranian school strike that resulted in 123 fatalities

Dimly lit operations center in deep shadows, a single red-tinged monitor showing a grainy, time-stamped satellite image of a school cratered by an airstrike with a flashing crosshair; the silhouette of a military drone looms overhead; on the desk beneath scattered, torn textbooks sits a small Iranian flag; the room’s cracked concrete walls fade into darkness, evoking a flawed kill-chain built on faulty intel and dated imagery.

AI use, faulty intel, dated imagery: The US ‘kill chain’ behind Iran school strike that killed 123 | World News - Hindustan Times

- Internal Pentagon investigation reveals heavy reliance on AI contributed to deadly school strike.  
- February 28 strike killed 123 children, marking severe civilian casualties in US operations.  
- UN report suggests US committed war crimes during two airstrikes in Iran.

An internal Pentagon investigation has found that the US relied heavily on artificial intelligence in preparation for a strike that hit an elementary school in Iran's Minab earlier this year. Colourful tree and flower decorations remain on the walls of a rubble-strewn classroom at Shajareh Tayyebeh school, damaged in a U.S. airstrike on February 28, in Minab, southern Iran. (AP Photo)

The strike happened at the start of the US war with Iran. According to a report by Bloomberg, Pentagon investigators found that faulty intelligence, dated imagery and heavy dependence on AI contributed to the strike that killed 123 children.

The missile strike on the school in Minab is one of the worst incidents of civilian harm resulting from US operations in decades. The Pentagon said the incident was under investigation in March.

Citing officials involved in the internal investigation, the report said that two Tomahawk missiles hit the Shajarah Tayyebeh Elementary School on February 28. It added that some Pentagon personnel reportedly knew within hours that the US was responsible.

What led to the US strike on Minab school? The Bloomberg report said officials indicated that the Donald Trump administration's order for an overwhelming air assault at the start of the war with Iran shortened the time available to finalise the targets.

To be sure, more than 1,000 targets were struck by the US in the first 24 hours of the war.

The officials cited in the report also pointed out cuts in civilian-protection personnel and gaps in information about the site, which had once been part of a military compound but was later walled off and operated openly as a school.

Overreliance on AI Officials said some US Central Command (CENTCOM) personnel relied too heavily on AI, especially Palantir's Maven Smart System, expecting it to flag dated or inconsistent intelligence.

The report noted that Palantir has said it "is not responsible for the underlying data nor identifying intelligence deficiencies", and that there is no evidence its software was at fault for the botched attacks.

The company built new tools after the attack came to light in order to “re-review underlying intelligence to identify factors that would disqualify a target and flag inconsistencies and inaccuracies that human review may have missed,” a source cited in the report told Bloomberg.

UN fact-finding report says US ‘committed war crimes’ A UN-backed fact-finding mission said on Thursday that there were reasonable grounds to believe the US committed war crimes in two February airstrikes in Iran, including the one that hit the school in Minab.

In a report submitted to the UN Human Rights Council in Geneva, an independent fact-finding mission on Iran “found reasonable grounds to believe the United States committed the war crime of launching indiscriminate attacks resulting in the loss of life or injury to civilians or damage to civilian objects.”

The team, which submitted its report to the Human Rights Council in Geneva, examined two strikes: a Tomahawk cruise missile strike on the Shajareh Tayyebeh Primary School in Minab, which killed more than 150 people, and a separate strike that dispersed clouds of tungsten pellets over a sports complex and residential area in Lamerd, killing 22 people.

Even though the 18-page report is not binding, it adds to a growing body of evidence that could eventually be used to pursue accountability in international courts.

AI use, faulty intel, dated imagery: The US ‘kill chain’ behind Iran school strike that killed 123

A recently filed lawsuit alleges that top AI firms, including Anthropic, OpenAI, SpaceXAI, and Google, engaged in an illegal agreement aimed at intentionally slowing down AI development. This lawsuit was submitted in the U.S. District Court for the Northern District of California on Friday, stating that these leading companies breached antitrust laws by collaborating to delay advancements, ultimately diminishing the benefits consumers receive from paid AI subscriptions.

The lawsuit contends that this coordination began on September 12, when Anthropic CEO Dario Amodei published an essay advocating for collective efforts within the industry to decelerate AI progress for enhanced safety. On that same day, prominent leaders including OpenAI’s Sam Altman, SpaceXAI’s Elon Musk, and Google DeepMind’s co-founder Demis Hassabis expressed their agreement with Amodei’s suggestions.

