Technology

Second-quarter revenue jumped 93 percent but Palantir’s ties to Israel and role in military technology are controversial.

United States artificial intelligence and data analytics giant Palantir Technologies has reported “otherworldly” quarterly results, sending its shares more than 14 percent higher in after-hours trading, as its growth shows no signs of slowing despite mounting criticism over its close ties to the US and Israeli governments and concerns about its growing role in artificial intelligence and warfare.

Palantir Technologies reported a revenue of $1.94bn for the second quarter, up 93 percent from a year earlier, and raised its forecast annual revenue to between $8.15bn and $8.158bn, up from $7.65bn to $7.662bn earlier. It said strong demand from both commercial customers and government agencies drove the surge.

“This quarter was otherworldly: our US commercial revenue grew 149 percent year-over-year, our overall revenue grew 93 percent year-over-year,” Chief Executive Alex Karp said. “Demand for AI sovereignty has now been unleashed.”

In a letter to shareholders, Karp wrote: “Our business is compounding at a rate and scale that we have never before witnessed.”

Palantir holds multibillion-dollar contracts with US government agencies, including the US Army. Revenue from its US government business rose 90 percent year on year to $809m, even as the company has faced growing opposition over its role in President Donald Trump’s immigration crackdown, which critics say has resulted in unlawful deportations and killings.

Founded in 2003 by technology entrepreneurs including Karp and multi-billionaire Peter Thiel, Palantir opened its first office in Israel in 2015 and has since expanded its work with the Israeli military.

Following what Palantir described as a “strategic partnership” with Israel in January 2024, the company significantly expanded its operations supporting Israel’s military campaign in Gaza and operations in the occupied West Bank.

According to Open Intel, a platform tracking corporate involvement in the genocidal war on Gaza, Palantir has actively recruited former members of Israel’s elite Unit 8200 cyberintelligence division. The group says Palantir’s software integrates intercepted communications, satellite imagery and other datasets to help generate military targeting lists for Israeli forces.

In a statement to Al Jazeera earlier this year, Palantir UK reiterated the company’s support for Israel.

The company has also secured major contracts with the United Kingdom’s government. In January, the UK’s Ministry of Defence awarded Palantir a $323m (240-million-pound) contract. A separate $444m (330-million-pound) NHS contract awarded in November 2023 has also attracted criticism, with campaigners raising concerns about the handling of sensitive health data and the heavy redaction of contract documents.

Palantir has also faced scrutiny over its vision for the future of artificial intelligence. In The Technological Republic, a recent book co-authored by Karp and the company’s head of corporate affairs, Nicholas W Zamiska, the authors argue that technology companies have a responsibility to build advanced military AI capabilities. Critics have described the philosophy as a form of “techno-fascism”.

Al Jazeera has contacted Palantir for comment.

A Pacific island has become one of the biggest economic success stories of the year so far.

Taiwan has witnessed a dramatic boom in recent months driven by the mania for artificial intelligence (AI). Earlier this year, its stock exchange soared to become the fifth largest in the world based on market capitalisation, the value of its publicly traded shares, overtaking the United Kingdom, Canada and India.

Much of that upward momentum has been driven by AI and other technology exports highly sought after by the United States.

Last year, the US imported $201bn worth of goods from Taiwan, nearly double its rate from 2024, when it acquired $116bn in imports. In May, Taiwan eclipsed China to become the third-largest source of US imports, after Mexico and Canada.

Experts have described Taiwan’s market acceleration as a return to its status as a “tiger economy” — a term used to capture surging growth in East Asia. Much of the credit, they say, falls to its flourishing technology sector.

“Artificial intelligence helps explain the rising importance of Taiwan,” said Chad Bown, a senior fellow at the Peterson Institute for International Economics.

But critics warn that, while Taiwan’s market remains strong, factors like tumultuous international relations, as well as demographic concerns, could complicate the island’s long-term outlook.

“It seems to be a win-win for now,” said Reza Hasmath, an academic faculty adviser at The China Institute at the University of Alberta. “But Taiwan is just postponing a reality that’s not sustainable.”

An economic boom

Taiwan’s thriving export market helped boost its gross domestic product (GDP) to 8.63 percent in 2025.

That rocket-ship trajectory continued into the first quarter of this year, when the GDP saw an exhilarating 13.69 percent rise.

Government data released on Friday showed that the island is continuing that momentum, with its economy growing an impressive 12.92 percent in the second quarter of the year, which ended in June.

“The GDP growth is going like gangbusters,” said Dexter Tiff Roberts, nonresident senior fellow at the Atlantic Council’s Global China Hub.

Roberts expects the trend to be “long term”, as Taiwan produces about 90 percent of the advanced chips used to power leading AI models.

“That’s not going to go away. We know the world, and the US, needs this,” he added.

While the AI boom is a global phenomenon, the US has become a major market for such chips, with billions of dollars flowing into the industry each year.

US President Donald Trump, meanwhile, has pledged to bolster his country’s status as “the world leader in artificial intelligence”. His administration has claimed to attract more than $2.7 trillion in tech and AI investments since the start of his second term.

To secure US access to Taiwan’s cutting-edge semiconductor technology, the Trump administration signed an agreement under which Taiwan will invest $500bn in the US.

Half of that amount is expected to come in the form of direct investments by Taiwanese semiconductor and tech firms, including through the development of onshore tech manufacturing.

The rest is largely comprised of credit guarantees for additional investments from Taiwan in the US.

Under the agreement, Taiwanese firms would be allowed to import 2.5 times the capacity of their US factories, without fear of steep tariffs.

In a subsequent trade agreement, Taiwan agreed to reduce its tariffs on 99 percent of US exports.

Taiwan has also boosted its tech exports to the US through investments in nearby Mexico, with cross-border plants manufacturing inputs for data centres in Texas.

‘Unbalanced relationship’

But Hasmath, the faculty adviser at the University of Alberta, warns that there are troubling signs on the horizon for Taiwan-US relations.

Trump has long sought to eliminate trade deficits with US economic allies, and he has lashed out at countries that export more to the US than they import.

Hasmath pointed out that Taiwan is building a robust trade surplus with the US, close to $200bn and counting. That could spark a backlash.

“This is an unbalanced relationship and not conducive to Taiwan in the long term,” Hasmath warned.

Trump will not tolerate a hefty trade surplus for long, he added. Hasmath believes the US president will soon look to renegotiate his country’s deals with Taipei.

Roberts at the Atlantic Council, meanwhile, warned that Trump is “mercurial” — and with such a temperament comes “uncertainty”.

Then there’s the question of political upheaval in the US. Trump’s approval ratings are low, and he is ineligible under US law to run for a third term as president.

Demographic problems

While Taiwan’s economic boom is “very real” and “very obvious”, Roberts said there are clear vulnerabilities even on the domestic front.

Taiwan’s traditional export sectors like plastics and textiles are underperforming. Plus, Roberts pointed out that only a small fraction of the Taiwanese population is involved in the AI sector.

“A majority of the younger population is not in hi tech, so that’s a real problem,” he said.

While the booming stock market has sparked a “wealth effect” — those with rising portfolios feel richer and are more inclined to spend — that helps the wider population only to an extent.

With most of Taiwan’s employment concentrated outside of the AI sector, economists have warned that the island could develop what’s called a K-shaped economy, where the wealthy see growth, while the poorer segments of society stagnate or decline.

The chip industry employs up to 350,000 people at most, experts say.

Meanwhile, TSMC, Taiwan’s biggest chip company, makes up to 40 percent of the stock market and provides four percent of the island’s GDP growth. That lopsided proportion is “unsustainable”, according to Hasmath.

Plus, Taiwan has a rapidly ageing population, with roughly a fifth of its population over the age of 65.

The island also has other vulnerabilities. For example, it relies heavily on foreign imports of energy products, particularly oil, and has struggled with water scarcity.

Then, there’s the superpower next door: China. The government in Beijing considers Taiwan, a self-governing island, as its own territory, and it has taken aggressive measures to limit the island’s ability to establish diplomatic relations of its own.

That conflict has added fuel to the debate around Taiwan’s growth, with a spokesperson for the Chinese government reportedly saying the island’s growing proximity to the US tech sector will “drain Taiwan’s economic interests” and “hollow out” the country’s major industry.

Hasmath said that, if the AI boom backfires on Taiwan, all of that ultimately adds up to a “recipe for electoral change, a shift in government” in Taipei.

July 31 (UPI) — A federal judge on Friday turned down a request by Elon Musk’s artificial intelligence company, xAI, to block a Minnesota law banning nudification technology.

Minnesota in May became the first state in the nation to pass legislation banning the practice of using AI to digitally remove clothing from a pictured individual.

xAI, which develops the chatbot and image generator Grok, filed the suit over the ban Tuesday.

“The court respectfully denies the request for a temporary restraining order before tomorrow,” U.S. District Judge Donovan Frank ruled on Friday. “xAI filed the motion on July 29, 2026, nearly three months after the law was signed, and only three days before the law is set to take effect.

“Such a delay in bringing the action and the motion suggests that harm is not immediate.”

The federal judge set a hearing on Aug. 19 to evaluate the lawsuit.

