عربي English Türkçe اردو فارسى
Logo
Latest News
Technology

AMD Shows off New Higher Performing AI Chip at CES Event

Lisa Su, CEO of AMD, holds an AMD Ryzen AI Halo Developer Platform during an AMD keynote address at CES 2026, an annual consumer electronics trade show, in Las Vegas, Nevada, US, January 5, 2026. (Reuters)
Lisa Su, CEO of AMD, holds an AMD Ryzen AI Halo Developer Platform during an AMD keynote address at CES 2026, an annual consumer electronics trade show, in Las Vegas, Nevada, US, January 5, 2026. (Reuters)
TT
TT

AMD Shows off New Higher Performing AI Chip at CES Event

Lisa Su, CEO of AMD, holds an AMD Ryzen AI Halo Developer Platform during an AMD keynote address at CES 2026, an annual consumer electronics trade show, in Las Vegas, Nevada, US, January 5, 2026. (Reuters)
Lisa Su, CEO of AMD, holds an AMD Ryzen AI Halo Developer Platform during an AMD keynote address at CES 2026, an annual consumer electronics trade show, in Las Vegas, Nevada, US, January 5, 2026. (Reuters)

Advanced Micro Devices CEO Lisa Su showed off a number of the company's AI chips on Monday at the CES trade show in Las Vegas, including its advanced MI455 AI processors, which are components in the data center server racks that the company is selling to firms like ChatGPT maker OpenAI.

Su also unveiled the MI440X, a version of the MI400 series chip designed for on-premise use at businesses. The so-called enterprise version is designed to fit into infrastructure that is not specifically designed for AI clusters. The ‌MI440X is a version ‌of an earlier chip that the US plans ‌to ⁠use in a supercomputer.

AMD ‌is one of Nvidia's strongest rivals but has struggled to have as much success. In October, AMD signed a deal with OpenAI that, in addition to the financial upside, was a major vote of confidence in AMD's AI chips and software. But it is unlikely to dent Nvidia's dominance, as the market leader continues to sell every AI chip it can make, analysts said.

At the Monday event, OpenAI President Greg Brockman joined Su on ⁠stage and said chip advancements were critical to OpenAI's vast computing needs.

Looking to the future needs of ‌companies like OpenAI, Su previewed the MI500 and said it ‍offered 1,000 times the performance of an ‍older version of the processor. The company said the chips would launch in ‍2027.

At the event, Su hosted Daniele Pucci, CEO of Generative Bionics, an Italian AI developer, who unveiled GENE.01, a humanoid robot.

"Our first commercial humanoid robot will be manufactured in the second half of 2026," Pucci said at the event.

Earlier on Monday, Nvidia showed off its next-generation Vera Rubin platform, which is made up of six separate chips. CEO Jensen Huang said it was in full production. It is expected to debut ⁠later this year.

In October, AMD signed the deal with ChatGPT maker OpenAI that will add billions of dollars to the company’s annual revenue. The first deployment of AI chips that incorporate AMD’s MI400 series will roll out this year. Nvidia has generated tens of billions of dollars in quarterly revenue from its AI chip sales, a feat that AMD has struggled to achieve thus far.

OpenAI is a key customer of AMD and executives at the Santa Clara, California-based company expect the deal to lead to significant additional new sales.

Also on Monday, AMD launched its Ryzen AI 400 Series processors for AI PCs, alongside Ryzen AI Max+ chips for advanced local inference and gaming. Intel held a launch event ‌earlier for its Panther Lake chips that it said would be available for order on Tuesday.

Most Viewed

 
 
Technology

India Plans AI 'Data City' on Staggering Scale

As India races to narrow the AI gap with the US and China, it is planning a vast new "data city" to power digital growth on a staggering scale. Arun SANKAR / AFP
As India races to narrow the AI gap with the US and China, it is planning a vast new "data city" to power digital growth on a staggering scale. Arun SANKAR / AFP
TT
TT

India Plans AI 'Data City' on Staggering Scale

As India races to narrow the AI gap with the US and China, it is planning a vast new "data city" to power digital growth on a staggering scale. Arun SANKAR / AFP
As India races to narrow the AI gap with the US and China, it is planning a vast new "data city" to power digital growth on a staggering scale. Arun SANKAR / AFP