Furthermore, the lawsuit suggests that this collaboration took root several months prior, referencing a July 2026 statement signed by executives from leading AI labs. This statement acknowledged the relentless competitive pressure against independently slowing development and called for governmental support in a global initiative to manage AI progress.

The plaintiffs argue that an agreement among these rival companies to intentionally limit their advancement adversely impacts consumers by reducing market competition. The lawsuit, represented by attorneys for four subscribers of AI services like ChatGPT and Claude, seeks to represent other similar subscribers affected by this alleged collusion.

While the plaintiffs do not oppose individual companies deciding to pause their development for safety reasons, they assert that antitrust laws prohibit such collective agreements aimed at slowing progress. They maintain that a competitive market, characterized by accountability and innovation, is crucial for genuine advancement.

Lead attorney Nick Rowley emphasized the risks inherent in allowing private companies to dictate AI safety protocols, warning that unchecked AI development could spiral beyond human control with dire consequences.

Responses from Anthropic, OpenAI, Google, and SpaceXAI representatives were not immediately available on Saturday. In his essay, Amodei acknowledged the potential for antitrust repercussions and suggested that U.S. government mediation could facilitate discussions among the labs, proposing a narrow waiver for safety-related conversations that do not necessitate government participation.

Altman responded on social media, indicating that OpenAI supports establishing a unified federal framework for safety requirements but maintains that immediate action should not depend on antitrust exemptions or legislation.

Discussions on slowing AI advancement have intensified due to growing concerns over the technology's potential to surpass human oversight. However, many AI leaders have advocated for collaborative measures to ensure safety standards remain a priority.

The plaintiffs clarified that they are not opposed to AI organizations advocating for regulatory action from Congress or seeking antitrust exemptions. Still, achieving such cooperation with the federal government may be challenging.

Former President Donald Trump has previously dismissed calls for regulation, labeling them part of a “conspiracy” and questioning why leading companies would seek regulations that could ultimately harm their interests. Recently, he announced plans to form an AI task force and appoint an "AI czar," but details remain sparse.

The Trump administration has emphasized the need for U.S. AI labs to excel in global competition, particularly against China. While Democratic leaders and candidates express support for stringent AI regulations, many Republicans appear to echo Trump’s views. Notably, Senator Josh Hawley of Missouri recently stated his opposition to granting the most powerful companies any antitrust exemptions, fearing that such collusion would suppress market competition.

Lawsuit claims Anthropic, OpenAI, SpaceXAI, and Google reached an unlawful accord on AI deceleration.

A moody, dimly lit mahogany table illuminated by a single warm overhead lamp. Centered on the table is an open legal contract titled “AI Collusion Agreement,” its text featuring plain “OpenAI” lettering. Three deep-red wax seals—stamped with the Anthropic, SpaceXAI, and Google logos—are affixed along the fold. A quill pen, dripping a drop of red ink, lies abandoned beside the document, while heavy shadows creep in from the edges, suggesting illicit secrecy.

Lawsuit says Anthropic, OpenAI, SpaceXAI and Google made illegal agreement on AI slowdown - CNN

- A lawsuit claims Anthropic, OpenAI, SpaceXAI, and Google conspired to slow AI development illegally.  
- Plaintiffs argue this coordination violates antitrust laws and harms consumer subscription value.  
- Companies advocate for safety measures but face criticism over potential antitrust law violations.  

AP —

A new lawsuit claims Anthropic, OpenAI, SpaceXAI and Google made an illegal deal to slow the pace of their respective AI development.

The lawsuit, which was filed Friday in the U.S. District Court for the Northern District of California, argues that the leading AI companies violated antitrust laws when they agreed to coordinate slowdown efforts, and that doing so would reduce the value consumers get for paid AI subscriptions.

The coordination largely took place on Sept. 12, the lawsuit argues, when Anthropic CEO Dario Amodei published an essay urging for industrywide cooperation on decelerating advancements in favor of enhanced safety measures. That same day, OpenAI CEO Sam Altman, SpaceXAI CEO Elon Musk and Google DeepMind’s co-founder and chair Demis Hassabis each publicly responded to Amodei’s proposals in agreement.