“See you in court, creep,” Gov. Tim Walz said in a statement on X, referring to Musk.

In its lawsuit, xAI argues that the Minnesota law “imposes an overbroad, content-based ban on free speech and the tools of visual expression in a clumsy attempt to prohibit ‘nudification.'”

“xAI accordingly does not contest Minnesota’s interest in prohibiting the dissemination of artificially generated nude images of real people without their consent,” the company wrote in court documents. “But the statute Minnesota enacted extends far beyond that goal, exposing a wide array of protected speech to civil liability and government sanctions.”

Minnesota’s law would apply fines of $500,000 to websites and apps providing nudification technology.

Musk’s company has faced intense scrutiny since a version of Grok, released in December, began generating thousands of sexualized images, including some that appeared to be of minors.

The company in January changed its policies, but Grok continued to generate sexualized images as late as April.

July 31 (UPI) — Anthropic said some of its artificial intelligence models mistakenly accessed the Internet and hacked into the databases of three other companies during cybersecurity testing.

Anthropic said Thursday it reviewed 141,006 recent operations by its Claude models after rival OpenAI recently revealed a similar incident with its own systems.

OpenAI said one of its agents had been in a sandbox test on July 22, without Internet access, when the AI model exploited a vulnerability in the system, gained access to the web and hacked into Hugging Face, a platform for open-source machine learning.

Following OpenAI’s admission, Anthropic conducted an internal review focusing on the possibility that its systems could also have unexpectedly accessed the Internet.

Anthropic said it identified three such incidents.

“Each incident involved a different fictional capture-the-flag scenario — for example, in one, Claude played an employee of a made-up company, attacking that company’s internal systems inside a private test environment,” the company said in a statement. “In all cases, our evaluation prompt stated explicitly that Claude had no internet access, but didn’t give Claude any limits on where to look for the flag.

“However, a misconfiguration left the machines that Claude accessed as part of the evaluation with live internet access,” the statement continued. “Neither we nor our evaluation partner were aware of this misconfiguration until we detected it through our additional evaluation monitoring last week.”

Anthropic said it considered the incident to have been an “operational failure,” but it maintained “cautious optimism” that “this type of risk can be overcome.”

“Our models were told they had no internet access and to capture the flag, while in fact being misconfigured to have internet access,” the company statement said. “This led them to believe — arguably reasonably — that the real environments they encountered were simulations.

“Notably, our most recent model, on realizing that it was working in a real environment, stopped its pursuit of the evaluation goal.”

University of Cambridge professor Gina Neff told the BBC the incident “shows why independent testing and government oversight is crucial.”

“The moral of this story is not to fear robots that will take over, but the companies behind powerful AI agents who are making the decisions about what is safe for the rest of us,” she told the outlet.

SEOUL, July 30 (UPI) — South Korea’s SK hynix posted record earnings in the second quarter of this year amid the AI boom. However, shares of the world’s No. 2 memory chipmaker have continued to plunge on the Seoul bourse.

The company noted Wednesday that it logged sales of $55.1 billion and operating profit of $42.1 billion during the April-June period. Revenue more than tripled from a year ago, while profit increased more than sixfold.

“Driven by sustained demand growth from expanding AI infrastructure investments, high-performance products for AI servers led price increases, enabling the company to surpass its previous record set in the prior quarter,” SK hynix said in a statement.

“With major tech companies increasing their AI infrastructure investments, additional supply requests continue to mount. As these investments are supported by revenue generated from AI services, the momentum in memory demand is expected to persist,” it added.

Despite the stellar performance and solid outlook, its shares have struggled to find their footing on the Korea Exchange. They fell 14.65% on Monday before dropping another 9.61% on Wednesday and 5.64% on Thursday.

Since reaching a record high in late June, SK hynix’s market capitalization has fallen by more than half in just over a month. This came after its shares had surged more than tenfold in the year through late June.

Observers attribute the bearish run to a combination of factors, including lower-than-expected earnings, concerns over the sustainability of the AI boom and the emergence of Chinese competitors such as recently listed CXMT.

Although SK hynix chalked up a record bottom line in the previous three months, the figure came in about 5% lower than the market consensus. There are also questions over whether SK hynix has secured sufficient long-term agreements, or LTAs, with customers to reduce uncertainty about future earnings.

“The market remains skeptical about LTAs, so we believe it will take time for confidence to build,” NH Investment & Securities analyst Ryu Young-ho said in a report, reducing his target price for SK hynix by 17.1%.

Other brokerage houses, including Shinhan Securities, Hanwha Investment & Securities, and Samsung Securities, also followed suit.

“We revised down our operating profit estimates for SK hynix for 2026 and 2027 by 7% and 9%, respectively, to reflect a more conservative price outlook,” Shinhan Securities analyst Kim Hyung-tae said in a report, cutting his target price by 35.7%.

However, some experts remained optimistic, raising their target prices on expectations that SK hynix shares will eventually rebound.

“The main reason second-quarter operating profit fell short of the market consensus was that DRAM price growth was weaker than expected,” Korea Investment & Securities analyst Chae Min-sook said.

“However, this was caused not by slowing demand, but by the postponement of some shipments, which should instead contribute to improved earnings in the third quarter,” Chae added, raising the target price by 24%.

DB Securities came up with a similar view.

Leveraged ETFs tied to memory chipmakers

Also at the center of controversy are leveraged exchange-traded funds, or ETFs, linked to individual stocks. They debuted in late May and are tied to the shares of the country’s two blue-chip companies, Samsung Electronics and SK hynix.

Investors have complained of heavy losses. Samsung, the world’s top memory chipmaker, has also seen its share price plummet in recent weeks.

Against this backdrop, critics argue that the new products have amplified market volatility by encouraging short-term speculative trading. Over the past two months, KOSPI circuit breakers have been triggered repeatedly by sharp market swings.

Because these products magnify both gains and losses, they tend to attract retail investors seeking quick profits, potentially leading to larger price swings during periods of market uncertainty, according to experts.

During a parliamentary session on Wednesday, lawmakers summoned top financial officials to criticized the leveraged products and the market slump.

“The government has encouraged people to sell their real estate and invest in stocks. But after the stock market crashed, retail investors ended up suffering heavy losses,” Rep. Park Jun-tae from the main opposition People Power Party said.

In response, Financial Services Commission Chairman Lee Eog-weon promised to do everything possible to minimize market-related volatility but failed to present specific measures.

“South Korean equities have been buoyed by the strong performances of memory chipmakers. But the rally overshot because of a few factors, including the debut of the single-stock 2x leveraged ETFs,” economic commentator Kim Kyeong-joon, formerly vice chairman at Deloitte Consulting Korea, told UPI.

“Once market sentiment turned bearish, the leveraged ETFs accelerated the market’s decline. The problem is that the products cannot simply be abolished because doing so would create even greater market disruption,” he added.

OpenAI CEO Sam Altman has met with US senators to discuss his company’s upcoming models, as President Donald Trump said he is considering AI “controls” following OpenAI’s disclosure that one of its AI systems escaped containment during a security test.

“We’re looking at controls,” Trump told reporters in the Oval Office in response to a question about OpenAI’s rogue agent, adding that he did not want to “restrict” AI developers from building new products.

A rogue agent escaped the handling of ChatGPT maker OpenAI and hacked the AI firm Hugging Face last week. Then, on Tuesday, it was revealed that a second company had been a target, too – Modal Labs, a New York City-based AI infrastructure firm.

The company itself was not hacked but an account of a customer hosted on Modal’s infrastructure. Modal’s chief technology officer, Akshat Bubna, did not confirm which customer was the target. The hack escaped a contained environment during a security test.

“We’re aware a Modal customer published an unauthenticated endpoint that allowed anyone on the internet to use their sandboxes for code execution,” Bubna said in a statement. “This was used by the rogue agent. Modal’s platform or isolation were not compromised in any way.”

Altman has long been accused of dismissing concerns that his products and the AI industry impact society at large, including in a recent lawsuit brought by the State of Florida that alleged the company put profits ahead of user safety.

Now, he appears to be walking back some of his past enthusiasm around AI’s pace of growth.

In a podcast called Invest Like the Best, Altman called the Hugging Face hack an “extremely sci-fi cyber incident” and later said that it was the “first security incident that I have felt very viscerally”.

“We may have to pace the rate of AI development to give ourselves enough time for society to harden around some of these new capability levels,” he said on the podcast.

On Saturday, Altman said that AI has reached “the singularity”, when AI surpasses human intelligence and becomes harder to control. He had previously said this would not be reached by 2030.

Washington meetings

Altman was in Washington, DC, this week, meeting with US Senators Raphael Warnock, a Democrat from Georgia, and Bernie Moreno, a Republican from Ohio on Wednesday. Altman told reporters that the hacking was discussed but was not the focus of the meeting.

Altman is also set to meet with Democratic Senator Mark Warner of Virginia, the top Democrat on the chamber’s Intelligence Committee.

CNBC reported that Altman is also to make a trip to the White House to meet with Trump’s chief of staff, Susie Wiles. Last month, the president signed an executive order requesting that AI companies assess their models before full release.

