Connect with us

Published

on

Blocked from the U.S. by tariffs, Chinese electric vehicle makers have looked elsewhere to sell their high-tech cars. But as Mexico has emerged as a hot spot for Chinese EVs, Washington officials worry the country may be used as a “backdoor” to the U.S. market.  

Last year, China was the leading car supplier to Mexico, exporting $4.6 billion worth of vehicles to the country, according to the Mexican Ministry of Economy. Even customers wary of EVs have been won over by affordable price tags. Tesla rival BYD sells its Dolphin Mini in Mexico for around 398,800 pesos, or about $21,300, a little over half the price of the cheapest Tesla. 

“The Chinese automakers came to the country very aggressively,” said Juan Carlos Baker, former Mexican deputy minister for international trade. “They have very good promotions. It’s a good product that sells at a very reasonable price.” 

Some Chinese EV makers, including BYD, have been looking for a further foothold in North America by exploring factory sites in the Mexican states of Durango, Jalisco and Nuevo Leon. The foreign investment would be an economic boost for Mexico. BYD has claimed that a plant there would create around 10,000 jobs.  

But U.S. officials worry this could be a part of a larger strategy by Chinese automakers to skirt trade restrictions and enter the American market.  

“Mexico is an attractive production platform, not only for Chinese companies, but for other companies as well, in part because of that free trade access that it has to the American market,” said Scott Paul, president of the Alliance for American Manufacturing. “And it can do something that in trade terms is called circumvention.” 

That free trade access is part of the United States-Mexico-Canada Agreement (USMCA), a revised iteration of the North American Free Trade Agreement (NAFTA) that removed tariffs on many goods traded between the North American countries starting in 2018. Under the agreement, if a foreign auto company manufactures in either Canada or Mexico and can prove that the building materials are sourced locally, the goods can be exported to the U.S. virtually duty-free.  

“We’ve seen China do this in other types of manufacturing as well, from appliances to auto parts to steel,” said Paul. “For more than a decade now, China, the United States have been playing a high-stakes game of whack-a-mole when it comes to trade policy tariffs.”

While meeting the USMCA requirements is difficult, the potential scenario terrifies U.S. lawmakers and auto companies.

“If [Chinese EV makers] are able to set up in Mexico, they would definitely pose an imminent threat to American automakers, if for no other reason, because their costs would be lower,” said Michael Dunne, CEO of Dunne Insights.  

In May, President Joe Biden announced a 100% tariff on Chinese EVs.

“We [the U.S.] are just starting to scale up our EV industry, so it’s what I call an ‘infant industry,'” said Paul. “And like any infant, it’s at a very delicate time in terms of development and has to be massively protected.” 

Experts say pressure from the U.S. leaves Mexico in a difficult position of maintaining its crucial relationship with America without being overly friendly to Chinese investment.  

Watch the video to learn more about how Mexico has become a hot spot for Chinese auto companies and how the next administration may impact EV trade policies.  

Continue Reading

Technology

Nvidia calls China’s DeepSeek R1 model ‘an excellent AI advancement’

Published

on

By

Nvidia calls China’s DeepSeek R1 model ‘an excellent AI advancement’

Jensen Huang, co-founder and chief executive officer of Nvidia Corp., during a news conference in Taipei, Taiwan, on Tuesday, June 4, 2024. Nvidia is still working on the certification process for Samsung Electronics Co.’s high-bandwidth memory chips, a final required step before the Korean company can begin supplying a component essential to training AI platforms. 

Annabelle Chih | Bloomberg | Getty Images

Nvidia called DeepSeek’s R1 model “an excellent AI advancement,” despite the Chinese startup’s emergence causing the chip maker’s stock price to plunge 17% on Monday.

“DeepSeek is an excellent AI advancement and a perfect example of Test Time Scaling,”  an Nvidia spokesperson told CNBC on Monday. “DeepSeek’s work illustrates how new models can be created using that technique, leveraging widely-available models and compute that is fully export control compliant.”