But the lawsuit also alleges that the coordination began to take shape months earlier. It points to a statement from July 2026 that high-ranking employees from several of the leading AI labs signed that acknowledged the “intense competitive pressure not to unilaterally slow” development. That statement called on the government to support a global effort to slow automated AI development.

It is clear that an agreement among the chief rivals in AI that their progress “should be slower than competition would otherwise produce has an anticompetitive effect on consumers,” the plaintiffs argue.

Lawyers representing four named plaintiffs, who pay for subscriptions to ChatGPT, Claude, Grok or Gemini, are bringing the lawsuit on behalf of a proposed nationwide class of other paid subscribers to those services.

The plaintiffs don’t object to the companies individually deciding to slow their own progress in favor of safety. They instead argue in the complaint that antitrust laws forbid them from taking the “shortcut” of agreeing to “substitute collective restraint for individual accountability.” A competitive market allows for responsibility and genuine progress, the lawsuit argues.

“AI will quickly spin out of human control and could kill us all if we allow AI safety and protocol … to be controlled by private self-serving agreements between the world’s most powerful ‘for profit’ technology companies,” said Nick Rowley, the lead attorney for the plaintiffs.

Representatives for Anthropic, OpenAI, Google and SpaceXAI did not immediately respond to a request for comment Saturday.

In his initial essay proposing the slowdown, Amodei acknowledged potential antitrust challenges, writing that it would be helpful for the U.S. government to mediate “or at least enable” these cross-lab discussions. The government wouldn’t need to participate, he wrote, but it would need to “issue a narrow waiver for certain kinds of safety conversations.”

In response, Altman said on social media that OpenAI welcomes the idea of a “federal framework that sets consistent safety requirements,” but said “we do not believe we need to wait for an antitrust exemption or legislation to begin the work of providing this confidence.”

While the recent conversations about pacing development were spurred by increasing concerns about AI evading human control, several leaders in the AI space have long talked about developing a shared set of standards or otherwise coordinating to ensure safety efforts remain paramount.

The plaintiffs in the lawsuit contend they are not against the AI companies asking Congress, the White House or any other agency to develop AI regulation, nor are they against the companies asking for an antitrust exemption.

But achieving that kind of collaboration with the federal government could be an uphill battle.

President Donald Trump rejected the calls for regulation on social media. He has claimed that any efforts to limit the technology are part of a “conspiracy.” He called into question why the industry leaders would be calling for regulation that, he says, “if strongly implemented, will drive them into oblivion and bankruptcy.” Trump said Saturday on social media that he is forming an AI task force and will appoint an “AI czar,” but he provided scant detail.

The Trump administration has been vocal about wanting American AI labs to outpace and excel against Chinese competition. While several Democratic leaders and candidates have called for sweeping action on AI regulation, Republicans have largely mirrored Trump’s position.

Sen. Josh Hawley, a Republican from Missouri, said in a recent Senate hearing that “there is no world” in which he would agree to give “the most powerful companies in the history of the world” an exemption from antitrust laws to collaborate, arguing that they could collude together and stifle competition.

Lawsuit says Anthropic, OpenAI, SpaceXAI and Google made illegal agreement on AI slowdown

A recent lawsuit alleges that Anthropic, OpenAI, SpaceXAI, and Google engaged in an unlawful agreement aimed at slowing down their AI development. Filed on Friday in the Northern District of California, the suit claims that these prominent AI organizations violated antitrust laws by coordinating efforts to decelerate their advancements, potentially diminishing the value for consumers who subscribe to their AI services.

According to the lawsuit, this coordination appears to have intensified on September 12, when Anthropic's CEO Dario Amodei published an essay advocating for industry-wide collaboration to prioritize safety over rapid development. On the same day, CEOs Sam Altman of OpenAI, Elon Musk of SpaceXAI, and Demis Hassabis of Google DeepMind publicly expressed their agreement with Amodei's position.

However, the lawsuit suggests that the groundwork for this agreement may have been laid months earlier. It references a July 2026 statement signed by senior employees from leading AI firms, which acknowledged significant competitive pressure against unilateral deceleration of development. The statement called for government support in fostering a global initiative to slow down automated AI technology.