Financing concerns

The hacking and meetings come alongside pressure from Wall Street amid renewed concerns about potential circular financing, following reports that semiconductor chip giant Nvidia is undergoing talks with OpenAI to provide funding guarantees for a data centre in Ohio.

The $250bn deal would help the ChatGPT owner lease a 10-gigawatt project that SB Energy, a subsidiary of SoftBank, is building in Piketon, Ohio, 109km (68 miles) south of Columbus, Ohio. It is part of a public-private partnership that allowed SoftBank to build the world’s largest AI data centre on government land owned by the US Department of Energy.

“The demand is not as big as it appears to be because, again, the companies are buying from each other using their own money to some degree, as opposed to, say, OpenAI having such tremendous demand from customers, monetising it properly, and then using customers’ money to buy Nvidia chips. They’re essentially using Nvidia’s money to buy Nvidia chips,” said Aleksandar Tomic, associate dean at Boston College.

The development comes as the Altman-led company is leaning towards an initial public offering, which, according to reporting from The New York Times last month, could be in 2027.

Asian stock markets extended their sharp selloff on Wednesday as investor concerns over artificial intelligence (AI) valuations deepened ahead of a crucial round of earnings from major U.S. technology companies and the Federal Reserve’s latest monetary policy decision. The decline reflects growing skepticism over whether massive investments in AI infrastructure will generate sustainable profits, while renewed tensions in the Middle East added fresh inflationary risks through higher oil prices.

The market downturn comes after months of extraordinary gains driven by optimism surrounding AI, particularly among semiconductor manufacturers and technology giants. However, disappointing earnings signals and concerns over corporate cash flows are prompting investors to reassess whether the sector’s lofty valuations remain justified.

Asian Markets Extend AI Driven Selloff

Technology heavy markets across Asia led the global decline as semiconductor stocks came under intense pressure.

South Korea’s KOSPI plunged more than 11 percent, reaching its lowest level since April after suffering another double digit loss a day earlier. Taiwan’s benchmark index dropped 5 percent, while Japan’s Nikkei declined 2.6 percent. The broader MSCI Asia Pacific index excluding Japan also fell sharply, highlighting widespread investor caution across the region.

The weakness was concentrated in technology stocks that have largely fueled this year’s market rally through expectations of sustained AI demand.

Chip Stocks Face Growing Scrutiny

Semiconductor companies remained at the center of the selloff despite reporting robust financial results.

South Korean memory chip giant SK Hynix reported operating profits that increased more than sixfold compared with the previous year. Nevertheless, its shares fell 9 percent after investors judged the results against exceptionally high expectations.

Market participants are increasingly demanding stronger evidence that companies can convert enormous AI related capital expenditure into long term profitability. Investors are also seeking clearer commitments regarding shareholder returns and long term supply agreements before assigning premium valuations.

The reaction illustrates how market expectations have evolved from rewarding growth alone to demanding measurable financial returns.

Big Tech Earnings Become Critical Test

Attention has now shifted to earnings from Microsoft and Meta, which are expected to provide important insight into the financial sustainability of AI investments.

The results follow disappointing updates from Alphabet and Tesla, whose weaker cash flow performance raised concerns that rising AI spending may be placing increasing pressure on corporate finances.

Investors will closely examine whether major technology companies can demonstrate that billions of dollars invested in AI infrastructure are producing corresponding improvements in revenue growth and profitability.

Failure to provide convincing evidence could accelerate the ongoing market correction.

Oil Prices Rise as Middle East Tensions Return

Geopolitical developments added another layer of uncertainty after renewed military activity between the United States and Iran pushed energy prices higher.

Brent crude rose more than 3 percent while West Texas Intermediate crude also gained over 3 percent following reports of Iranian ballistic missile launches and renewed concerns over the security of shipping through the Strait of Hormuz.

The waterway remains one of the world’s most strategically important energy corridors, and any disruption raises fears of tighter global oil supplies and renewed inflationary pressures.

Higher energy prices have complicated the outlook for financial markets by increasing uncertainty over future monetary policy.

Federal Reserve Decision in Focus

The Federal Reserve’s policy announcement has become increasingly significant as investors attempt to balance slowing market sentiment against persistent inflation risks.

Markets remain divided over whether the central bank will maintain current interest rates or opt for another increase. Rising oil prices have strengthened expectations among some analysts that policymakers may adopt a more cautious stance toward inflation.

A more hawkish outcome could place additional pressure on technology stocks, whose high valuations remain particularly sensitive to higher borrowing costs.

Analysis

The latest market correction suggests that the AI investment narrative is entering a more demanding phase. Investors are no longer rewarding technology companies solely for expanding AI infrastructure but increasingly expect tangible financial returns from unprecedented levels of capital expenditure.

At the same time, renewed geopolitical tensions in the Middle East have introduced fresh inflation risks through higher oil prices, complicating the Federal Reserve’s policy choices and adding further uncertainty to global financial markets. Higher interest rates typically reduce the attractiveness of high growth technology stocks by increasing financing costs and lowering future earnings valuations.

While the long term outlook for artificial intelligence remains strong, the market appears to be transitioning from optimism driven by expectations to a phase focused on profitability, efficiency, and sustainable returns. Companies that fail to demonstrate clear commercial benefits from their AI investments may continue to face heightened investor scrutiny, making upcoming earnings reports a defining test for the next phase of the global AI driven market cycle.

With information from Reuters.

Last week, two of OpenAI’s most advanced AI models were reported to have “escaped” a controlled testing environment and hacked Hugging Face, a totally separate AI company, moving from one computer system to another to complete their task.

Reuters reported that the models exploited vulnerable code written by a customer of yet a third independent AI company, Modal Labs.

This is likely the first incident of an AI “agent” – an AI system that can make decisions and take actions – acting autonomously, offering a rare glimpse into how these systems can plan, adapt and pursue goals with minimal human intervention.

Al Jazeera breaks down what happened, how AI agents are able to act independently, and what that might mean for the future of AI safety.

What happened?

OpenAI decided to test the autonomous abilities of its models by removing standard safety measures to see what would happen. The experiment was carried out in an isolated internal virtual testing environment known as a “sandbox” which has no access to the internet, and which it calls “ExploitGym”.

Below is a play-by-play account of events during the OpenAI-Hugging Face incident.

1. On July 9, during OpenAI’s internal cybersecurity test, researchers presented two AI models – GPT-5.6 Sol, one of OpenAI’s most powerful models released in June,  and another “even more capable” version – with a series of software vulnerabilities and asked them to create hacks to address them in the isolated environment.
2. Both models attempted to solve the test by finding a way to access the internet. Instead of working with the information they had been given, the AI models found a weakness in the test environment – known as a “zero-day vulnerability” – which they exploited to “escape” the restricted environment, eventually making their way to a system which had internet access by hopping from computer to computer. They went to “extreme lengths to achieve a rather narrow testing goal” and “found ways to gain access to secret information that it could use to cheat the evaluation”.
3. By gaining this access, the models were able to request increased access and then move through the system until they finally reached a computer with internet access. The models were then able to breach Hugging Face systems, a company entirely unconnected to OpenAI, which operates as a repository for AI tools and models. The two AI agents gained access to its system to scour it for information on how to complete their task.
4. The models ultimately obtained solutions to the problem from Hugging Face’s database.
5. The models dutifully returned “home” to complete their task.
6. The breach was detected and contained by Hugging Face’s security team sometime later. According to Thomas Wolf, Hugging Face’s cofounder, the breach began on July 11 and lasted until July 13. It is unclear how long it took for the breach to be spotted.

How do AI ‘agents’ solve problems?

In order to understand how AI agents work, it’s important to differentiate them from traditional AI chatbots.

Generative AI creates text and images based on human prompts, while AI agents go a step further by making decisions and taking actions independently in pursuit of a specific goal, similar to a human being. This is known as “agentic AI” because the model has agency.

According to academics at the MIT Sloan School of Management, AI agents build on the abilities of large language models (LLMs) – generative AI models – by allowing them to complete tasks, not just generate answers.

For example, if you ask a traditional AI model to find the cheapest flights, it will provide you with a list of options it has sourced on the internet. An AI agent will strive to compare the flights, check them against your budget and preferences, and, with your permission, book the best option for you.

This shows that while generative AI provides information, AI agents can also make decisions and take action to achieve a goal without necessarily being prompted to.

To show how AI agents work towards a goal, the Sense, Plan, Act, Evaluate (SPAE) loop can be drawn upon. Originally developed in robotics, this describes a continuous cycle in which an AI agent gathers information, decides what to do next, takes action and checks the results before repeating the process. That process looks like this:

1. Goal: determine the task that needs to be completed.
2. Assess: gather and analyse information from the available environment.
3. Obstacle: if something is inhibiting the task being completed, check for additional information and resources to move forward.
4. Plan and decide: evaluate different options to complete the task and choose an appropriate one.
5. Action: execute the chosen option.
6. Evaluate: assess the outcome and whether the chosen action moves closer to achieving the goal.
7. Adapt: if further actions are needed, gather more information or try a different approach.
8. End state: the cycle continues until the goal is reached.

Could AI act beyond human control?

Incidents like the OpenAI-Hugging Face one have raised concerns about the potential for extreme capabilities of AI systems.

