TheChek News Flow

The AI firm is allowing shoppers to buy goods without leaving the chatbot—and laying the groundwork for AI agent-based shopping

OpenAI is letting ChatGPT users buy things through its popular artificial intelligence chatbot, all without leaving the confines of its platform.

The San Francisco-based AI company said Monday that U.S.-based ChatGPT users will be able to buy goods from online marketplace Etsy’s domestic sellers, as well as some merchants on Shopify’s e-commerce platform. The service, called Instant Checkout, currently only supports single-item purchases.

OpenAI also unveiled an open-source technical standard for merchants to build integrations with ChatGPT, called Agentic Commerce Protocol, which the company hopes will draw more merchants onto its chatbot platform. The protocol allows merchants to make their products shoppable inside ChatGPT.

Amazon and Walmart, the nation’s two largest digital retailers, aren’t currently using the protocol, OpenAI said.

OpenAI’s announcements come as the company continues expanding the capabilities and reach of its flagship chatbot, which ignited the AI boom in late 2022. Over one in 10 people who use ChatGPT have some intent or interest in making a purchase, said Michelle Fradin, OpenAI’s product lead for commerce in ChatGPT.

That makes ChatGPT an ideal place for users to actually complete their purchases, rather than needing to leave the platform to finish buying something, OpenAI said.

OpenAI’s Operator AI agent, released early this year, can also buy goods on behalf of users, but requires them to manually enter their payment information at checkout. ChatGPT’s Instant Checkout feature is meant to be easier to use within the chatbot, plus, more users are using ChatGPT to search for products than Operator, Fradin said.

Merchants pay a small fee on completed purchases through ChatGPT. OpenAI declined to specify the amount of that fee, citing confidentiality agreements with its partners.

For tech giants like Google and Amazon, there’s a lot at stake as search and e-commerce are radically upended by tools like ChatGPT. People are turning to AI chatbots and services to act as personal shopping assistants, or even asking AI agents to do their purchasing for them.

As of August, OpenAI counted roughly 700 million people—9% of the world’s population—as weekly users of ChatGPT, up from 500 million in March.

Late last year, OpenAI pushed directly into Google’s territory with the launch of a search engine for ChatGPT. The company added product recommendations inside ChatGPT in April, making it easier for chatbot users to compare products and services.

For online merchants, too, there’s a risk of losing a direct connection to shoppers who might otherwise be visiting their websites.

“The merchant gets the sale, that’s the benefit, but they lose the [customer] loyalty,” said Emily Pfeiffer, a principal analyst focused on commerce at IT research and consulting firm Forrester. “A lot of retailers and brands are just very nervous because this is all moving super quickly.”

But that’s a threat OpenAI says won’t bear out.

“ChatGPT serves a very different purpose than a merchant’s website,” Fradin said.

The chatbot also sends customer payment information through the payments processor Stripe, which forwards that data to the merchant. Fradin said that allows merchants to maintain control over their customer relationships.

OpenAI’s latest move also serves to lay the groundwork for future AI agent-based shopping, which is still in its infancy. While e-retailers like Amazon and eBay have released their own AI shopping agents, they are neither ubiquitous nor fully autonomous.

For the most part, AI agents still need more technical infrastructure to work across apps and services. Agents will need permission to access apps, APIs and websites if they are ever going to call an Uber or book a flight. The effort has been helped along by Model Context Protocol, or MCP, an open-source standard introduced by Anthropic last year.

OpenAI’s Operator, which is a computer-use agent, is limited to using other apps and services through the web browser, and still requires human intervention to access them.

OpenAI said it aims for its own Agentic Commerce Protocol to serve as a foundation for merchants to reach more customers who might be searching for their products in ChatGPT.

Article argues output of laser weapon unveiled at Beijing military parade exceeds American Helios and LWSD Mark 2 MOD 0 laser weapons

A Chinese military magazine has shed light on the mysterious laser weapon unveiled at the country’s Victory Day parade in Beijing this month, and argues that its output exceeds that of its American counterparts.