The comments come after DeepSeek last week released R1, which is an open-source reasoning model that reportedly outperformed the best models from U.S. companies such as OpenAI. R1’s self-reported training cost was less than $6 million, which is a fraction of the billions that Silicon Valley companies are spending to build their artificial-intelligence models. 

Nvidia’s statement indicates that it sees DeepSeek’s breakthrough as creating more work for the American chip maker’s graphics processing units, or GPUs. 

Read more DeepSeek coverage

“Inference requires significant numbers of NVIDIA GPUs and high-performance networking,” the spokesperson added. “We now have three scaling laws: pre-training and post-training, which continue, and new test-time scaling.”

Nvidia also said that the GPUs that DeepSeek used were fully export compliant. That counters Scale AI CEO Alexandr Wang’s comments on CNBC last week that he believed DeepSeek used Nvidia GPUs models which are banned in mainland China. DeepSeek says it used special versions of Nvidia’s GPUs intended for the Chinese market.

Analysts are now asking if multi-billion dollar capital investments from companies like Microsoft, Google and Meta for Nvidia-based AI infrastructure are being wasted when the same results can be achieved more cheaply. 

Earlier this month, Microsoft said it is spending $80 billion on AI infrastructure in 2025 alone while Meta CEO Mark Zuckerberg last week said the social media company planned to invest between $60 to $65 billion in capital expenditures in 2025 as part of its AI strategy. 

“If model training costs prove to be significantly lower, we would expect a near-term cost benefit for advertising, travel, and other consumer app companies that use cloud AI services, while long-term hyperscaler AI-related revenues and costs would likely be lower,” wrote BofA Securities analyst Justin Post in a note on Monday.

Nvidia’s comment also reflects a new theme that Nvidia CEO Jensen Huang, OpenAI CEO Sam Altman and Microsoft CEO Satya Nadella have discussed in recent months.

Much of the AI boom and the demand for Nvidia GPUs was driven by the “scaling law,” a concept in AI development proposed by OpenAI researchers in 2020. That concept suggested that better AI systems could be developed by greatly expanding the amount of computation and data that went into building a new model, requiring more and more chips.

Since November, Huang and Altman have been focusing on a new wrinkle to the scaling law, which Huang calls “test-time scaling.” 

This concept says that if a fully trained AI model spends more time using extra computer power when making predictions or generating text or images to allow it to “reason,” it will provided better answers than it would have if it ran for less time. 

Forms of the test-time scaling law are used in some of OpenAI’s models such as o1 as well as DeepSeek’s breakthrough R1 model.

WATCH: DeepSeek challenging sense of U.S. exceptionalism priced into markets, fund manager says

Continue Reading

Technology

Apple turns its AI on by default in latest software update

Published

on

By

Apple turns its AI on by default in latest software update

Tim Cook, chief executive officer of Apple Inc., greets customers during the first day of in-store sales of Apple’s latest products at Apple’s Fifth Avenue store in New York, US, on Friday, Sept. 20, 2024. 

Victor J. Blue | Bloomberg | Getty Images

Apple on Monday released a software update for iPhones, iPads and Macs that turns Apple Intelligence on by default for users with supported devices.

The updates, iOS 18.3, iPadOS 18.3 and macOS Sequoia 15.3, also disable AI summaries for news apps, which have gained a reputation for twisting news push notifications to display inaccurate facts.

The release is a milestone in the rollout of Apple Intelligence, the company’s suite of artificial intelligence features. Apple Intelligence is a critical service for Apple as it seeks to distinguish its products from competitors with an AI system integrated into iPhones and its other devices.

While Apple Intelligence is already featured in the company’s marketing for the latest iPhones, the rollout has been deliberate and limited. Apple says that is to allow it to test new features and make sure it has enough server capacity. The entire Apple Intelligence suite is still officially in beta, and it’s only available in a handful of English-speaking countries.

Apple’s move to turn Apple Intelligence on by default is a step toward a more complete rollout of the feature. Previously, users with supported iPhones — models that were released in 2024 as well as 2023 “Pro” level models — were prompted to turn on Apple Intelligence when setting up their phone, a process that included downloading AI models from the internet and some installation.

