Connect with us

Published

on

We worked hard to keep up with the 'various abilities' of ChatGPT, says Naver Cloud CEO

South Korean internet giant Naver released its own generative artificial intelligence tool on Thursday, joining other companies in launching large language models to compete with OpenAI’s ChatGPT.

Naver is dubbed the Google of South Korea, as the operator of the country’s top search engine. The company said its large language model HyperClova X can improve search as well as marketing and shopping experiences for advertisers and users.

HyperClova X powers a slate of new applications Naver announced Thursday including generative AI search service Cue: and chat app Clova X.

Naver said Clova X can increase work productivity by summarizing documents and running language translation. Users can also use its AI writing tool to draft emails and blog posts. Naver said Clova X will start beta testing Thursday.

“In terms of productivity, this will save 10% to 50% of time spent on those tasks,” Kim Yuwon, CEO of Naver Cloud, said in an exclusive interview with CNBC aired on Thursday.

An AI boom led by ChatGPT — the AI chatbot which has gone viral for its ability to generate humanlike responses to user prompts — has sparked clones around the world such as Microsoft’s Bard, Meta’s Llama, Alibaba’s Tongyi Qianwen and Baidu’s Ernie Bot.

“ChatGPT is a very challenging competitor. I think it’s an honest confession that we worked hard to keep up with the various abilities of ChatGPT,” said Kim.

“We tried to improve our ability to understand and respond to the intentions of the user. Korean proficiency is HyperClova’s competitive edge, but we tried to improve various language capabilities such as English to match the level of global competitors,” said Kim.

Naver’s Korea-listed shares jumped 6.03% Thursday after the announcement.

James Lim, senior research analyst at Dalton Investments, said that Naver is “still behind” other global players.

“There are really global top tier players and how do you get to a large number of parameters? First, you have to make that much of a investment and that’s where I think Naver might feel a little bit challenged because you’re going [against] the global leaders,” Lim told CNBC’s “Squawk Box Asia” on Thursday.

Naver CEO Choi Soo-yeo said that it has invested a total of 1 trillion Korean won ($756.8 million) in the last three to four years toward the research and development of AI.

Developing an AI chip with Samsung

Kim said Naver’s AI investment is “reaching the level of full-fledged commercialization” and the firm needs a “low-cost and efficient semiconductor chip to improve our service quality.”

“We’ve done the large language model artificial intelligence training but to apply it to search and shopping, the cost for this service will increase tremendously. We came to the conclusion that it would be difficult to provide quality service to our users without reducing this cost,” said Kim.

Kim said that Naver is co-developing “low-cost and efficient” AI semiconductor chips with Samsung Electronics, the world’s largest dynamic random-access memory chipmaker. SK Hynix, another South Korean DRAM chipmaker, is the second-largest DRAM chipmaker globally.

“So this is not just about further commercializing our technology. We are doing this because we think we need such a low-cost and efficient semiconductor chip to improve our service quality,” said Kim.

Analyst discusses Naver and SK Telecom and their development of AI services for enterprise customers

Naver expects to further refine and develop its generative AI model, CEO Choi said earlier this week.

“We are excited about the potential value that a generative AI-powered search, marketing and shopping experience can create for our users and advertisers,” said Choi.

“Naver has already historically showed that they can utilize the AI technology to improve their services such as increasing the efficiency of their e-commerce recommendations, and I think that will accelerate with the adoption of their more newly developed AI technology,” said Lim of Dalton Investments.

Continue Reading

Technology

Nvidia sending 18,000 of its top AI chips to Saudi Arabia

Published

on

By

Nvidia sending 18,000 of its top AI chips to Saudi Arabia

Tareq Amin, CEO of Humain, and Jensen Huang, CEO of NVIDIA, attend the Saudi-U.S. Investment Forum, in Riyadh, Saudi Arabia May 13, 2025.

Hamad I Mohammed | Reuters

Nvidia will sell over 18,000 of its latest artificial intelligence chips to Saudi Arabian company Humain, CEO Jensen Huang announced on Tuesday.

The announcement was made as part of a White House-led trip to the region that includes President Donald Trump and other top CEOs.

The cutting-edge Blackwell chips will be used in a 500 megawatt data center in Saudi Arabia, according to remarks at the Saudi-U.S. Investment Forum in Riyadh on Tuesday. Nvidia said its first deployment will use its GB300 Blackwell chips, which are among Nvidia’s most advanced AI chips at the moment, and which were only officially announced earlier this year.

Tuesday’s announcement underscores the importance of Nvidia’s chips as a bargaining tool for the Trump administration as countries around the world clamor for the devices, which are used to train and deploy advanced AI software such as ChatGPT.

Read more CNBC tech news

“I am so delighted to be here to help celebrate the grand opening, the beginning of Humain,” Huang said. “It is an incredible vision, indeed, that Saudi Arabia should build the AI infrastructure of your nation so that you could participate and help shape the future of this incredibly transformative technology.”