According to Ordnance Industry Science Technology, a Chinese military journal, the Liaoyuan-1 (LY-1) shipborne laser weapon displayed in the military parade on September 3 will be tasked with “close-range interception of incoming missiles and drones”, serving as the “last line of defence” within the shipborne air-defence system.

The article, published in the latest issue of the journal, added that the LY-1 “may form a complementary range structure alongside the long-range HHQ-9C, medium-range HQ-16C and short-range HQ-10A [naval air-defence missile] systems”.

It also compared the LY-1 with the United States’ Helios laser weapon, saying the Chinese system’s overall structure was “considerably more robust” than the Helios.

“Particularly the LY-1’s lens aperture, which is nearly twice the diameter of the Helios’,” the article added. “Its auxiliary equipment, sensors, elevation mechanism and other subsystems are also far more advanced.”

The article also mentioned a larger LWSD Mark 2 MOD 0 laser weapon that was tested by the US Navy aboard the USS Portland, a San Antonio-class amphibious transport dock ship, in 2020.

It compared China’s laser weapon with the new US system and speculated that the LY-1’s power output was between 180 kilowatts and 250kW, which “slightly exceeds the naval laser systems already undergoing practical trials by the US Navy”.

Laser weapons are an effective countermeasure against rapidly advancing unmanned aerial warfare, offering advantages such as high operational endurance and low running costs. Many nations are investing in the development of similar directed-energy weapons.

The USS Preble destroyer has been equipped with Helios, a Lockheed Martin weapon designed to intercept aerial threats, such as drones, missiles and small boats.

The LY-1 is regarded as China’s response to Helios. Beijing has said the weapon has entered service, although it is not known which vessels are carrying it.

The article in Ordnance Industry Science Technology cited unnamed military experts as saying the LY-1 “has sufficient space to accommodate power units, implying greater potential output”.

“Not only can it intercept incoming maritime unmanned systems, but it could even be used against anti-ship missiles, with a very low per-interception cost,” one expert said.

The cited experts also noted that the LY-1 “could gradually replace close-in weapon systems such as the 630, 730 and 1130 on warships”, referring to these systems as a liability for surface combatants.

However, military analyst and former PLA instructor Song Zhongping said shipborne laser weapons would “coexist long-term” with close-in weapon systems and short-to-medium-range air defence missiles because their “excessive cost” necessitated “coordination and complementarity with conventional weapons rather than replacement”.

Song added that, for naval forces, laser weapons primarily intercepted low-altitude, slow-moving, small targets, including cruise missiles, “but are powerless against ballistic missiles”.

“Though laser weapons are promising, their deployment at sea, on land or in the air faces significant constraints. After all, the Earth’s atmosphere is not a vacuum. Therefore, laser weapons cannot become the sole offensive or defensive armament. China has given this matter thorough consideration,” he said.

Ordnance Industry Science and Technology journal also revealed the Chinese origin of the LY-1 designation as a name derived from a quote by Mao Zedong.

Liaoyuan means flames spreading across the wilderness.

In 1930, in a letter of encouragement to the Communist revolutionary army, Mao told frontline officer Lin Biao: “A spark can start a prairie fire.”

The phrase became a metaphor for new developments that begin modestly yet possess vast potential for future growth.

Artificial intelligence is hungry for power at a scale that defies belief.

Last week, OpenAI and Nvidia said they would work together to develop 10 gigawatts of data center capacity over an unspecified period. Inside OpenAI, Sam Altman floated an even more staggering number: 250 GW of compute in total by 2033, roughly one-third of the peak power consumption in the entire U.S.!

Let that sink in for a minute. A large data center used to mean 10 to 50 megawatts of power. Now, developers are pitching single campuses in the multigigawatt range—on par with the energy draw of entire cities—all to power clusters of AI chips.

