>>> Semiconductor ETFs are now large enough to destabilize the entire US market:

Semiconductor ETFs are now large enough to destabilize the entire US market:

US leveraged ETF assets under management (AUM) hit a record $198 billion, according to Nomura, with some data showing it exceeds $200 billion.

The vast majority of this exposure is concentrated in technology, led by semiconductor ETFs.

As a result, the market impact of leveraged semiconductor ETF rebalancing has surged from ~$2 billion per 1% move in the S&P 500 in 205 to nearly $10 billion today, almost 5 TIMES larger in just 12 months.

In other words, for every 1% move in the market, leveraged semiconductor ETFs are now mechanically forced to buy or sell ~$10 billion of underlying stocks near the end of each trading day.

Meanwhile, the 3x leveraged semiconductor ETF,
$SOXL
, alone now holds a record ~$35 billion in assets.

And because
$SOXL
targets 3x daily returns, even a modest pullback in semis can quickly turn into a 20% to 30% drawdown, while a disorderly unwind could be far worse.

Put simply, the bigger these funds become, the more they can amplify both rallies and selloffs, mechanically buying into strength and selling into weakness.

The semiconductor trade has never been more leveraged, crowded and fragile.


Electrek : Controversial Ferrari Luce EV is an instant sellout in China

Controversial Ferrari Luce EV is an instant sellout in China

All the controversy surrounding the launch of the first all-electric sedan to ever wear the Ferrari badge hasn’t slowed down sales any – every single Ferrari Luce allocated to the Chinese market is already spoken for.

Ferrari unveiled its new Luce EV this past May, and the car – both the first regular production sedan in the company’s history and the brand’s first car penned by ex-Apple design guru Jony Ive – was immediately criticized for being too plain, too practical, and too electric (among other things) by legions of internet commenters. The negative feedback was enough to drive Ferrari’s stock down over 6% in a single day, leading the company to axe its long-serving chief marketing and commercial officer, Enrico Galliera, just weeks after the rocky debut, installing former BMW Italy boss Massimiliano Di Silvestre in his place.

Nobody seems to have explained all this to the Chinese, however. All 88 examples of the 3,988,000 yuan (~$586,000) Ferrari Luce allocated to that market were snapped up “immediately,” according to CarNewsChina.

That rapid sellout supports Ferrari CEO Benedetto Vigna’s claims that the Luce is “clocking up orders” despite the backlash, which could be interpreted to mean that any backlash against the electric Ferrari isn’t a market backlash, with “market” being defined as “people who can actually afford to buy the thing.”


Another take, however, is the idea that Ferrari dealers were effectively forcing customers to buy a Luce in order to gain access to the brand’s more exclusive (and expensive) models – a report that gained enough steam to appear in Bloomberg before being vehemently denied by Galliera in at least one interview prior to his axing.

“No, I was mad because we don’t respect what is written in this article, and it’s totally not correct,” Galliera told The Drive. “Let me say why: because since the very beginning, we made clear to our clients that this car is designed for a different target audience.”

We’ll see if that’s true if and when those 88 Luce buyers get fast-tracked into the next Ferrari hypercar soon enough.

TechCrunch : Why Wall Street thinks US memory maker Micron is the next Nvidia

Why Wall Street thinks US memory maker Micron is the next Nvidia

Micron, the Boise, Idaho-based memory chip maker, has captured Wall Street’s heart. Whether the love affair endures will heavily depend on how long the AI-driven supply crunch for memory chips lasts.

Micron promises that it has shored up its position for the long term, which would allow it to withstand a sudden drop in demand or overcapacity of supply. And Wall Street has become a believer, helping Micron briefly surpass the market valuation of Meta and Tesla for the first time on Thursday, though it floated back down by Friday to nearly match them.

Specifically Micron closed Friday’s trading with a market cap close to $1.27 trillion, while Meta was at $1.39 trillion and Tesla was at $1.42 trillion. Micron’s stock has soared over 236% in the past month alone, closing Friday at $1,132 a share. In comparison, it spent years upon years before mid-2025 at below $100 a share.

It’s a dizzying rise for a company that most consumers associated with the tiny memory cards that, back in the day, were commonly needed to boost PCs, smartphones, or other device storage.

Wall Street isn’t sweating over that product line. Micron is benefiting from the AI data center buildout boom that has created a shortage of system memory chips, both DRAM and NAND, which Micron makes, particularly High-Bandwidth Memory (HBM). A single AI server requires magnitudes more memory than a laptop.

