Technology

Baidu, once China’s generative AI leader, is battling to regain its position

Published

7 months ago

March 18, 2025

admin

Pictured here is the Ernie bot mobile interface, with the Baidu search engine home page in the background.

Future Publishing | Future Publishing | Getty Images

Chinese tech giant Baidu has released two new free-to-use artificial intelligence models as it vies to regain its leading position in the country’s fiercely competitive AI space.

The Baidu models launched Sunday included the company’s first reasoning-focused model, and come ahead of plans to move toward an open-source strategy.

However, experts told CNBC that while the release of the models is a positive development for Baidu, they also highlight how it is playing catch up as its Ernie bot — one of China’s earliest versions of a ChatGPT-like chatbot — struggles to gain widespread adoption.

“The new models make Baidu more competitive since the company has been lagging behind in a reasoning model release,” Lian Jye Su, chief analyst at Omdia, told CNBC.

A reasoning model is a large language model that breaks down tasks into smaller pieces and considers multiple approaches before generating a response. It is designed to process complex problems in a similar way to humans.

Chinese startup DeepSeek upended the global AI race and transformed China’s ecosystem in January when it released its R1 reasoning model, which rivaled American competitors despite costing a fraction of the price.

Baidu has said its new ERNIE X1 reasoning model “delivers performance on par with DeepSeek R1 at only half the price,” and has “stronger understanding, planning, reflection, and evolution capabilities.” CNBC has not been able to independently verify this claim.

According to Wei Sun, principal analyst of artificial intelligence at Counterpoint Research, Baidu’s future competitiveness could hinge on whether its new models deliver on the promised performance and cost advantages.

“Baidu is clearly in catch-up mode, largely due to its slow innovation pace and underestimating rapid shifts in market dynamics,” Sun said.

What happened?

Baidu rolled out its first generative AI platform to the public in 2023, giving China one of its first answers to OpenAI’s popular AI chatbot ChatGPT.

However, despite initial momentum, Baidu’s Ernie product has since been eclipsed by competitors including startups as well as large-tech companies such as Alibaba and ByteDance.

Experts list a number of reasons for Baidu’s struggles and slow rate of innovation.

“Baidu fell behind when they tried to build proprietary models and compete for funding for AI,” Ray Wang, principal analyst and founder of Constellation Research, told CNBC. He added that the company has also suffered from recent government crackdowns and was distracted by “regulatory nonsense.”

CFOTO | Future Publishing | Getty Images

Proprietary models keep their source code and underlying architecture confidential, in contrast to models from the likes of DeepSeek, whose source code is made freely available on the open web for possible modification and redistribution.

“Using a closed-source approach means that [Baidu] was training its model from scratch whereas the open-source models were able to leverage certain parts that were communal to developers,” said Kai Wang, a senior equity analyst for Morningstar.

Baidu, however, said last month that it would make its next-generation AI model Ernie open-source from June 30, according to Reuters.

“Baidu has always been very supportive of its proprietary business model and was vocal against open source, but disruptors like DeepSeek have proven that open source models can be as competitive,” said Omdia’s Su.

He added that Baidu is “merely following the footstep” of its biggest competitors in China, namely Alibaba, DeepSeek, and Tencent, which have all now released open-source models.

Baidu’s advantages

According to Morningstar’s Wang, AI players like Baidu are coming up with new models every month or so, with competitors constantly trying to outperform each advancement.

However, “the differentiator in the Chinese models remains the data used for the models, and the application use-case,” Wang said.

In these areas, experts see some key advantages for Baidu, which operates a wide range of popular applications and services, including its flagship search engine, which offers access to AI tools, as well as its online forum app, Baidu Tieba.

“Due to its large user base in the consumer space, particularly in AI search and recommendation systems, the company can still maintain its market leadership position as long as it keeps up with the innovation cycle,” said Omdia’s Su.

During an earnings call late last year, Baidu CEO Robin Li said he expects generative AI to turn Baidu search into the “new killer app in the age of AI.”

“The secret to AI is data, chips, good math, and cheap energy … You have to have data to play. They have the data,” said Constellation Research’s Wang.

Correction: This article has been updated to state that Baidu’s new AI models are free to use.