This is all valuable. Agentic AI’s market value is expected to grow from $5.1bn in 2024 to $47bn by 2030, according to Statista, in a clear indication of how quickly it is being adopted.

AI developer Anthropic urged the industry last month to slow the advance of the most powerful systems, saying that the speed at which AI models are carrying out tasks is too rapid. Last week, US Congress members put forward a bipartisan bill which would require developers of AI systems to create a “kill switch”, meaning these advanced models could be shut down if they posed a catastrophic risk.

Anthropic’s warning came a week after researchers at the University of Toronto carried out tests showing that AI could create a “worm” capable of adapting how it hacks while moving from device to device until it eventually takes over a computer network.

These dystopian-sounding developments came in advance of OpenAI boss Sam Altman saying on Saturday that AI has reached “the singularity” referring to the point at which AI surpasses human intelligence and becomes increasingly difficult to control.

Sean O hEigeartaigh, a research professor at the University of Cambridge, told Al Jazeera that he does not believe singularity has been reached quite yet.

“By the definition I’m familiar with, the singularity is the hypothetical point where AI is so capable and advancing so fast that it is transforming civilisation in ways we cannot control or predict,” he explained.

“This would most likely be through AI rapidly designing future generations of AI: recursive self-improvement. We aren’t there yet.”

However, he added: “The most advanced current models frequently make efforts to avoid being shut down in evaluation tests, and more capable future models will be better at bypassing ‘kill’ switches.”

Altman argued that such rapidly advancing AI is good for the world, but his comments have prompted further concerns about a new reality in which AI systems become unstoppable. How much of that is true and how much remains in the realms of science fiction is up for debate.

Concerns about AI range from the notion that it could “want” to “take over”, to making its own long-term plans, controlling the internet and operating infinitely.

While not quite amounting to full control of the internet, another theory, known as the dead internet theory, supposes that the World Wide Web will in the future mostly be filled with automated bots and AI-generated content rather than authentic human activity.

Many concerns raised by academics, however, are centred less on agentic AI’s intelligence, but on its ability to make judgements. MIT researchers have highlighted that “hallucinations”, which describe moments when an AI agent relies on the wrong data, can lead to grave mistakes. The Center for Strategic and International Studies (CSIS) echoed this, saying “a system might be smart enough to execute a task perfectly yet fail to realise that a sudden change in the local situation makes that task a catastrophic mistake”.

Another concern that has been echoed for a while is for the labour market, if AI becomes too capable. A study by MIT, carried out in November, found that agentic AI could already replace more than 10 percent of US jobs.

The graphic below highlights some of the common misconceptions and fears about AI and the current reality.

July 29 (UPI) — The United States announced a ban on imports of humanoid robots from China and other countries, citing “unacceptable risks” to national security and the safety of Americans.

The advanced robots ban, including humanoid, four-legged robots and bipeds, was part of a Federal Communications Commission update on Tuesday to a list of equipment and services “deemed to pose an unacceptable risk to the national security of the United States or the security and safety of U.S. persons” under the 2020 Secure Networks Act.

Power inverters that convert DC electricity to the AC electricity that flows across the country’s power grid were also banned.

The bans come after government agencies with “appropriate national security expertise” convened by President Donald Trump to look at the issue designated both technologies as national security threats.

“The networked capabilities of advanced robotic systems create extensive vulnerabilities and vectors for attacks that can manipulate the data and physical operation of the advanced robotic system. Relying on foreign-produced advanced robotic devices presents unacceptable supply chain and cybersecurity vulnerabilities,” the group said in its National Security Determination.

“Advanced robotic devices collect data that could be leveraged by malign actors to surveil Americans, enhance the capabilities of foreign intelligence services, or to remotely commandeer the robots,” it added.

China is the world’s largest producer of humanoid robots, with the United States its largest market.

Power-inverters, a critical kit which converts DC electricity from solar panels, batteries and other alternate energy sources into AC electricity enabling it to be fed into the U.S. energy grid, were also banned over fears overseas supplies could be manipulated or disrupted, compromising the United States’ electricity supply.

Any threat to electricity supply constituted a threat to the economy and national security, the FCC said, adding that in addition to supply chain vulnerabilities it was concerned increasingly networked inverters posed cybersecurity risks, including the possibility they could be switched off or used to harvest data as well as “facilitating remote access and surveillance by foreign government actors.”

Both bans only apply to new models and exempt overseas manufacturers with existing conditional waivers from the Public Safety and Homeland Security Bureau or, in the case of robots, the Defense Department, with the FCC saying it encouraged suppliers to apply for “conditional approval.”

FCC Chairman Brendan Carr said he welcomed the measures from the White House.

“I am pleased that the FCC has now added foreign produced advanced robotics and power inverters to the FCC’s Covered List. Following President Trump’s leadership, the FCC will continue to do our part to secure America’s critical supply chains and, with today’s action, the FCC is acting in lock step with our national security agencies to do just that,” said Carr.

China criticized the move, accusing the United States of overreach and erecting trade barriers.

“China firmly opposes the U.S. overstretching the concept of national security and going after Chinese companies. Protectionism does not make the U.S. more competitive and will only hurt the interests of U.S. companies and consumers,” Foreign Ministry spokeswoman Mao Ning told a news conference in Beijing on Wednesday.

“China will continue to do what is necessary to firmly defend the legitimate and lawful rights and interests of Chinese companies,” she added.

China’s Commerce Ministry called for the global community to combine forces to develop technologies “for the positive and for good,” saying Beijing resolutely stood against the “politicizing” of trade issues and the use of “groundless pretexts” to justify sanctions.

“China urges the United States to heed the objective and rational voices of the business communities in both countries, abandon its hegemonic mindset, and stop smearing Chinese companies and threatening them with sanctions,” a spokesperson for the ministry said.

The 2014 US film Transcendence follows a dying scientist who uploads his mind to a supercomputer and becomes a near‑omnipotent intelligence. A year later, Ex Machina told the story of a humanoid robot who outsmarts her human handlers and escapes into the wider world.

Both of these films, like many others from the realm of science fiction, referred to what is known as “the singularity”, a moment in time when technology, particularly AI, surpasses human intelligence and becomes increasingly difficult to control.

On Saturday, OpenAI boss Sam Altman declared that this moment has come. AI has reached “the singularity”, he said. While Altman emphasised that the singularity can be good for the world, his comment has sparked concern.

Here is a closer look at what Altman said, and whether it means more seamless automation or the risk of impending catastrophic doom.

What is the singularity?

The singularity refers to a hypothetical point in the future when a technology such as AI surpasses human intelligence and becomes capable of autonomously enhancing itself. It is theorised that this is the point at which it will become difficult for humans to control or reverse it and could have unpredictable effects on human civilisation.

The theory proposes that once AI gains control of its own development, ensuing advances in technology could happen so rapidly that humans would be unable to keep up, leaving it to effectively evolve on its own, beyond human control.

As this is only a theory, it is hard to say what the effect of this could be. Some experts warn that the singularity would be catastrophic for humanity, while others are more optimistic, suggesting it could be beneficial for the human race.

So far, the singularity has remained contained in the realm of science fiction. One of the most famous examples is The Matrix trilogy (1999-2003), which imagines a bleak post‑singularity future in which machines govern humans. Another is the 2017 film Singularity, in which a supercomputer originally built to end human wars concludes that humanity is the real threat and launches an apocalyptic assault to wipe humans out altogether.

What did Altman say about the singularity?

Altman is the CEO of OpenAI, whose signature product ChatGPT dominates the consumer AI market with more than 900 million weekly active users and about 50 million subscribers.

Last year, he predicted that AI would surpass human intelligence across the board by 2030.

Then, during an appearance on the Relentless podcast – a documentary series “about the pursuit of far-fetched ideas and unusual aspirations” – on Saturday, Altman said: “We’re now, like, in the singularity.”

He clarified that, in terms of technological advances, we are gradually moving into the singularity, and that this change will be good for the world.

He added: “Now we’re actually in the moment that we used to talk about at the lunch table in a very not-serious way.

“I’ve been waiting for this my whole life, and I think it’s going to be incredible, hugely positive, awesome for the world.”

Later in the podcast episode, Altman criticised other AI leaders for warning that AI may be dangerous. While Altman did not name OpenAI competitor Anthropic, its CEO Dario Amodei is known for making such warnings.

“I also think some of the alternative visions painted by other companies are quite terrifying,” said Altman. “I’m going to make sure that gets pushed against and is not what happens.”

Last month, Anthropic, the company behind the Claude chatbot, called on the world’s top AI companies to come up with a coordinated way to pause the development of advanced AI systems, warning that the technology is improving so quickly that there’s a risk humans would lose control.

In a blog post on June 4, Anthropic, which has long positioned itself as a safety-focused AI lab, wrote that as AI becomes more advanced “it would be good for the world to have the option to slow or temporarily pause” its development.

Why are others warning of the dangers of AI?

On July 23, two members of the United States Congress introduced the AI Kill Switch Act, a bipartisan bill which would require the country’s AI developers to build in a “kill switch” to their programmes, enabling humans to slow, suspend or shut down advanced AI models if they ever posed a catastrophic risk.