With the latest updates, Apple Intelligence will be turned on by default when the latest software update is installed, expanding the number of users who are exposed to the software. Apple Intelligence is marketed as being able to rewrite text, generate images, and summarize long emails and message threads.

“For users new or upgrading to iOS 18.3, Apple Intelligence will be enabled automatically during iPhone onboarding,” Apple said in the developer release notes for the update.

Users will have to navigate to the Apple Intelligence page in the Settings app to turn it off, according to Apple.

The latest software update also includes one of the most notable examples so far of Apple rolling back an AI feature after it generated controversy and subpar results. Apple joins Google and Microsoft as companies that were forced to recall new AI features after they generated harmful content or “hallucinations” that weren’t based in reality.

Apple Intelligence can take stacks of notifications and simplify them into a three-sentence notification. However, the BBC and other news outlets discovered in December that the feature can twist news headlines into inaccurate information.

Earlier this month, Apple’s system conflated notifications from BBC’s sports app to say that “Brazilian tennis player, Rafael Nadal, comes out as gay.” Nadal is Spanish and is married to Maria Francisca Perello.

The latest update disables Apple Intelligence for news and entertainment apps. Apple Intelligence also has been updated to show any notifications generated by its AI in italics, signifying which notifications are created by generative AI and which were pushed by the app itself.

“We’re pleased that Apple has listened to our concerns and is pausing the summarization feature for news,” a BBC spokesperson told CNBC.

Apple told CNBC earlier this month that its notification summaries for news apps would return in a future update.

WATCH: Apple’s superficial problem is there’s not enough demand, says Jim Cramer

Apple's superficial problem is there's not enough demand, says Jim Cramer

Continue Reading

Technology

DeepSeek hit with large-scale cyberattack, says it’s limiting registrations

Published

on

By

DeepSeek hit with large-scale cyberattack, says it's limiting registrations

Dado Ruvic | Reuters

DeepSeek on Monday said it would temporarily limit user registrations “due to large-scale malicious attacks” on its services, though existing users will be able to log in as usual.

The Chinese artificial intelligence startup has generated a lot of buzz in recent weeks as a fast-growing rival to OpenAI’s ChatGPT, Google’s Gemini and other leading AI tools.

Earlier on Monday, DeepSeek took over rival OpenAI’s coveted spot as the most-downloaded free app in the U.S. on Apple‘s App Store, dethroning ChatGPT for DeepSeek’s own AI Assistant. It helped inspire a significant sell-off in global tech stocks.

Buzz about the company, which was founded in 2023 and released its R1 model last week, has spread to tech analysts, investors and developers, who say that the hype — and ensuing fear of falling behind in the ever-changing AI hype cycle — may be warranted. Especially in the era of the generative AI arms race, where tech giants and startups alike are racing to ensure they don’t fall behind in a market predicted to top $1 trillion in revenue within a decade.

Read more CNBC reporting on AI

DeepSeek reportedly grew out of a Chinese hedge fund’s AI research unit in April 2023 to focus on large language models and reaching artificial general intelligence, or AGI — a branch of AI that equals or surpasses human intellect on a wide range of tasks, which OpenAI and its rivals say they’re fast pursuing.

The buzz around DeepSeek especially began to spread last week, when the startup released R1, its reasoning model that rivals OpenAI’s o1. It’s open-source, meaning that any AI developer can use it, and has rocketed to the top of app stores and industry leaderboards, with users praising its performance and reasoning capabilities.

The startup’s models were notably built despite the U.S. curbing chip exports to China three times in three years. Estimates differ on exactly how much DeepSeek’s R1 costs, or how many graphics processing units went into it. Jefferies analysts estimated that a recent version had a “training cost of only US$5.6m (assuming US$2/H800 hour rental cost). That is less than 10% of the cost of Meta‘s Llama.”

But regardless of the specific numbers, reports agree that the model was developed at a fraction of the cost of rival models by OpenAI, Anthropic, Google and others.

As a result, the AI sector is awash with questions, including whether the industry’s increasing number of astronomical funding rounds and billion-dollar valuations is necessary — and whether a bubble is about to burst.

Don’t miss these insights from CNBC PRO

Continue Reading

Trending