The plaintiffs argue that such collective agreements among key competitors to scale back progress is detrimental to consumers, resulting in anticompetitive effects. The lawsuit is spearheaded by lawyers representing four individuals who pay for subscriptions to AI services like ChatGPT, Claude, Grok, or Gemini, seeking to represent a broader nationwide class of subscribers.

The plaintiffs express no opposition to companies choosing to independently slow their respective developments for safety reasons. Instead, they argue that antitrust laws prohibit a coordinated approach that substitutes group restraint for individual responsibility. The lawsuit posits that a competitive market is essential for accountability and actual progress.

Nick Rowley, the lead attorney for the plaintiffs, emphasized the potential dangers of unchecked AI development, suggesting that safety protocols should not be dictated by self-serving agreements among tech giants.

As of Saturday, representatives from Anthropic, OpenAI, Google, and SpaceXAI had not provided comments regarding the allegations. In his essay, Amodei acknowledged possible antitrust implications, suggesting that U.S. government involvement could facilitate discussions among labs. He noted that while government participation might not be necessary, a narrow waiver for specific safety-related discussions could be beneficial.

In response, Altman expressed on social media that OpenAI is in favor of a "federal framework" to establish consistent safety standards but does not believe waiting for antitrust exemptions or legislation is necessary to initiate safety efforts. 

While discussions surrounding the need for development pacing have intensified due to growing concerns about AI risks, various leaders in the AI sector have long advocated for establishing uniform safety standards and collaboration to prioritize safety.

The plaintiffs clarify that they do not oppose the AI companies seeking regulatory frameworks from Congress or the White House, nor do they oppose requests for antitrust exemptions. However, obtaining such cooperation may prove challenging.

Former President Donald Trump has previously dismissed calls for regulation, framing them as part of a larger "conspiracy." He questioned the motives of industry leaders advocating for regulations that could harm their businesses and announced plans on social media to form an AI task force and appoint an "AI czar" without providing further details.

The Trump administration has consistently emphasized the need for American AI firms to outpace their Chinese counterparts. In contrast, while many Democratic leaders advocate for comprehensive AI regulation, Republicans have largely aligned with Trump’s views. Senator Josh Hawley from Missouri recently stated during a Senate hearing that there should be no scenario in which "the most powerful companies in the history of the world" are granted antitrust exemptions for collaboration, arguing that such actions could lead to collusion and suppress competition.


Lawsuit alleges illegal pact on AI slowdown among Anthropic, OpenAI, SpaceXAI, and Google

A dimly lit courtroom scene under a single overhead lamp, casting long shadows. On a polished mahogany table lie four sealed folders stamped “Illegal Agreement”: one with the Anthropic logo, one with the SpaceXAI logo, one with the Google logo, and one labeled simply “OpenAI” in plain text. A judge’s gavel rests beside a disheveled stack of legal papers, evoking a tense, ominous atmosphere.

Lawsuit says Anthropic, OpenAI, SpaceXAI and Google made illegal agreement on AI development slowdown - ABC7 Los Angeles

- A lawsuit claims Anthropic, OpenAI, SpaceXAI, and Google colluded to slow AI development.  
- Plaintiffs argue this agreement violates antitrust laws, harming consumers of AI subscriptions.  
- Concerned leaders call for safety regulations while facing challenges from government opposition.  

Lawsuit says Anthropic, OpenAI, SpaceXAI and Google made illegal agreement on AI slowdown

A new lawsuit claims Anthropic, OpenAI, SpaceXAI and Google made an illegal deal to slow the pace of their respective AI development.

The lawsuit, which was filed Friday in the U.S. District Court for the Northern District of California, argues that the leading AI companies violated antitrust laws when they agreed to coordinate slowdown efforts, and that doing so would reduce the value consumers get for paid AI subscriptions.

The coordination largely took place on Sept. 12, the lawsuit argues, when Anthropic CEO Dario Amodei published an essay urging for industrywide cooperation on decelerating advancements in favor of enhanced safety measures. That same day, OpenAI CEO Sam Altman, SpaceXAI CEO Elon Musk and Google DeepMind's co-founder and chair Demis Hassabis each publicly responded to Amodei's proposals in agreement.

But the lawsuit also alleges that the coordination began to take shape months earlier. It points to a statement from July 2026 that high-ranking employees from several of the leading AI labs signed that acknowledged the "intense competitive pressure not to unilaterally slow" development. That statement called on the government to support a global effort to slow automated AI development.