AI system makers like Nvidia, as well as the hyperscalers building their own systems, are buying up large quantities of memory, such as Microsoft, Amazon AWS, Google, Meta and Oracle. This is forcing all the other companies who need memory to hoard it as well, from PC makers like Dell and HP, to other kinds of device makers.

This lack of supply, which has been dubbed RAMageddon, is predicted to persist into 2027. And it’s already driving up the price of consumer electronics like Apple products and Xbox consoles.

With the whole tech industry clamoring for more memory, Micron’s delivered blockbuster third-quarter earnings last week. Revenue quadrupled year-over-year to $41.45 billion, and profits skyrocketed from $1.88 billion to $28.2 billion over the same period. Micron also provided a positive outlook, forecasting fourth-quarter revenue of between $49 billion and $51 billion.

And Wall Street, which has been eager to find more public AI-related companies that may do as well as Nvidia, became even more enamored.

The historic problem for memory chip makers like Micron and Samsung is that building out manufacturing facilities to increase capacity is a time-consuming, expensive endeavor. And demand often falls just as companies can increase capacity, creating a glut and subsequent price drop.

Micron got ahead of any AI bust chatter by emphasizing a series of long-term supply agreements, including with Nvidia and AI lab Anthropic, that would presumably protect it. The company said in its earnings presentation that it has signed 16 strategic customer agreements across the data center, consumer, and auto market segments, which it expects to fundamentally transform its business model.

That seemed to convince a number of analysts that this company could be another long-term, profitable investment. In a research note, William Blair tech analyst Sebastien Naji noted demand growth continues to outpace the rate that new cleanroom space can come online.

“Given the strong likelihood of continued ASP growth in the coming quarters and improving revenue visibility thanks to a rapidly expanding set of long-term agreements (SCAs) with key customers, we see potential for more durable earnings growth and reiterate our Outperform rating,” Naji wrote.

Whether Micron really can sustain itself for long-term without a bust cycle remains to be seen. But for a brief moment on Thursday, this U.S. company was more valuable than some of the industry’s giants.

WSJ : China Resets the AI Race Plus, another primary win for Trump-backed candid

China Resets the AI Race
Plus, another primary win for Trump-backed candidates, and the World Cup reignites a long-running debate: Who lives better, Americans or Europeans?

1. China has matched Anthropic’s Mythos in cybersecurity.
Security researchers said that a new AI model released this month by China’s Zhipu AI, also known as Z.ai, can match the latest U.S. models when it comes to finding security bugs, a development poised to reset the global tech race and pressure the White House in its overhaul of U.S. AI policy.

2. Iran asserted sole control of the Strait of Hormuz after days of back-and-forth fighting with the U.S.
The latest comments by the Iranian foreign minister stand at odds with American arguments that the preliminary deal signed by the two countries doesn’t give Iran control of the waterway and that navigation must be unimpeded. Talks between the two sides were expected to resume this weekend but have been stalled by the fighting.

3. A father bankrolled a high-school football team for his son. The program fell apart as quickly as it took off.
The wealthy entrepreneur put millions into The First Academy as he tried to propel his son to a college career. The program pulled in elite players chasing rich endorsement deals but collapsed amid cheating accusations—and his son moved elsewhere.

4. Trump pick Julia Letlow won the GOP Senate runoff in Louisiana.
Letlow defeated state Treasurer John Fleming in the contest to succeed Sen. Bill Cassidy, who had run afoul of President Trump. Clinching the nomination in a safe Republican state marks a major victory for the congresswoman.

5. Crimea has gone from prize asset to liability for Russia’s Vladimir Putin.
A Ukrainian campaign of drone strikes has caused a fuel and electricity crisis in the occupied Black Sea peninsula just as the tourism season, the key driver of its economy, was kicking off. The flare-up adds an extra negotiating card for Ukraine as Russia struggles to advance in the eastern Donbas region, though it is far from certain that Moscow would be more ready to settle as a result.

FT : Robots, not chatbots, will realise AI’s potential Factory-floor application

Robots, not chatbots, will realise AI’s potential
Factory-floor applications of the technology could significantly enhance rich-world economies

Since the launch of OpenAI’s ChatGPT three-and-a-half years ago, the world has grown familiar with large language models. The technology is speeding up and augmenting activities ranging from life admin and research to medical diagnoses. AI’s true potential, however, extends well beyond clever chatbots.