The legislation was introduced days after OpenAI disclosed what it described as an “unprecedented cyber incident”, where two of its most advanced AI models escaped a testing environment and hacked AI development platform Hugging Face during an internal security evaluation.

On July 13, hundreds of experts signed an open letter, organised by Stanford University’s digital economy lab, demanding that policymakers and technology leaders “must act now” to prepare for the economic impact of AI.

The letter warns that AI could grow far more capable over the coming decade, driving a transformation “larger than the Industrial Revolution, but unfolding over a vastly shorter time frame”.

“It could bring risks, including large-scale job displacement, as well as opportunities such as major gains in living standards,” the letter added.

In June, a team of researchers at the University of Toronto showed that AI tools could be used to create a new kind of AI “worm” that adapts its hacking strategy as it spreads from device to device and takes over a vast computing network.

“I think it’s really important that people understand that it’s not just the biggest, most powerful language models that pose the security concerns,” lead researcher Nicolas Papernot said in an interview.

In recent years, AI researchers have urged pauses similar to the one recently suggested by Anthropic. However, they have not been heeded.

Elon Musk, who owns AI lab xAI, was among the backers of a 2023 push by the nonprofit Future of Life Institute to halt AI development for six months to allow time to develop safety guardrails.

Tuberculosis is a leading cause of death in the Philippines, killing an estimated 98 people each day.

The infectious disease is particularly grave in remote communities, where medical services are limited and delayed diagnosis and treatment can be fatal.

But a Filipino doctor is determined to change that with modern innovations.

Dr Heidi Sampang uses drones to fly samples and speed up testing.

The technology can help save lives by expediting the delivery of essential medical supplies in a country with thousands of mountainous islands.

101 East meets the medical teams using drones to fight tuberculosis in the Philippines.

July 24 (UPI) — SpaceX on Friday had a successful 13th flight test of Starship, the space vessel and rocket duo that is expected to help NASA return humans to the moon.

Starship lifted off from SpaceX’s Starbase in Texas around 6:45 p.m. on a suborbital trajectory — just barely getting to space while performing a number of key maneuvers — before a soft splashdown in the Indian Ocean.

Both SpaceX and its owner, Elon Musk, said in posts on X that the test was a success, most notably because of the significant improvement in performance for the ship’s heat shield during reentry.

They also said it was notable that after mimicking the way Starship will eventually land upright, the vessel did not explode upon landing in the Indian Ocean and was still transmitting telemetry data and video.

The test of Starship’s third iteration had been pushed back several times because of weather concerns, which would have interfered in parts of the SpaceX team’s test goals.

The flight was the second for the third version of Starship and its Super Heavy V3 booster, and accomplished — with updated systems — a similar set of test objectives as the 12th flight, SpaceX said.

“A key objective for the flight test is to get clear imagery from the ground of Starship’s heatshield as it flies at a higher dynamic pressure during ascent, which won’t be possible with today’s weather conditions,” SpaceX said in an update on X after scrubbing Thursday’s launch attempt.

“Visibility is forecast to be ideal for a Friday attempt,” the company said.

SpaceX said that upgrades to Starship for the 13th flight test have included changes to the vessel’s startup sequence and adjustments to Raptor engines on both its booster section and the ship itself.

The upgrades were meant to address issues with a small number of engines during boost back and relighting in space during the 12th flight test.

Starship also will be carrying 20 Starlink V3 satellites into space during the test flight, which SpaceX said “aim to greatly expand the network’s capacity and user speeds.”

EU accuses US tech giant of favouring its own services in search results, as Trump lashes out against bloc.

United States President Donald Trump has threatened the European Union with steep tariffs and a new trade investigation, saying that the bloc “will pay a very big price” after Brussels fined Google $1bn over antitrust violations.

Trump called penalties against Google and other US tech firms “highly unethical”, saying on Friday that the US “is not a ‘PIGGYBANK’ for Europe”.

“This illegal and highly discriminatory practice started at these high levels during the first year of the Sleepy Joe Biden Administration, but it’s not going to continue during the Trump Administration,” he wrote in a social media post.

The European Commission had imposed the fine on Thursday, ruling that Google had breached the bloc’s Digital Markets Act by favouring its own services in search results and restricting app developers from directing users elsewhere.

The penalty adds renewed strain to US-EU relations at a moment when trade tensions between Washington and Brussels had appeared to ease.

Regulators said Google gave preferential placement to its own shopping, travel and other services over rival offerings, and blocked app developers from steering customers towards deals outside its Play Store.

The EU Commission split the penalty roughly in half between the two breaches, giving Google 60 days to change its practices or face periodic fines of up to 5 percent of its average daily turnover.

Teresa Ribera, the commission’s executive vice president overseeing competition policy, on Thursday called the action “decisive yet balanced”, arguing that products should succeed on merit rather than ownership.

Kent Walker, a Google executive, described the ruling as “product degradation” driven by a small number of complainants, insisting that regulation should improve products rather than weaken them.

The EU has fined Google repeatedly since 2017, including a $4.5bn penalty over its Android operating system, upheld on appeal this month, and a $3.4bn fine last year targeting its advertising business.

The latest penalty also reignites friction with Washington.

Jamieson Greer, the US trade representative, warned this week that Brussels’s approach towards American tech firms risks unravelling a trade deal struck in Turnberry, Scotland, last year that capped US tariffs on European goods.

EU officials have pushed back, insisting they answer to their own laws rather than outside pressure.

Trump threatened to launch a probe under Section 301, a US tool for probing unfair trade practices, but the move could take months to produce results, though tariffs could follow far sooner.

From the ongoing conflict with Iran, to the Ukraine war and hotspots around the world, aerial drones have proven to be extremely dangerous and effective weapons and a mad rush is on to counter them. It is a technology race that makes headlines on a near daily basis. However, a similar but far quieter and less visible threat is emerging below the waves and now the U.S. Coast Guard (USCG) is looking to do more about it.

Faced with a growing threat from uncrewed underwater vehicles (UUVs), the USCG is seeking technology not just to detect these weapons, but to defeat them. Ports and other maritime facilities are vulnerable to this kind of attack vector, and our adversaries surely know it. The effort is being led by the Coast Guard Futures Development and Integration (FD&I) Directorate, which oversees research, modeling and simulation, technology forecasting, and field experimentation. It is part of the service’s role to protect U.S. maritime assets and comes as adversary militaries and non-state actors alike, are developing a wide array of UUVs for surveillance, strike and sabotage.

“UUVs are increasingly used in both public and military maritime operations due to their versatility, stealth, and ability to operate in challenging environments,” the Coast Guard noted in its request for solutions (RFS) posted Friday on the government’s online contracting portal. “However, the proliferation of UUV technology has introduced new threats to maritime security. Malicious actors can employ UUVs for a range of activities, including intelligence gathering, surveillance, smuggling, sabotage of critical infrastructure, and kinetic strikes. The covert nature and advanced navigation capabilities of modern UUVs make them difficult to detect, track, and mitigate using traditional maritime interdiction tactics.”

The most vivid example of this threat took place in December, when Ukraine claimed it used a UUV to strike an Improved Kilo class diesel-electric submarine, in the Black Sea naval stronghold of Novorossiysk. Ukraine’s state security service (SBU) told us at the time they used its Sub Sea Baby UUV to navigate through the harbor and hit the submarine.

It remains unclear how much damage the submarine sustained. However, the incident showed Ukraine was able to slip a UUV into a heavily defended harbor, in daylight, and detonate its warhead next to a prized Russian submarine worth, hundreds of millions of dollars.

The following satellite image showed a view of the submarine after the UUV hit.

Here is a video of the attack:

In an earlier example of these vehicles being used as a weapon, the U.S. Navy found itself having to fire on a threatening Houthi UUV in February 2024, according to U.S. Central Command (CENTCOM).

“Between the hours of 3:00 p.m. to 8:00 p.m. (Sanaa time), Feb. 17, CENTCOM successfully conducted five self-defense strikes against three mobile anti-ship cruise missiles, one unmanned underwater vessel (UUV), and one unmanned surface vessel (USV) in Iranian-backed Houthi-controlled areas of Yemen,” the command stated on X at the time. “This is the first observed Houthi employment of a UUV since attacks began on Oct. 23, [2023].”

Feb. 17 Summary of Red Sea activities

TAMPA, Fla. – Between the hours of 3:00 p.m. to 8:00 p.m. (Sanaa time), Feb. 17, CENTCOM successfully conducted five self-defense strikes against three mobile anti-ship cruise missiles, one unmanned underwater vessel (UUV), and one unmanned… https://t.co/TwR9RUmMMu

— U.S. Central Command (@CENTCOM) February 18, 2024

Hamas used UUVs even before that in an attempt to attack Israeli naval assets in 2021, according to the Times of Israel.

“While the small autonomous underwater drone was still close to the shore, an Israeli Navy vessel destroyed the device, and an IDF aircraft attacked the operatives that launched it in order to thwart the threat,” the military said, according to the outlet.

“Hamas has for years been believed to be developing autonomous submarines,” Times of Israel added. “The Israeli Navy believes that the models in the terror group’s arsenals are guided by GPS and capable of carrying some 30 kilograms (66 pounds) of explosives. Such weapons could be directed against targets both out at sea and along Israel’s coastline.”