Nvidia shares rose 4% in trading on Tuesday.

Last week, the Department of Commerce said that it was going to scrap what it called President Joe Biden’s rule, and implement a “much simpler rule.” Nvidia has also been required to seek an export license for its AI chips since 2023 because of national security concerns. 

Humain will be owned by Saudi Arabia’s Public Investment Fund, and will work on developing AI models as well as building data center infrastructure, according to a press release. Humain’s plans eventually include deploying “several hundred thousand” Nvidia GPUs. 

“Saudi Arabia is rich with energy, transforming the energy through this giant versions of these Nvidia AI supercomputers, which are essentially AI factories,” Huang said.

Continue Reading

Technology

Microsoft is cutting 3% of its workforce

Published

on

By

Microsoft is cutting 3% of its workforce

Microsoft CEO Satya Nadella leaves after attending a meeting with Indonesian President Joko Widodo at the Presidential Palace in Jakarta, Indonesia, on April 30, 2024.

Willy Kurniawan | Reuters

Microsoft on Tuesday said that it’s laying off 3% of employees across all levels, teams and geographies.

“We continue to implement organizational changes necessary to best position the company for success in a dynamic marketplace,” a Microsoft spokesperson said in a statement to CNBC.

The company reported better-than-expected results, with $25.8 billion in quarterly net income, and an upbeat forecast in late April.

Microsoft had 228,000 employees worldwide at the end of June, meaning that the move will affect thousands of employees.

It’s likely Microsoft’s largest round of layoffs since the elimination of 10,000 roles in 2023. In January the company announced a small round of layoffs that were performance-based. These new job cuts are not related to performance, the spokesperson said.

Read more CNBC tech news

One objective is to reduce layers of management, the spokesperson said.

Last week cybersecurity software provider CrowdStrike announced it would lay off 5% of its workforce.

In January, Microsoft CEO Satya Nadella told analysts that the company would make sales execution changes that led to lower growth than expected in Azure cloud revenue that wasn’t tied to artificial intelligence. Performance in AI cloud growth outdid internal projections.

“How do you really tweak the incentives, go-to-market?” Nadella said. “At a time of platform shifts, you kind of want to make sure you lean into even the new design wins, and you just don’t keep doing the stuff that you did in the previous generation.”

On Monday, Microsoft shares stopped trading at $449.26, the highest price so far this year. They closed at a record $467.56 last July.

WATCH: Microsoft will continue as the best performing mega-cap stock in 2025: D.A. Davidson’s Gil Luria

Microsoft will continue as the best performing mega-cap stock in 2025: D.A. Davidson's Gil Luria

Continue Reading

Technology

Hinge Health aims to raise up to $437 million in IPO, pricing at $28 to $32 per share

Published

on

By

Hinge Health aims to raise up to 7 million in IPO, pricing at  to  per share

Hinge Health co-founders Gabriel Mecklenburg (left) and Daniel Perez (right).

Courtesy of Hinge Health

Hinge Health said in a filing on Tuesday that it plans to raise up to $437 million in its upcoming initial public offering.

The digital physical therapy startup filed its initial prospectus in March, and it updated the document with an expected pricing range for its Class A common stock of $28 to $32 per share. Hinge said it plans to sell about 13.7 million shares in the offering.

Based on the number of Class A and Class B shares outstanding after the offering, the deal would value the company at $2.42 billion in the middle of the range, though that number could be higher on a fully diluted basis.

Hinge, founded in 2014, uses software to help patients treat acute musculoskeletal injuries, chronic pain and carry out post-surgery rehabilitation remotely. The company was co-founded by CEO Daniel Perez and Executive Chairman Gabriel Mecklenburg, who have both experienced personal struggles with physical rehabilitation.

More CNBC health coverage

Three weeks after Hinge filed its initial prospectus, President Donald Trump announced a sweeping tariff policy that plunged U.S. markets into turmoil. That volatility has caused several companies, including online lender Klarna and ticket marketplace StubHub, to delay their long-awaited IPOs.

Hinge is forging ahead anyway, and a second digital health startup, virtual chronic care company Omada Health, filed to go public on Friday. Both IPOs will be closely watched by the digital health sector, which has been mostly devoid of public offerings since 2021.

During its first quarter, Hinge said that revenue climbed 50% to $123.8 million, up from $82.7 million during the same period last year. Hinge reported $117.3 million in revenue during its fourth quarter, up 44% from the same period in 2023. 

The company plans to trade on the New York Stock Exchange under the ticker symbol “HNGE.”

Hinge has raised more than $1 billion from investors including Tiger Global Management and Coatue Management, and it boasted a $6.2 billion valuation as of October 2021, the last time the company raised outside funding. The biggest institutional shareholders are venture firms Insight Partners and Atomico, which own 19% and 15% of the stock, respectively, according to its prospectus.

WATCH: IPO window likely to open in first half of 2026: PitchBook

IPO window likely to open in first half of 2026: PitchBook

Continue Reading

Trending