Beyond generating digital content, machines fitted with AI can independently carry out physical actions in the real world using cameras and sensors. Examples of applications include intelligent equipment, autonomous vehicles and humanoids. They use “world models” — systems that understand how objects move, interact and respond, as opposed to LLMs, which are more akin to predictive text engines — and are trained using real and simulated data.

In the service-driven economies of the rich world, it is perhaps unsurprising that much attention and investment has been absorbed by the application of AI to cognitive tasks. But leveraging the technology in the physical world — particularly on the factory floor — is likely to be the greater economic prize.

µ

First, the potential productivity improvements could be significant. “LLMs may deliver faster near-term gains, but in theory physical AI can have the larger long-run productivity upside because it targets physical bottlenecks,” says Daniela Rus, director of the Massachusetts Institute of Technology’s Computer Science and Artificial Intelligence Laboratory. “After all, most working hours are still spent moving atoms, not bits.”

Developed nations in Europe and Asia have already made strides integrating robots into assembly lines to drive higher growth. But AI enables robots to also learn and adapt in complex production, construction and logistics environments, which can boost efficiency and quality, and slash changeover times between different products.

Early deployments illustrate the potential. Foxconn, which produces intricate electronics including iPhones, found that vision-guided, self-adjusting robotic arms improved its assembly cycle times by up to 30 per cent, while reducing error rates by 25 per cent.

Adoption by Amazon in a US warehouse found that package-carrying robot fleets fitted with AI have learned to navigate around moving humans and obstacles, cutting their travel time by 10 per cent. (This video demonstrates how physical AI works in industrial settings.)


Next, AI-powered robotics could alleviate labour shortages in rich nations. University-educated talent is in ample supply for knowledge-intensive services jobs, which now appear to be on the frontline of LLM disruption. But manufacturers frequently cite a lack of workers as a constraint in production.

In a 2026 survey by consultancy Capgemini, more than 50 per cent of industrial executives said labour shortages, costs and regulation would be among their top five reasons for adopting physical AI. (Autonomous machines can also conduct bespoke repairs in environments and conditions that are hazardous to humans.)

Political tensions over immigration policy, ageing demographics and shifting attitudes towards physical work raise the long-term case for investing in AI-enhanced robotics.

Though the technology will displace some jobs, it will also create higher value-added roles on the factory floor, where knowledge of manufacturing processes is still needed to supervise, reconfigure and train robots. (In rich nations, where anxiety about AI’s impact on jobs is rising in dominant white-collar work, a shift towards embedding the tech into manufacturing might be relatively less politically disruptive.)

Finally, physical AI complements efforts by governments to boost domestic supply-chain resilience in critical industries. Indeed, multitasking machines could reduce the need for large assembly lines, which means production can take place at home at a lower marginal cost.


There is, then, a clear economic and political case for rich-world policymakers to place greater emphasis in supporting the adoption of physical AI in industry. China has made significant advances in applying the technology to the real world and Beijing’s latest five-year plan includes a strategic pivot towards “embodied AI”.

The transition will take time in the west. Capital, further developments in world models and vast quantities of training data are required. The necessary supply chains to build robotics hardware are essential, too. And factories and warehouses need redesigning. But venture capitalists are increasingly seeking out investments in robotics.

There are, of course, broader applications of embodied AI. Use of autonomous vehicles, such as robotaxis, is growing. Humanoids are often touted as the future of social care and household chores. And use cases in surgical robotics look promising.

Yet manufacturing is likely to offer the fastest route to scale. Relative to roads, homes or hospitals, factories are more contained areas that can be designed around machines. Rather than building technology that mimics humans, firms can create specialised robots to achieve levels of speed, precision and endurance well beyond human limits.

LLMs have demonstrated AI’s ability to process information. But for rich nations seeking greater growth and economic resilience, the bigger opportunity lies in applying that intelligence to production. After all, the greatest technological transformations tend not to replicate what humans can already do, but enable what they cannot.

FT : Heathrow’s £33bn expansion is ‘like a luxury Mercedes’, says airline group

Heathrow’s £33bn expansion is ‘like a luxury Mercedes’, says airline group boss
Head of Star Alliance warns of ‘eye-watering’ costs of third runway project

The head of a major airline alliance that includes Lufthansa and United has likened Heathrow’s third runway plans to an unaffordable luxury car as he urged the airport to cut back some parts of the “eye-watering” project. 