Or think of it this way: A typical nuclear power plant generates around 1 GW of power. Altman’s target would mean the equivalent of 250 plants just to support his own company’s AI. And based on today’s cost to build a 1 GW facility (around $50 billion), 250 of them implies a cost of $12.5 trillion.

“We are in a compute competition against better-resourced companies,” Altman wrote to his team last week, likely referring to Google and Meta Platforms, which also have discussed or planned large, multigigawatt expansions. (XAI CEO Elon Musk also knows a thing or two about raising incredible amounts of capital.)

“We must maintain our lead,” Altman said.

OpenAI expects to exit 2025 with about 2.4 GW of computing capacity powered by Nvidia chips, said a person with knowledge of the plan, up from 230 MW at the start of 2024.

Ambition is one thing. Reality is another, and it’s hard to see how the ChatGPT maker would leap from today’s level to hundreds of gigawatts within the next eight years. Obviously, that figure is aspirational.

Then again, OpenAI’s fast-rising server needs surprised even Nvidia executives, said people on both sides of the relationship.

Before the events of last week, OpenAI had contracted to have around 8 GW by 2028, almost entirely consisting of servers with Nvidia graphics processing units. That’s already a staggering jump, and OpenAI is planning to pay hundreds of billions of dollars in cash to the cloud providers who develop the sites.

To put it into perspective, Microsoft’s entire Azure cloud business operated at about 5 GW at the end of 2023—and that was to serve all of its customers, not just AI. (Azure is No. 2 after Amazon’s cloud business.)

Bigger Is Still Better

Data center developers tell me most of OpenAI’s top competitors are asking for single campuses in the 8 to 10 GW range, an order of magnitude bigger than anything the industry has ever attempted to build.

A year and a half ago, OpenAI’s plan with Microsoft to build a single Stargate supercomputer costing $100 billion seemed like science fiction. Barring a seismic macroeconomic change, these types of projects now seem like a real possibility.

The rationale behind them is simple: Altman and his rivals believe that the bigger the GPU cluster, the stronger the AI model they can produce. Our team has been at the forefront of reporting on some of the limitations of this scaling law, as evidenced by the smaller step-up in quality between GPT-5 and GPT-4 than between GPT-4 and GPT-3.

Nevertheless, Nvidia’s fast pace of GPU improvements has strengthened the belief of Altman and his ilk that training runs conducted with Blackwell chip clusters this year and with Rubin chips next year will crack open significant gains, according to people who work for these leaders.

In the early days of the AI boom, it was hard to develop clusters of a few thousand GPUs. Now firms are stringing together 250,000, and they want to connect millions in the future.

That desire runs into a pretty important constraint: electricity. Companies are already trying to overcome that hurdle in unconventional ways, by building their own power plants instead of waiting for utilities to provide grid power, or by putting facilities in remote areas where energy is easier to secure.

Still, the gap between company announcements and the reality on the ground is enormous. Utilities by nature are conservative when it comes to adding new power generation. They won’t race to build new plants if there’s a risk of ending up with too much capacity—no matter who is asking.

‘Activating the Full Industrial Base’

OpenAI’s largest cluster under development, in Abilene, Texas, currently uses grid power and natural gas turbines. But other projects it has announced in Texas will use a combination of natural gas, wind and solar.

Milam County, where OpenAI is planning one of its next facilities, recently approved a 5 GW solar cell plant, for instance. And gas is expected to be the biggest source of power for the planned sites, this person said.

To accomplish its goals, OpenAI and its partners will need the makers of gas and wind turbines to greatly expand their supply chains. That’s not an easy task, given that it involves some risk-taking on the part of the suppliers. Perhaps Nvidia’s commitment to funding OpenAI’s data centers while maintaining control of the GPUs will make those conversations easier.

Altman told his team that obtaining boatloads of servers “means activating the full industrial base of the world—energy, manufacturing, logistics, labor, supply chain—everything upstream that will make large-scale compute possible.”