Technology

Microsoft CEO Satya Nadella’s annual pay jumps to $96.5 million

Published

2 hours ago

October 22, 2025

admin

Microsoft CEO Satya Nadella's annual pay jumps to .5 million

Satya Nadella, CEO of Microsoft, speaking on CNBC’s “Squawk Box” outside the World Economic Forum in Davos, Switzerland, on Jan. 22, 2025.

Gerry Miller | CNBC

Microsoft CEO Satya Nadella is getting a big bump in his compensation, as the company’s stock price has continued to rally, propelled by the boom in artificial intelligence.

Nadella’s total pay for fiscal 2025 climbed 22% to $96.5 million from $79.1 million last year, Microsoft said in a proxy filing after the close of regular trading on Tuesday. That includes more than $84 million in stock awards and over $9.5 million in Nadella’s cash incentives.

The pay plan is largely tied Microsoft’s share performance. So far in 2025, Microsoft’s stock price has risen by 23%, topping the S&P 500’s 15% gain. The shares have more than doubled in valued over the past three years.

Microsoft is scheduled to report results for the fiscal first quarter next week. In its fourth-quarter disclosure in July, the company reported better-than-expected earnings and revenue, with sales climbing 18%, the fastest growth in more than three years. Microsoft Azure business is driving expansion as companies’ cloud infrastructure needs grow to meet AI demand.

In fiscal 2024, Nadella’s pay jumped 63% from 48.5 million the prior year, with 90% of his compensation coming from stock awards. Nadella was eligible for a $10.66 million cash incentive last year, but he asked the board’s compensation committee to reduce that number to $5.2 million as a result of a series of cyberattacks that the company endured.

Despite Microsoft’s strong financial and stock performance, the company has seen turmoil among its workforce in recent months. In July, Nadella penned a memo to employees saying that the company’s elimination of more than 15,000 employees in 2025 had “been weighing heavily” on him.

Microsoft has also terminated several activist employees who protested the company’s work with the Israeli military.

WATCH: Microsoft is trending toward a $5T market cap, says Wedbush’s Dan Ives

Technology

Meta partners with Blue Owl Capital on $27 billion AI data center project

Published

4 hours ago

October 21, 2025

admin

Meta partners with Blue Owl Capital on billion AI data center project

Mark Zuckerberg, chief executive officer of Meta Platforms Inc., during the Meta Connect event in Menlo Park, California, US, on Wednesday, Sept. 17, 2025.

David Paul Morris | Bloomberg | Getty Images

Meta said Tuesday that it formed a joint venture agreement with Blue Owl Capital in a deal worth $27 billion to fund and develop the social media company’s massive Hyperion data center in rural Louisiana.

As part of the deal, the asset management firm will own 80% of the joint venture, while Meta will retain a 20% stake and oversee the construction and property management services of the data center, which is being built in Richland Parish, Louisiana. Blue Owl contributed about $7 billion in cash as part of the joint venture, while Meta received a one-time payout of $3 billion.

The partnership provides the “the speed and flexibility” Meta needs to build the data center and support its “long-term AI ambitions,” the social media company said in a statement.

Meta in December announced that it chose Louisiana to host what would be its largest data center. Construction of that facility, which is being built on a site the size of roughly 1,700 football fields, is expected to finish by 2030.

Local utility Entergy told CNBC in June that the new data center could consume about twice as much electricity as the city of New Orleans on a peak day.

Meta has been spending heavily on artificial intelligence amid a broader race with other tech giants like Alphabet and ChatGPT-maker OpenAI, which are also developing gigantic data centers to power future AI models.

OpenAI, Oracle and Softbank in January formed the Stargate joint venture that will see the companies invest $500 billion to develop data centers over the coming years. The first Stargate data center site came online in September 180 miles west of Dallas in Abilene, Texas.

Last week, Google said that it would invest $15 billion on a data center project in southern India that will be the search giant’s largest AI hub in the world outside of the U.S.

WATCH: A rotation out of the US is a bet against the AI trade.

A rotation out of the US is a bet against the AI trade: Morgan Stanley's Andrew Slimmon

Technology

CNBC exclusive: First look at Shield AI’s new AI-piloted military fighter drone

Published

4 hours ago

October 21, 2025

admin

CNBC exclusive: First look at Shield AI's new AI-piloted military fighter drone

Shield AI is trying to shake up the defense industry.