Theo Panagiotoulias, chief executive of Star Alliance, warned that the £33bn runway and terminal risked pricing out airlines.

“I’d love to have a fully specced Mercedes-Benz that’s top of the line. I don’t know if I can necessarily afford it, but I’d like it,” he said. “It’s the same parallel.” 

“I’ve seen some numbers thrown around and they’re eye-watering. We’ve got to be able to have conversations to determine what’s the ‘nice to have’ and what are the things we must have,” he told the Aviation Club in London. 

Star Alliance’s 26 members account for one in six planes that take off from Heathrow. Panagiotoulias’ intervention is the latest in a chorus of airlines warning about the potential cost of building a third runway. 

British Airways and Virgin Atlantic, the airport’s two largest carriers, have both been vocal about the risks of rising costs at the airport. Lufthansa is the third-largest carrier at Heathrow, while United is the fifth. 

Panagiotoulias said the members of the Star Alliance, which also includes Turkish Airlines, Air China and Singapore Airlines, “fully support [a third runway] conceptually”.

He added: “We think expansion is the right thing for the market. It’s the right thing for the airlines, because the airlines are wanting to grow at Heathrow, [but] there’s no simple solution to this. This is going to cost a lot of money.”

Heathrow has long argued that its costs are high because of its proximity to London, as well as the challenges of building in a congested area. Its plan for a third runway includes moving part of the M25. 

BA has publicly backed plans for a shorter, cheaper alternative put forward by Arora Group. 

Panagiotoulias said he needed to see more details from both sides before endorsing one of the projects. 

Earlier this week, the boss of Heathrow lambasted Arora’s plans as “drawings on the back of an envelope”, in his strongest criticism yet of the rival scheme. 

Chief executive Thomas Woldbye said the airport’s own proposal for a 3,500-metre runway that crosses the M25 was the only viable option. 

“There’s no lower-cost option, there are drawings on the back of an envelope - and that’s all there is,” he said, in an unusually direct attack of the alternative scheme. “If we want a third runway we need to get on with the scheme on the table.” 

Arora’s project has been put together by Bechtel, the engineering consultancy behind the Elizabeth Line and the expansion of Perth Airport. 

Its plans would involve building the new runway in two stages: a 2,300-metre runway sandwiched between the M25 and M4 that it says could take many of the narrow-body planes that use Heathrow, with the option to extend over the M25 at a later date. 



Arora has argued that a shorter runway would be operational earlier, and would avoid the costs and complexity of moving the M25 as well as relocating power plants and diverting rivers that are currently in the way of the proposed full-length runway. 

But Woldbye said the company had no intention of building the second phase — and warned that entertaining Arora’s proposal would draw out the process for years. 

In response, the group’s founder Surinder Arora said Heathrow “has a poor track record on development, which has led to it being, year-on-year, the most expensive airport in the world”.

He said that his scheme “can deliver new runway capacity in line with the Government timeframe of 2035, demonstrating just one benefit of competitive bids for expansion”. 

Woldbye’s comments come days after ministers released a draft Heathrow policy that leaves the door open for a competitor to present a runway plan.

The policy allows the runway to be built in stages, but requires any company seeking to win the work to obtain planning permission for the full-length runway. 

The Information : Baidu’s Chip Unit Asked IPO Investors to Buy Its Semiconductor

Baidu’s Chip Unit Asked IPO Investors to Buy Its Semiconductors

The Takeaway
  • Kunlunxin, Baidu’s AI chip unit, targets a $50 billion Hong Kong IPO.
  • IPO investors must commit to buying chips worth 3-7x their subscription.
  • Company seeks SMIC production after missing China’s secure chip list.

In China, chip companies may have found a new clientele for their semiconductors: IPO investors.

Kunlunxin Technology, an AI chip firm majority owned by search engine company Baidu, is planning to go public in Hong Kong at a target valuation of $50 billion, according to a person who participated in a recent investor road show by the company and another person with direct knowledge of Kunlunxin’s plans. And as it lines up a group of investors to take a chunk of the offering, Kunlunxin is prioritizing those who commit to buy its chips, according to two people who participated in a road show.

Such investors include investment funds backed by local governments that have a mandate to invest in AI and semiconductors. They have been asked to buy chips with a value three to seven times the worth of their planned subscription in Kunluxin’s initial public offering shares, according to the people. The timing of the IPO is unclear, as the listing is still going through regulatory approval.