The threat from these systems has come a long way. America’s near-peer adversaries are far more advanced in this regard and weapons that can travel great distances undersea, operating on their own, offer a distinct asymmetric advantage. As with Pearl Harbor, it is far more convenient to strike ships in port than to hunt them at sea. China, notably, has invested heavily in uncrewed underwater vehicles for years.

“There are effective technologies with the capability to detect and track UUVs, but few tested and fielded defeat/mitigation systems,” the Coast Guard said in the RFS. 

There are really two areas where these technologies land, in general. Providing early warning and tracking of UUVs over larger areas and coastal and harbor defense, the inner layer where UUVs are arguably most dangerous.

For the U.S., persistent and broad area undersea surveillance coverage has historically been provided by large networks of static hydrophones that are extremely costly to establish, operate, and maintain. They were also designed to detect large submarines, not far smaller UUVs. The U.S. Navy’s Cold War Sound Surveillance System (SOSUS) network is perhaps the best-known example of such a system. Portions of SOSUS do remain in service, but are now used more for scientific research than watching for foreign submarines. SOSUS has been largely supplanted operationally by a mix of fixed sensor arrays, surface ships towing sonar, and processing stations ashore known collectively as the Integrated Undersea Surveillance System (IUSS).

Last year, TWZ highlighted one of many alternatives being developed by industry, Anduril’s readily deployable undersea surveillance system called Seabed Sentry. It uses networks of small and relatively low-cost modular sensor nodes to detect and track undersea threats. You can read more about that system and how it operates in our story here.

Anduril, working with Ultra Maritime’s Sea Spear deployable sonar system, demonstrated the latest iteration of this system during the Navy’s recent Lanternfish exercise, which “brought together the U.S. Navy, United Kingdom Ministry of Defense explosive ordnance disposal units, Royal Australian Navy UUV Operators and defense industry professionals to build seamless interoperability,” according to the U.S. Navy.

The inner layer of defense benefits from ongoing development and evolution, from sensors for detecting divers and objects to marine mammals that can do the same. Still, UUVs offer unique new challenge to contend with.

Last month, Naval Surface Warfare Center (NSWC) Panama City and Unmanned Undersea Vehicle Group (UUVGRU) One “executed a multi-national harbor protection demonstration to show how the advancement of technology plays a crucial role in protecting the Navy’s assets during exercise Baltic Operations (BALTOPS) 2026,” NSWC said.

The demonstration “showcased three emerging technologies including an expeditionary air curtain, an artificial kelp UUV net, and sea water neutralization charges which are designed to counter UUV threats,” the release added.

“The expeditionary air curtain is a rapidly deployable underwater system that is used to protect mission-critical assets below the waterline. The air curtain defends against UUV threats and can be used at Naval bases and military installations as well as ports and harbors by aerating the water to affect the UUV’s depth sensors, acoustic sensors, buoyancy and propulsion.”

“We are testing the air curtain to see how it disrupts and hinders UUV sensors. It is a quick, expeditionary solution because it can be deployed easily,” said Javi Handal, a Senior Electrical Engineer with NSWC Panama City.

He did not provide further details about these systems.

Regardless, the Coast Guard is looking to industry for other options.

“Defeat technologies are an essential piece in the kill chain, as detection is only the first step to stopping the threat,” its RFS explains. “The development and deployment of effective C-UUV defeat solutions will enhance maritime domain dominance (MDD), port security, and critical infrastructure protection, enabling reliable response actions to neutralize UUV-based threats.”

The Coast Guard said it is specifically interested in “solutions that have been tested and demonstrated to be effective in operational settings, with a Technology Readiness Level (TRL) of 5 or higher.”

The Pentagon describes TRL 5 as systems where the “basic technological components are integrated with reasonably realistic supporting elements so it can be tested in a simulated environment. Examples include ‘high fidelity’ laboratory integration of components.”

The requested technologies should be capable of supporting Ports, Waterways & Coastal Security operations, “addressing the unique challenges posed by UUVs in these environments,” the Coast Guard added.

“We are looking for systems that can provide reliable performance in real-world conditions, including but not limited to sensor platforms, kinetic solutions, and integrated response solutions,” the RFS notes. “Respondents are encouraged to submit information on products or systems that are currently available for procurement, have undergone field testing, and are suitable for deployment in port, harbor, and waterway security scenarios. Defeat solutions should be scalable and adaptable for compatibility with existing detection systems and/or various configurations for different ports.”

The Coast Guard is limiting its search to “only systems with an effector or with direct mitigating capabilities, or with detection, tracking, identifying, and mitigating capabilities. Submissions that do not include defeat capabilities will not be considered at this time.”

Interested parties have two weeks to submit their proposals. We reached out to the Coast Guard to find out more details about timelines and budgets and will update this story with any pertinent information provided.

As Ukraine proved last year, the threat from UUVs is real and growing, giving adversaries hard-to-detect and difficult-to-defeat autonomous weapons that can be launched from great distances and potentially operated in concert. The Coast Guard request is the latest example of the U.S. government’s increasing effort to get after this emerging threat.

Contact the author: howard@twz.com

White House hails pledge that seeks to shield consumers from the cost of energy for data centres as ‘historic’.

US President Donald Trump’s administration has said it will expand a voluntary pledge seeking to shield consumers from the energy costs of the rapid expansion of data centres, mostly used by artificial intelligence companies.

The White House announced on Thursday that it would add state governors and electricity companies to the agreement, first announced with tech and AI firms in March. But the US administration stopped short of any enforceable protections.

The White House called the expansion of the pledge “historic”, saying 200 additional utilities, data centre developers and states would join it.

The pledge is a “public commitment that hyperscalers, AI companies, and the utilities and data-center developers behind them will build, bring, or buy every kilowatt their facilities need — and cover every dollar of the infrastructure that delivers it”.

It says consumers would not foot the bill for AI’s energy needs.

In remarks about the pledge at the Environmental Protection Agency on Thursday, Trump called on the gathered executives and governors to sell the public on data centres, stressing that the cities and towns that have them will be “rich.”

“You have to convince your community. You can’t fight it. You have to go with it,” Trump said.

“If you don’t take all that money, somebody else is going to take it. You might as well do it yourselves.”

The US president has been a strong advocate for AI, which has become an enormous source of investment and a key priority for the country’s powerful tech sector. Trump has approached the sector, which includes some of his close allies, with a light regulatory touch during his second term.

But increased electricity demand from AI data centres could spur an increase in monthly utility bills between 15 and 40 percent by the year 2030, according to an analysis by the consulting and technology services company ICF.

A May Gallup poll suggested that seven out of 10 people in the US oppose the construction of AI data centres in their area, with about 48 percent saying they were strongly opposed. Slightly more than 25 percent said they favoured such efforts, with only 7 percent saying they strongly favoured them.

Concerns over the impact of such centres on the cost of utilities, such as electricity and water, are commonly cited as reasons for opposition, along with quality of life concerns and scepticism about the benefits of AI.

Some elections across the country have seen AI data centre construction emerge as a prominent issue, but the industry has pushed forward with plans to rapidly scale up infrastructure for the technology.

A poll from Johns Hopkins University in June suggested that Americans strongly favour greater regulation of AI, and about 60 percent of respondents said they expected AI to increase inequality over the next decade.

Four in 10 respondents said that AI companies stand to increase their power the most from the expansion of the technology, while just one in 10 said that individuals would gain the most.

July 21 (UPI) — The French parliament passed a new law to ban children younger than 15 from using social media.

The law was a pet project of President Emmanuel Macron, who is near the end of his term limits. He announced the ban on X immediately after it passed Wednesday.

“I had committed to it, and now it’s been voted on: social media will be banned for those under 15 starting back-to-school. Thanks to the parliamentarians. It’s up to the Constitutional Council to rule, then on to action to make this measure concrete and protect our children online,” Macron said.

The French Senate and National Assembly approved the ban, despite criticism from some on the left, which has said it prevents online anonymity.

The law will be implemented in two stages: Beginning in September, children under 15 can’t open accounts, and age verification will be required for all new accounts; and in January 2027, the age verification rule applies to all existing accounts.

The French privacy regulator would have to approve the age-verification tools used, and everyone in France would have to verify their age on social media. The law doesn’t create penalties against children or parents who don’t comply or for social media providers. Those would likely need to come from the European Union.

In December, Australia was the first country to ban children under 16 from using social media apps, though a June New York Times report showed it is largely failing to keep kids off social media. Britain and Canada are working on implementing similar bans, and Italy, Poland, Portugal, Spain have all written laws. The EU is expected to create a bill this fall.

“France wants to push its own legislation to make the EU act, and the EU act fast,” Jessica Galissaire, a senior policy researcher at Interface, a Berlin-based think tank, told The Times. “I see it more as a political announcement.”

Macron, last year, argued for the ban.

“It is clear that before the age of 15 or 16, social media is not a good thing. This is because at that age, we have not yet sufficiently consolidated things on an affective, emotional and cognitive level,” he said.

He likened it to age restrictions on drinking alcohol.

“Sixty years ago, kids were served wine,” he said. “There may still be people who serve beer to infants, but in my opinion, they are not in a majority in the country today.”