There are other bottlenecks, such as getting enough chipmaking machines from ASML and getting enough manufacturing capacity from Taiwan Semiconductor Manufacturing Co., which produces Nvidia’s GPUs. Negotiating for that new capacity will fall to Nvidia.

Predicting the future is notoriously difficult, but a lot of things will need to go right for OpenAI and its peers to get all the servers they want. In the meantime, they will keep making a lot of headlines in their quest to turn the endeavor into a self-fulfilling prophecy.

Agreement with Dar Global comes just months after US president’s visit to region

Donald Trump’s family business is expanding in the Gulf region’s fast-growing real estate sector, agreeing a deal with a Saudi developer for a $1bn development project in Saudi Arabia’s biggest coastal city Jeddah.

The project, the Trump Organization’s second planned tower in Jeddah, will include offices, high-end apartments and town houses. In a nod to the US president’s New York real estate background, it will form part of a larger scheme dubbed “Manhattan” featuring a Central Park-inspired park.

The planned Trump Plaza with Dar Global follows Trump’s visit to Saudi Arabia and its neighbours in May in the first planned foreign trip of his second term. It resulted in trillions of dollars in investments and business deals for US companies — from defence to artificial intelligence.

The Trump Organization has also licensed its name to Dar Global for projects in Dubai, Oman and Qatar, and has announced plans in Riyadh. The collaborations range from Trump-emblazoned golf courses and hotels to skyscrapers, with the Trump Organization collecting fees for the use of its brand and for managing some properties. The financial risk of buying and developing the land is borne by Dar Global.

Some cities in the region, including Riyadh and Dubai, have seen property prices soar in recent years, attracting increased interest from foreign investors.

The Trump Organization, managed by Trump’s son Eric, has increased activities in the Gulf following last year’s election, raising concern that the president’s family is using ties with Gulf leaders to benefit financially while he’s in office. But Trump’s holdings are held in trust until he leaves office.

Eric Trump said he was “honoured” to launch a new development in Saudi Arabia. “Together with Dar Global, we are creating a destination which will set a new benchmark,” he said.

The latest Trump family deal in Saudi Arabia comes as his administration is struggling to achieve one of its key foreign policy objectives — convincing the kingdom to normalise diplomatic ties with Israel. Saudi Arabia has condemned Israel’s war in Gaza and said it would not normalise relations without the establishment of a Palestinian state.

The new project in Jeddah will be built on a 1mn square metre piece of land that Saudi-listed Dar Al Arkan, Dar Global’s parent company, bought with other partners this year in a $1.19bn deal. That was a record-breaking land transaction for the city. Dar Global launched a $533mn Trump Tower in Jeddah last December.

Dar Global chief executive Ziad El Chaar called Trump Plaza in Jeddah “one of the most ambitious developments” the company has yet attempted.

“This is truly a once-in-a-lifetime development opportunity,” Dar Al Arkan chair Yousef al-Shelash said in a statement.

US president boosts shares in sector by sharing video advocating medical use of hemp-derived cannabidiol

Shares of cannabis companies leapt on Monday after Donald Trump drew attention to the potential benefits that use of hemp-derived cannabidiol could bring to senior healthcare.

The US president used his Truth Social account on Sunday to share a video from The Commonwealth Project, an advocacy group for the integration of medical cannabis into mainstream healthcare for over-65s, several weeks after saying his administration was reviewing the possibility of reclassifying marijuana as a less dangerous drug.

Shares in Tilray Brands, a medical cannabis company, were up 40 per cent by late morning in New York. Cannabis-focused Canopy Growth jumped 16 per cent and peer Cronos Group gained 15 per cent.

Trump did not comment on the video, which outlined potential benefits the use of cannabidiol (CBD) could have for seniors, including reduced pain and stress, improved sleep and the slowing of disease progression.

The video said it was time to educate doctors on the body’s endocannabinoid system, which regulates things such as mood and sleep, and “provide Medicare coverage for CBD.”