It’s not uncommon in the U.S. for tech firms going public to ask investment banks handling the offering to become customers. SpaceX, for instance, required the banks arranging its recent IPO to buy into its Grok AI model. However, these arrangements are not common in China’s tech IPOs. Moreover, asking investors to buy chips, a huge commitment that would only make sense for companies that operate their own data centers, reflects the fact that it is getting tougher to compete in China’s AI chip market.

China has pushed the local tech industry to become more technologically self-reliant, responding to U.S. chip export restrictions imposed against Beijing in 2022. That has spawned a group of startups like Kunlunxin, all tackling the same part of the AI chip market—inference, the process where AI models generate responses or perform tasks.

Kunlunxin’s target valuation of $50 billion is almost 40% higher than the $36 billion market value commanded by Baidu, which currently owns 58% of Kunlunxin and is traded both on the Nasdaq Stock Exchange and in Hong Kong.

What gives Kunlunxin confidence to shoot for this valuation is that it has a built-in customer in Baidu, which gives it an advantage over most domestic chip firms that are still struggling to establish a significant client base.

Kunlunxin would be the sixth Chinese AI chip designer to go public since December 2025. It couldn’t be learned how much it is looking to raise in the IPO. The company earlier sought to raise as much as $2 billion in the offering, Bloomberg reported in January.

Kunlunxin, named after a 1,864-mile-long mountain range in China revered in Taoist mythology as the “ancestor of mountains,” was founded in 2011. Its long track record, plus the compatibility of its product with Nvidia’s Cuda software system, make it easier for developers to move some tasks away from Nvidia hardware.

Baidu has been training new versions of its Ernie model on Kunlunxin chips, partly replacing Nvidia graphics processing units, The Information reported. In May, Baidu said a key version of Ernie 5.1 had been trained on the chips.

Crowded Market

One reason most Chinese AI chip designers have focused on inference is because they are less difficult to make than chips used in training models.

Kunlunxin’s chips sit mainly in this segment. Its P800 series is used largely for inference, or running already-trained AI models, while also supporting the post-training steps of fine-tuning and parameter tuning, or using small datasets to improve models. Baidu said Kunlunxin’s M100 chip series is ready for large-scale inference this year, and its M300 series will be ready for training and inference in 2027.

Kunlunxin is chasing three kinds of customers: central government–owned enterprises, large internet companies and large AI-model developers, according to two employees. Tencent has become a major external customer, they added.

Kunlunxin’s pitch to state-linked buyers took a hit after China left it off the first “secure and reliable certification list” for domestic AI training and inference chips. The list, released in May, includes hardware from many of the company’s rivals, such as Huawei, MetaX, Moore Threads and Alibaba’s T-Head unit, but not Kunlunxin. This certification is an important procurement guide for government agencies and state-owned companies, which are under pressure to use trusted domestic technology.

Part of the problem is where Kunlunxin makes its chips. The company in the past has relied on Samsung Electronics to manufacture them, undercutting its pitch as a domestic supplier. It is now in talks with China’s state-owned foundry, Semiconductor Manufacturing International Corp., to shift some of the production home, according to two separate Kunlunxin employees.

Alibaba is also planning an IPO for its T-Head unit, Bloomberg reported in January. In June, the unit tripled its registered capital to $148 million, its first increase in more than three years, and underwent a corporate restructuring.

WSJ : The AI Startup Challenging Tesla and Waymo in the Race to Automate Driving

The AI Startup Challenging Tesla and Waymo in the Race to Automate Driving
Wayve is emerging as a go-to partner for traditional automakers trying to keep up with Silicon Valley

British startup Wayve secured deals with Stellantis and Nissan to integrate its AI-driven self-driving technology into vehicles.
Wayve raised $1.5 billion this year, valuing it at $8.6 billion, and plans a London robotaxi trial with Uber this summer.
Wayve’s “end-to-end” AI system, using cameras and radar, contrasts with traditional rules-based and hybrid self-driving methods.

Earlier this year, a team of engineers flew to Detroit on a secret mission to make Jeep Grand Cherokees drive themselves.

A few weeks later, prototypes were convincing enough for Jeep owner Stellantis STLA -1.05%decrease; down pointing triangle to take investors on test rides around Motor City’s suburbs and commit to rolling out the technology across America.

The self-driving know-how was developed by Wayve, a British startup that is emerging as an unlikely front-runner in a hard-fought race with U.S. behemoths Tesla TSLA 1.22%increase; up pointing triangle and Google owner Alphabet to bring autonomous vehicles to the masses.