The company, which is valued at $5.3 billion after securing $240 million in its latest round of funding, is set to unveil its next generation, autonomous fighter jet known as the X-Bat on Wednesday.

CNBC got exclusive access to the company’s headquarters ahead of the launch.

Shield AI says the unmanned aircraft has a jet engine, will have a 2,000 mile range, can fly up to 50,000 feet and has the ability to take off and land vertically, enabling it to operate in remote locations without a runway — like on a ship in the middle of the ocean.

The X-Bat will be piloted by an AI software developed by Shield AI called Hivemind. The company is now hinging a lot of its future on artificial intelligence development.

“The software is a cornerstone and foundation for everything we do,” said Shield AI CEO Gary Steele. “It will ultimately be the long term growth driver of this business because it enables the development of this next generation aircraft.”

Shield AI ranks number 38 on the 2025 CNBC Disruptor 50 list.

X-Bat combines some of the defense industry’s most advanced technologies into one fighter jet. There have been experimental aircrafts built as early as the 1950s with vertical takeoff and landing capabilities but they required pilots. Shield AI has also used Hivemind to fly the F-16, one of the most widely used modern fighter jets, autonomously.

“But those two things — AI piloted and vertical takeoff launch and land — have never come together in the form of a next generation aircraft,” said Brandon Tseng, Shield AI president and co-founder.

The company says its on track to produce the X-Bat for around $27 million, which is a fraction of what advanced military aircrafts typically cost. For example, the F-35 fighter jet that’s currently in use by the U.S. government and allies, costs more than $100 million to produce.

Unlike Shield AI’s previous aircrafts, the X-Bat is designed for combat and can be equipped with missiles.

“We fundamentally believe we can save service members’ lives by reducing the risk that you have of putting people in danger,” ” said Steele. “What I’m particularly excited about is the mission we’ve been on, and the opportunity that it unlocks from a business perspective.”

Shield AI has been around since 2015 and has already landed some major defense contracts. In 2024, the company secured a nearly $200 million contract with the U.S. coast guard for a drone it produces called the V-Bat.

But the startup is still proving itself in a competitive industry. Although it has grown quickly, the company is relatively small compared to defense primes like Lockheed Martin and Northrop Grumman and its biggest startup competitor, Anduril, which is valued at over $30 billion.

Despite generating billions of dollars in revenue, Shield is not yet profitable. In 2023, Forbes reported that the company was on track to reach profitably by 2025. However, those targets were thrown off track when a U.S. service member had his fingers partially severed during a Shield AI drone landing demonstration in 2023.

“Through that process, there were some loss of confidence from customers,” Steele said. “But I think we’ve done a phenomenal job of recovering from that and rebuilding momentum. And today as we sit here, we’re very confident in our ability to deliver great products that are safe.”

Drones have been used in war zones as early as World War I, but their prevalence has grown dramatically in recent years. The war in Ukraine has helped show the general public the scale and prevalence of drone usage on the battlefield today.

“What we see from the war in Ukraine and the Middle East, they are tactically, operationally and strategically absolutely important weapons,” said Oleksandra Molloy, a drone expert and senior aviation lecturer at UNSW Canberra. “We have seen a lack of those systems from the U.S., and particularly, we have not really seen the presence of many American companies in the real battlefield.”

But the U.S. government is now trying to change that. In June 2025, President Donald Trump issued an executive order called Unleashing American Drone Dominance that aims to accelerate commercialization of drone technologies and integrate them into the National Airspace System. Although no direct dollar amount was attached to that order, the Big Beautiful Bill has allocated billions of dollars in unmanned aerial systems and AI development.

“We have to empower the defense industrial base with the exact same development tools, infrastructure and pipelines that Shield AI has used to make AI autonomy,” Tseng said. “We work directly with the major defense prime contractors of the world. We want to see them wildly successful building AI and autonomy, because at the end of the day, that’s what the warfighter needs. That’s what the United States and our allies need.”

Watch the video to learn more about how Shield AI is making a name for itself in the defense sector.