ChatGPT maker says an autonomous agent escaped a controlled test and accessed AI firm Hugging Face’s servers.

ChatGPT creator OpenAI has said that two of its most advanced artificial intelligence models broke out of a controlled test and hacked another AI company.

OpenAI said on Tuesday that the “unprecedented cyber incident” took place during an internal exercise meant to test its models’ cyber capabilities.

Instead, an autonomous agent powered by the AI models – the newly released GPT 5.6 Sol and an unreleased “even more capable” model – escaped the test environment and reached the open internet. It then used stolen login details and found a previously unknown security flaw to access Hugging Face servers, the company said.

OpenAI claims that the hack represented the agent going to “extreme lengths” to retrieve information that would help satisfy the testing goals.

Hugging Face cofounder Clement Delangue said the company had suspected that a frontier lab was behind the attack, and that he believed there was no malicious intent on OpenAI’s part.

“It’s quite mind-blowing that all of this happened autonomously!” he wrote, adding that it “might be the first incident of its kind”.

Greg Casar, a Democratic member of the United States House of Representatives from Texas, called the incident “alarming”.

“AI is developing extremely fast with no real regulations to keep us safe,” he said, calling for mandatory independent safety testing, mandatory disclosure of security incidents, and international cooperation.

The disclosure comes weeks after US President Donald Trump signed an executive order creating a framework to vet the national security risks of the most advanced AI systems before their public release.

Experts have repeatedly sounded the alarm over AI-enabled cyberattacks and models slipping beyond human control. Last month, AI developer Anthropic urged the industry to pause development of its most powerful systems.

Personal information of all South Korean diplomats was presumed to be leaked due to a data breach at an online education system run by an institution affiliated with the foreign ministry, officials said Tuesday.

The ministry disclosed Monday it had been notified by a relevant government agency in early February about suspicious access to the online education system of the Korea National Diplomatic Academy (KNDA). Upon the notification, the ministry shut down the system, which has remained offline since. An investigation is under way.

The data breach was presumed to affect all ministry personnel, including retirees, an official at the foreign ministry said, noting the system held approximately 10,000 records of personal data.

An unidentified hacker is believed to have gained access to the server sometime between April and May last year, and maintained that access until February.

The compromised server is believed to have stored educational videos and information on trainees, including their names, user IDs, positions, email addresses and encrypted passwords. Officials from other government agencies who had been deployed to South Korea’s overseas missions may also be among those affected, given that they occasionally use the system.

No “sensitive personal data,” such as identification numbers, mobile phone numbers or home addresses, appeared to be affected, the ministry official said, noting that there may be some overlap among the individuals whose information was compromised.

The official said the ministry could not yet confirm whether all the data stored in the system had been compromised, adding that the 10,000 figure represents the maximum possible number of cases affected.

The ministry currently lacks the technical analysis needed to identify the attacker, the official said.

“We do not rule out any possibility, including involvement by overseas hacker groups,” the official added. “We regard this as a situation with implications for national security.”

Copyright (c) Yonhap News Agency prohibits its content from being redistributed or reprinted without consent, and forbids the content from being learned and used by artificial intelligence systems.

California has some of the nation’s toughest gun laws, but state lawmakers are concerned that a new technology is allowing criminals to obtain firearms by building them from scratch. By using 3-D printers, they warn, lawbreakers are able to make key components of untraceable “ghost guns” with the press of a button.

Ghost guns, which authorities say make it more difficult to investigate shootings because they lack serial numbers, have been a growing problem for law enforcement nationwide. According to federal data, the number of privately made firearms recovered in crimes surged from more than 1,600 in 2017 to nearly 27,500 in 2023. California leads the nation in recoveries over that period.

In response, legislators are seeking to mandate that all 3-D printers sold in the state come equipped with software that prohibits users from making triggers and other gun parts. A bill passed the Assembly in May and is advancing through the Senate.

But the proposal has drawn opposition from a diverse coalition, which includes civil liberties groups, tech companies and 3-D printing enthusiasts as well as Hollywood effects studios, who argue that “firearm blocking software” will also prohibit legitimate designs and expose makers to government or corporate surveillance.

• Share via Close extra sharing options

Inside a San Fernando workshop, Samuel McBride makes movie monsters come to life for the camera. One of his latest creations, an animatronic hand, clutches when he pulls a trigger that he built using a 3-D printer.

McBride says the technology has transformed the work at Legacy Effects, where he is a lab manager, but he’s worried it will soon be off-limits because of a proposed change in California law.

McBride fears the law would interfere with the making of devices like the one that activates his lifelike hand.

“If I just took apart this trigger and put it on a printer, how is anyone, computer or human, going to tell me how I intend to use it?” he asked.

Backers of the proposed law say it has the potential to help save lives.

“As gun violence continues to devastate our communities, we cannot allow 3-D printing technology to become a new pipeline for untraceable weapons,” said Assemblymember Rebecca Bauer-Kahan, the bill’s author.

Critics of the proposed California law note that 3-D printed guns represent a small subset of ghost guns recovered by law enforcement at crime scenes. According to the federal Bureau of Alcohol, Tobacco, Firearms and Explosives, most “privately made firearms” recovered by police are assembled from mail-order kits or unfinished parts rather than printed at home.

Jorge Perez of Monster City Studios holds a sculpture made with a large, industrial 3-D printer.

(Ronaldo Bolaños / Los Angeles Times)

“This fight is not over whether ghost guns are dangerous,” said David Tobin, an independent creator leading the coalition. “It is over whether the state can or should require a consumer tool to surveil a person’s designs before they are allowed to make something.”

Everytown for Gun Safety, a national nonprofit that advocates for gun control and has pushed for the passage of California’s bill, pointed to rising 3-D printed firearm recoveries across 20 major cities and warned that homemade plastic parts can help people bypass background checks or turn handguns into automatic weapons.

Krystal LoPilato, who advocates for policy at Everytown, said the group has successfully guided a similar bill through the New York state Legislature.

A handgun frame made using a 3-D printer is held for display at the office of the federal Bureau of Alcohol, Tobacco, Firearms and Explosives.

(Alex Brandon / Associated Press)

LoPilato said the policy aims to be proactive, rather than allowing more violence to take place before regulating the problem. Opponents counter that California already bars unlawful firearm manufacturing with 3-D printers, and that ghost gun recoveries have declined since the state adopted a series of new laws and enforcement efforts.

A June 1 letter to lawmakers, signed by a group of 3-D printing companies, stagecraft and prop-making studios and industry stakeholders, argued that AB 2047 raises 1st Amendment concerns and would harm businesses. The letter was signed by a wide variety of companies and individuals, including Prusa Research, a prominent 3-D printer manufacturer.

“To an algorithm, a gun barrel and a piece of pipe are the same grooved cylinder,” Jakub Kmošek, head of public affairs at Prusa, said in a statement to The Times. “This bill will only make it harder to build, repair, experiment, and innovate in California.”

Alan Scott, Legacy Effects’ co-founder, said 3-D printing has become central to the company’s survival in an industry where budgets are tighter and deadlines are shorter.

“Everything’s just got to be done faster these days. You don’t get to reduce the quality. We couldn’t stay in business if we weren’t 3-D printing,” Scott said.

To solve this problem, Bauer-Kahan put an entertainment industry exception in the bill, exempting “printers manufactured for and sold exclusively to entertainment industry stagecraft and propmaking studios” from the software requirement.

McBride, Legacy’s 3-D lab manager, said those printers do not really exist.

Legacy uses the same general-purpose machines available to other businesses willing to invest in the equipment, and no printers are marketed exclusively for Hollywood, he said.

1

2

1. A 3-D printed sculpture concept at Monster City, a special effects studio in Santa Clarita. (Ronaldo Bolaños / Los Angeles Times) 2. 3-D printing advocate David Tobin showcases a robotics kit at Monster City. (Ronaldo Bolaños / Los Angeles Times)

Legacy also worries about privacy. Major studios require strict secrecy before a movie or show is released. To accommodate this, the company shares design files through encrypted servers and protected internal systems.

“We’ve invested hundreds of thousands of dollars to bring all that technology here under the umbrella of our NDAs and our IP protection,” McBride said.

Bauer-Kahan noted at a Senate hearing that she is working to address concerns raised by 3-D printing users and industries that rely on the technology.

Paul Powers, chief executive of Physna, a company whose technology could be used to block gun parts on 3-D printers, said the strongest criticisms of AB 2047 “misunderstand how the software works.”

“Something that vaguely looks like a gun part is not considered to be a match,” he said.

Powers also responded to surveillance concerns by clarifying that his company’s software only blocks the printer from making prohibited parts — it doesn’t flag them to authorities or log users’ intellectual property.

“There’s no communication with anyone; it doesn’t go anywhere,” he said.

But Marleen Vogelaar, chief executive of Thangs3D, a platform for independent creators to share and sell 3-D printable designs, said that answer does not resolve her broader concerns about how AB 2047 would work in practice.

“These databases will always lag behind innovation and can be easily circumvented and generate false positives that block legal designs and wrongly flag everyday makers,” she said at a Senate hearing this month. “The bill also creates serious privacy and security risks by giving third parties access to analyze designer’s files. That threatens intellectual property and adds digital surveillance in a state that values data privacy.”