The video’s call for coverage under Medicare, the federal health insurance programme for people in the US aged 65 and older, comes as the White House and some senior lawmakers contemplate policy changes that could shake up the CBD market.

In August, Trump said his administration was reviewing the possibility of reclassifying marijuana at the federal level, a move that would follow on from President Joe Biden’s efforts to shift cannabis from a so-called Schedule 1 drug to Schedule III. About 40 states allow medical marijuana use.

The reclassification of the drug would allow cannabis companies to take advantage of standard business tax deductions from which they are barred at present.

Trump’s remarks — and the reposted CTP video — have given the sector some hope after uncertainty about the regulatory environment knocked share prices over the past year.

Mitch McConnell, the Republican senator from Kentucky, who championed the federal legalisation of hemp via the 2018 farm bill during the first Trump administration, had this year sought to roll back that legislation by banning hemp derivatives with a quantifiable amount of tetrahydrocannabinol (THC), the primary psychoactive ingredient found in cannabis.

A bill in California banning the sale of intoxicating hemp products outside authorised dispensaries is awaiting the signature of Democratic governor Gavin Newsom.

The difficulty of producing CBD products without at least some level of THC means a ban on cannabis with any trace amount of the compound could have adverse ramifications for US hemp growers.

Switzerland agrees with US not to manipulate its currency

Rare joint statement with US Treasury seen as ‘green light’ for further currency intervention by Swiss central bank

Switzerland has agreed with the US that it will not manipulate its currency, in a rare joint statement widely seen as a “green light” for further intervention by the country’s central bank to restrain a rise in the franc.

The Swiss National Bank and the US Treasury on Monday issued a declaration formally aligning their views on foreign exchange matters, with both countries promising not to “target exchange rates for competitive purposes”. The statement also recognised that market interventions are a valid tool for addressing currency volatility or “disorderly” moves in exchange rates.

“The US is recognising the SNB’s right to intervene. So it could be seen as a green light if they choose to do so again in future,” said Lee Hardman, currency strategist at MUFG.

The announcement follows years of tension over Switzerland’s currency interventions, which have seen the SNB frequently wading into markets to curb the strength of the franc. The currency often rises in times of market stress due to its role as a safe haven for investors, which weighs on exports and drags down inflation, making the exchange rate a critical focus for the SNB.

The Alpine country was placed on a US Treasury watch list for currency and economic practices in June, having been designated a currency manipulator by the US during President Donald Trump’s first term in 2020.

The Swiss franc has surged this year as investors sought refuge from Trump’s trade war, driving the dollar down 12 per cent to under SFr0.80 for the first time since the franc’s shock appreciation back in 2015.

The rise in the currency, combined with the threat from US tariffs, has threatened to push inflation, which dipped briefly into negative territory earlier this year, further below the SNB’s target. 

But the agreement with Washington gives the central bank greater room for manoeuvre if the franc’s rise resumes, analysts say.

Stefan Gerlach, chief economist at EFG, said easing tension between Bern and Washington reduces the risk of a political backlash if the SNB needs to weaken the franc once more.

It also lessens the risk that the central bank will need to push interest rates below zero in a bid to prevent the currency climbing further, he added.

At its decision last week to hold the interest rate steady, the SNB highlighted that inflation had risen slightly from minus 0.1 per cent in May to 0.2 per cent in August. 

Traders have trimmed some of their bets on a move to negative interest rates, with the market now ascribing a roughly one in four chance of such a move, according to levels implied by derivatives markets.

The timing of the announcement is notable as Switzerland battles higher tariffs from the US. The Trump administration in August raised duties on key Swiss exports to 39 per cent, including watches and cheese, creating fresh strains in bilateral ties.

“I think that this statement — despite being not legally binding and confirming existing practices — can be interpreted as a renewal of bilateral relations and practices in a complicated time,” said Edoardo Beretta, a professor specialising in political economy at the Università della Svizzera italiana.