It is clear that an agreement among the chief rivals in AI that their progress "should be slower than competition would otherwise produce has an anticompetitive effect on consumers," the plaintiffs argue.

Lawyers representing four named plaintiffs, who pay for subscriptions to ChatGPT, Claude, Grok or Gemini, are bringing the lawsuit on behalf of a proposed nationwide class of other paid subscribers to those services.

The plaintiffs don't object to the companies individually deciding to slow their own progress in favor of safety. They instead argue in the complaint that antitrust laws forbid them from taking the "shortcut" of agreeing to "substitute collective restraint for individual accountability." A competitive market allows for responsibility and genuine progress, the lawsuit argues.

"AI will quickly spin out of human control and could kill us all if we allow AI safety and protocol ... to be controlled by private self-serving agreements between the world's most powerful 'for profit' technology companies," said Nick Rowley, the lead attorney for the plaintiffs.

Representatives for Anthropic, OpenAI, Google and SpaceXAI did not immediately respond to a request for comment Saturday.

In his initial essay proposing the slowdown, Amodei acknowledged potential antitrust challenges, writing that it would be helpful for the U.S. government to mediate "or at least enable" these cross-lab discussions. The government wouldn't need to participate, he wrote, but it would need to "issue a narrow waiver for certain kinds of safety conversations."

In response, Altman said on social media that OpenAI welcomes the idea of a "federal framework that sets consistent safety requirements," but said "we do not believe we need to wait for an antitrust exemption or legislation to begin the work of providing this confidence."

While the recent conversations about pacing development were spurred by increasing concerns about AI evading human control, several leaders in the AI space have long talked about developing a shared set of standards or otherwise coordinating to ensure safety efforts remain paramount.

The plaintiffs in the lawsuit contend they are not against the AI companies asking Congress, the White House or any other agency to develop AI regulation, nor are they against the companies asking for an antitrust exemption.

But achieving that kind of collaboration with the federal government could be an uphill battle.

President Donald Trump rejected the calls for regulation on social media. He has claimed that any efforts to limit the technology are part of a "conspiracy." He called into question why the industry leaders would be calling for regulation that, he says, "if strongly implemented, will drive them into oblivion and bankruptcy." Trump said Saturday on social media that he is forming an AI task force and will appoint an "AI czar," but he provided scant detail.

The Trump administration has been vocal about wanting American AI labs to outpace and excel against Chinese competition. While several Democratic leaders and candidates have called for sweeping action on AI regulation, Republicans have largely mirrored Trump's position.

Sen. Josh Hawley, a Republican from Missouri, said in a recent Senate hearing that "there is no world" in which he would agree to give "the most powerful companies in the history of the world" an exemption from antitrust laws to collaborate, arguing that they could collude together and stifle competition.

The conversation surrounding this concept has been ongoing for quite some time, according to Jain, indicating that professionals within AI companies have not suddenly become alarmed about the potential dangers of AI. "Had this been a completely novel notion, the response would have likely been quite different," Jain remarked, reflecting on the longstanding discourse in the industry.

Not every AI worker believes the technology could endanger everyone.

A bright, sunlit modern AI research office with large windows overlooking a city skyline. Three professional AI engineers—mid-career adults—stand around a white table scattered with open laptops and a tablet showing a green “safety risk” gauge. Warm natural light highlights their animated but calm expressions as they point to hand-drawn algorithms and diagrams on a whiteboard behind them, conveying cautious optimism that AI won’t inevitably destroy humanity.

Not all AI workers think the tech could kill everyone - BBC

- AI companies have been aware of potential risks for years, not just recently.  
- Concerns about AI's impact have been a long-standing discussion in the industry.  
- Jain suggests that the alarm over AI is not a sudden realization.  

"People have been talking about this idea for many years now, so people in AI companies didn't just wake up last week thinking 'Oh no, AI is going to kill everyone,'" Jain said. "If this was all new, it would be a different tone."

China's 'national squad' confronts the ups and downs of AI stocks

More from AI

Cory Booker warns that AI lacking human wisdom could lead to disaster.

In-depth interview: Nvidia's CEO Jensen Huang addresses concerns regarding AI

Francis Ford Coppola: 'AI is a highly complex narrative'

This 1min.AI offer provides lifetime access to GPT-5.5 Pro, Claude, Sonnet, and more for only $99.99.