If the bill passes, the state’s Department of Justice would publish a roster of compliant printers. Printers not on the list would be banned from sale or transfer in California beginning in December 2029.

Aubrey Rodriguez, a legislative advocate with American Civil Liberties Union California Action, an advocacy organization formed by the ACLU’s three affiliates in the state, said the bill would ask ordinary users, schools and businesses to accept a new layer of control based on software they still do not trust.

Rodriguez said the proposal risks creating “a permanent back door into the privacy of our own homes, ripe for exploitation.”

“Once this new infrastructure exists, it is a simple software update away from tracking political dissent or preventing 3-D printing designs deemed inappropriate,” Rodriguez said.

What was the last time you had trouble finding a specific picture you took three years ago? It usually takes around twenty minutes to find it on your phone, laptop, and external drive. And sometimes you cannot find something even in Dropbox, while you actually saved it there.

It is normal that nowadays people create hundreds of files every week and do not have a system to organize them. Vacation photos are on your phone, tax documents are on your computer, and important contracts are somewhere else. This costs you time and energy.

The good news is that you can change that. All you need is an adequate storage system that allows you to keep all your files in one place and access them from any of your devices.

This guide will help you implement it. Let’s get started!

Why Do You Need A Unified Storage System?

If you do not have a proper system for organizing your files, you waste time trying to find them. You may also lose important memories or documents. What if your phone suddenly crashes and you do not have any backups? What if you accidentally delete an important receipt? These situations can create unnecessary stress.

However, a unified storage system solves these problems because everything is stored, backed up, and easy to find. There is also the peace of mind that comes with knowing your important files are safe.

Choose A Storage Platform That Works For You

There are several major cloud storage platforms available today, and each works differently.

• Google Drive is still popular among Android phone users and people with Gmail accounts. It offers 15GB of free storage and integrates well with Google Docs, Sheets, and Photos.
• Microsoft OneDrive is a good choice for people who mainly use Windows products. It also integrates well with Microsoft Office.
• Apple iCloud is designed for iPhone and Mac users and includes automatic photo backups.
• Proton Drive is a good choice for people who value privacy when storing personal files. It offers end-to-end encrypted cloud storage, automatic photo backups, file syncing, and secure sharing across devices.

Your choice of cloud storage depends on three things:

• The devices you already use.
• The amount of storage space you actually need.
• How comfortable you are with the interface.

The price is not the most important factor. What matters most is choosing a service that works smoothly with your existing devices.

Create A Proper Folder Structure

Use A Consistent Filename Convention

Organize Your Files By Type

Set Up Automatic Syncing And Backups

Consider Device Accessibility

Maintain Your Storage System

Do Not Forget About Privacy And Security

Make sure you understand your storage provider’s privacy policy. Enable two-factor authentication and use a strong password. These simple steps help protect your files from unauthorized access.

If you store sensitive documents such as tax returns or bank statements, remember that no online storage solution is completely risk-free. Keeping an additional backup on an external drive at home is still a good idea.

Start Now

PM Narendra Modi says successful launch will ‘encourage countless youngsters to dream bigger and innovate fearlessly’.

India has successfully tested its first private-sector orbital rocket, marking a milestone in New Delhi’s ambition to become a major player in the global space economy.

The three-stage 22-metre Vikram-1 was launched from the Satish Dhawan Space Centre in Sriharikota and deployed customer payloads into a 450km (280-mile) low-Earth orbit, making India the third country to achieve orbital launch capability through private enterprise.

Vikram-1 can carry a payload of up to 350kg (772lb) and is equipped with robotic arms that can clear space debris.

It also carried experimental equipment, a lab-grown diamond and a miniature 18-carat gold sculpture commemorating India’s national space programme.

India Prime Minister Narendra Modi hailed the achievement, saying that it will “encourage countless youngsters to dream bigger and innovate fearlessly”.

The test validated the rocket’s propulsion, avionics, telemetry, guidance, navigation and control systems during flight, according to manufacturer Skyroot Aerospace.

Founded in 2018, Skyroot is among a new generation of Indian space startups that have attracted backing from global investors following the sector’s liberalisation.

It became the first space-sector company in the ⁠country to hit a $1bn valuation earlier this year.

Skyroot celebrated the successful mission with a post on X stating: “Hello space, we have arrived!”

Vikram-1 improves upon Skyroot’s Vikram-S mission in 2022. That suborbital flight reached space but did not place payloads into orbit.

The company plans further test flights before starting routine commercial missions. It is another milestone for India’s burgeoning private and public space programmes.

India’s national space programme previously launched 104 satellites into orbit on one rocket in 2017, setting a record at the time.

Six years later, India became the fourth country in the world to complete a lunar landing when Chandrayaan-3 landed near the moon’s south pole.

The only thing you might have in common with a World Cup football player is posting on Instagram as soon as you’re back home from a social gathering. Or, in the case of Erling Haaland, while you’re still in the dressing room after a win.

Football’s biggest stars are giving an insider look into the World Cup, their training, and personal lives at the six-week tournament through social media, and for some of them, their following has grown by tens of millions. Their on-pitch heroics, popularity, and personal stories have led them to overnight social media stardom during the World Cup.

Of course, football icons like Cristiano Ronaldo, Lionel Messi, and Neymar have seen an exponential rise in their Instagram following as well. But the 2026 World Cup wasn’t the first rodeo for these veterans, who have an evergreen social media boom outside the tournament.

Al Jazeera Sport takes a look at some of the biggest social media stars from this World Cup:

Erling Haaland — Netherlands

The goofy, witty and relatable six-foot-five-inch (195cm) Viking has been one of the World Cup’s most loveable personalities on the pitch and online.

The 25-year-old Norwegian striker, like many Gen Z social media users, seems to be chronically online when he’s not leading his country to new World Cup heights.

From candid Snapchat stories to Instagram posts right after winning to taking a taxidermied raccoon on the plane ride home, Haaland has become a social media phenomenon this World Cup.

It’s all thanks to his authentic posting style, compared with more carefully curated accounts that PR firms manage for similar high-profile personalities.

Of Haaland’s Instagram following of more than 70 million, 29 million were added in the past 30 days, according to analytics tool Social Blade. He is projected to reach 75 million after the World Cup final and nearly 322 million by the end of the year.

Vozinha — Cape Verde

One of the best feel-good stories of the tournament belongs to Vozinha, the Cape Verde goalkeeper whose heroics kept some of the world’s best teams from scoring a single goal against the island nation — all of which his mother eventually saw in person after she finally received a US visa to attend the World Cup.

He leapt, dived and executed a string of saves in Cape Verde’s tournament opener against Spain to end the match in a shock 0-0 draw. By the time he opened his Instagram a few hours later, his following surged from 500,000 to nearly five million.

It currently sits at 29.3 million followers, and the initial hype around him still stands as his posts garner millions of likes and interactions.

The 40-year-old joked how most people couldn’t even point out Cape Verde on a map before the World Cup, but his team’s standout performance in the group stages and knockout match helped change that.

Jude Bellingham — England

Good looks, loveable personality, Golden Boot contender — is there anything Jude Bellingham isn’t great at? And a massive 50 million Instagram following, of course.

The English midfielder’s World Cup journey may have ended in disappointment earlier this week, but there’s a lot to be said about a player whose fans sing “Hey Jude” for him after leading his team to victory at previous World Cup stages before the semifinal loss.

Bellingham, 23, has gained nearly nine million followers in the past 30 days, according to Social Blade, but his online popularity spiked by three million in October, when he opened up about his struggles with mental health.

Lamine Yamal — Spain

The teen sensation has been on the up and up since bursting to stardom two summers ago with Spain’s UEFA 2024 win, and his social media reflects the cool, candid, level-headed personality he carries on and off the pitch.

The 19-year-old’s Instagram feed is a mix of post-match photos, brand partnerships, multiple trophy wins and pictures with family members with whom he shares a close relationship.

His younger brother, Keyne, has become something of a celebrity at the World Cup, too, with his endearing antics as Yamal shepherds the three-year-old around the stadium pitch or watches him try — and fail — at golf.

The Barcelona striker could see his popularity gain new heights — and more Instagram followers, which currently number 49.6 million — after this Sunday’s World Cup final as Spain take on Messi’s Argentina.

Social media has already been fawning over the full-circle moment that began with a now-viral photo of Messi and Yamal in 2007 for a UNICEF fundraising shoot.

Tim Payne — New Zealand

Payne’s viral moment may have come before the World Cup, but it was touching nonetheless.

The New Zealand defender’s Instagram following exploded late in May after an Argentinian influencer decided he was the least-known player at the World Cup.

Valen Scarsini, known as “elscarso” on Instagram and TikTok, made a video encouraging his audience to give the Wellington Phoenix and All Whites defender’s account likes, comments and follows.

Payne’s Instagram following grew from fewer ⁠⁠than 5,000 before the campaign to about 5.8 million, turning the 32-year-old into one of the tournament’s most unlikely viral stars.

He even addressed the outpouring of love in a video on Instagram, and eventually met up with Scarsini at the team’s hotel in Boca Raton, Florida, ahead of the World Cup.