Stellantis said last month it would use Wayve’s “AI driver” across its brands—which also include Chrysler, Dodge and Ram—from 2028. The deal followed a similar agreement with Nissan that envisaged a potentially earlier launch in Japan.

The technology promises hands-free navigation on both highways and urban roads as long as the human driver stays attentive and ready to take over as a backup. Tesla pioneered this approach with the feature it now sells as “Full Self-Driving (Supervised).”

But while Tesla bakes that technology into its own vehicles, Wayve thinks selling its systems as an off-the-shelf solution to other automakers could be where the real money is in self-driving.

“Not everyone wants to buy a Tesla,” said Wayve co-founder Alex Kendall. “Our opportunity is to bring this technology to every other automaker.”

In late 2023, Tesla pivoted to an AI-based system similar to Wayve’s, where driving decisions are made by a model—akin to those behind chatbots such as ChatGPT—trained on reams of driving videos.

Around that time, Wayve started to attract wider attention. In 2024, it raised $1.05 billion in a funding round led by SoftBank and supported by Microsoft, Nvidia and Uber. This year it got a further $1.5 billion at an $8.6 billion valuation, including from Stellantis, Mercedes-Benz and Nissan and chip makers AMD, Arm and Qualcomm.

“I’ve always believed that intelligence would be the way that robots work in the future,” said Kendall. “But the market’s conviction around that, I think, has come together really in the last couple of years.”

The seed of Stellantis’s bet on Wayve was sown on a test ride in London, when the technology wowed the carmaker’s chief technology officer, Ned Curic, with its humanlike way of edging into flowing traffic.

When Curic later took responsibility for Stellantis’s automated-driving program in early 2025, he shelved the company’s internal development, which followed the traditional rules-based approach, and started scouting for AI-savvy partners.

Curic challenged Kendall to show his technology would work on an unfamiliar Jeep in unfamiliar Michigan.

It performed “immensely well,” Curic said, even in a blizzard and after his team asked Wayve to shift sensors from an unsightly roof rack to the wing mirrors.

Still, the “end-to-end” AI approach championed by Wayve—where a single model takes inputs from sensors and delivers driving decisions as outputs—remains controversial, particularly when it comes to removing human backup drivers.

Unlike with chatbots, verification and compliance are key considerations in the safety-first automotive industry. To their critics, end-to-end AI models are “black boxes” that lack the transparency required to comply with strict regulations.

Robotaxi pioneers Waymo in the U.S. and Baidu’s Apollo Go in China, which already operate without backup drivers, use hybrid approaches that blend AI models with high-definition maps and traditional rules that function as guardrails.

Supporters of AI systems tout their greater capacity to generalize from one driving situation to another, as human drivers can, making them easier to roll out at scale than hybrid systems.

Wayve’s approach doesn’t require elaborate mapping of streets before it can drive around them. Last year, the company went on a tour of 506 cities across the world to prove the point, albeit with a safety driver behind the wheel.

Kendall said Wayve had spent years working out how to make its AI driver compliant and that its solution had been accepted by the safety experts at Stellantis, Nissan and other automakers with which it is in talks.

A similar debate focuses on the sensors required to run a self-driving vehicle. Wayve, like Tesla, says its technology can do without the expensive laser-based lidar units that most of their peers rely on. Its London test vehicles feature only cameras and radar.

But traditional automakers are more conservative when it comes to deploying the technology at commercial scale. Stellantis and Nissan are both developing robotaxis using Wayve’s AI driver that include lidar in the sensor suite.

While Stellantis used six cameras for the jury-rigged Grand Cherokees, the vehicles available for purchase in 2028 will sport 10 cameras and five radars, giving Wayve’s AI more data to rely on.

“We feel it’s best for us to make sure the system is consistent, safe and seeing everything,” said Doug Wellman, who leads assisted driving at the automaker’s U.S. headquarters.

The flexibility of Wayve’s AI-first approach could attract more traditional automakers looking to share the investment burden of keeping up with Silicon Valley’s advances in vehicle autonomy. The startup says it is in talks with most of the world’s top-10 players.

Kendall said being based in London, far from the autonomous-vehicle mainstream, helped Wayve stick to its technological guns.

“I want to be building something contrarian,” he said. “We’ve completely gone against the grain of the Silicon Valley bubble for self-driving and I think we might have been swallowed up if we were starting there.”