As India races to narrow the artificial intelligence gap with the United States and China, it is planning a vast new "data city" to power digital growth on a staggering scale, the man spearheading the project says.
"The AI revolution is here, no second thoughts about it," said Nara Lokesh, information technology minister for Andhra Pradesh state, which is positioning the city of Visakhapatnam as a cornerstone of India's AI push.
"And as a nation... we have taken a stand that we've got to embrace it," he told AFP ahead of an international AI summit next week in New Delhi.
Lokesh boasts the state has secured investment agreements of $175 billion involving 760 projects, including a $15 billion investment by Google for its largest AI infrastructure hub outside the United States.
And a joint venture between India's Reliance Industries, Canada's Brookfield and US firm Digital Realty is investing $11 billion to develop an AI data center in the same city.
Visakhapatnam -- home to around two million people and popularly known as "Vizag" -- is better known for its cricket ground that hosts international matches than cutting-edge technology.
But the southeastern port city is now being pitched as a landing point for submarine internet cables linking India to Singapore.
"The data city is going to come in one ecosystem... with a 100 kilometer (60 mile) radius," Lokesh said. For comparison, Taiwan is roughly 100 kilometers wide.
- 'Whole nine yards' -
Lokesh said the plan goes far beyond data connectivity, adding that his state had "received close to 25 percent of all foreign direct investments" to India in 2025.
"It's not just about the data centers," he explained while outlining a sweeping vision of change, with Andhra Pradesh offering land at one US cent per acre (three per hectare) for major investors.
"I'm chasing the companies that make those servers that go sit in those data centers, the companies that make the entire air conditioning, the water-cooling system -- the whole nine yards."
The 43-year-old, Stanford-educated minister is the son of Andhra Pradesh Chief Minister N. Chandrababu Naidu, who helped turn Hyderabad into a major technology hub that is dubbed "Cyberabad".
They are allies of Prime Minister Narendra Modi, who will host the AI Impact Summit from Monday.
India is now third in a global AI power ranking -- sitting above South Korea and Japan -- based on more than 40 indicators from patents to private funding calculated by Stanford University's Institute for Human-Centered AI.
With more than a billion internet users, India has seen a surge of investment as generative AI players seek inroads to the world's most populous country.
Microsoft said in December it will invest $17.5 billion to help build the country's artificial intelligence infrastructure, with CEO Satya Nadella calling it the firm's "largest investment ever in Asia".
But critics say India lags in access to high-end computing power or commercial AI deployment, and remains more a consumer than creator of the cutting-edge technology.
Some question whether data centers will create meaningful employment when up and running, but Lokesh rejects that.
"Every industrial revolution has always created more jobs than it has displaced," he said.
"But it has created those jobs in countries that have embraced the industrial revolution."
- 'Learned from China' -
Lokesh argues that the jobs and economic benefits would more than compensate for the giveaway cost of land.
He said the state government had accounted for the vast electricity and water demands for the energy-hungry industry, and would tap "surplus water" that drains into the Bay of Bengal to cool the massive data centers.
"It's a crime that so much water during monsoons goes into our oceans," he said.
He cited China as an inspiration -- admiring how India's rival had "been able to systematically bring people out of poverty" at speed.
The state's plan to create industrial clusters was something he had "learned from China".
With a target of six gigawatts of data center capacity -- three already signed and another three in the pipeline -- Andhra Pradesh is betting that speed and scale will give it an edge.
New Delhi last year agreed to "in-principle approval" for six 1.2 GW nuclear power plants at Kovvada in Andhra Pradesh.
"We are on a journey," Lokesh said. "We will execute these projects at a pace that the country has never seen".

Technology

SDAIA President Leads Saudi Delegation to India AI Impact Summit 2026

File photo of the Saudi flag/AAWSAT
File photo of the Saudi flag/AAWSAT
TT
TT

SDAIA President Leads Saudi Delegation to India AI Impact Summit 2026

File photo of the Saudi flag/AAWSAT
File photo of the Saudi flag/AAWSAT

The Kingdom of Saudi Arabia, represented by the Saudi Data and Artificial Intelligence Authority (SDAIA), is participating in the India AI Impact Summit 2026 in New Delhi on February 19 and 20.

Held under the theme "People, Planet, and Progress," the summit brings together global leaders to discuss the developmental impact of artificial intelligence, focusing on social empowerment, AI safety, and inclusive access to resources, SPA reported.

The Saudi delegation, led by SDAIA President Dr. Abdullah Alghamdi, aims to highlight Vision 2030 initiatives that leverage advanced technologies for both humanitarian and economic benefits.

By engaging in this international dialogue, Saudi Arabia reaffirms its commitment to strengthening strategic ties with India and supporting the United Nations Sustainable Development Goals through responsible AI innovation and global cooperation.

Technology

China’s ByteDance Releases Doubao 2.0 AI Model for 'Agent Era’

This picture taken on February 5, 2026 shows advertising promoting ByteDance's cloud and AI service platform 'Volcano Engine' and chatbot 'Doubao' at the Beijing Capital International airport in Beijing. (Photo by Adek BERRY / AFP)
This picture taken on February 5, 2026 shows advertising promoting ByteDance's cloud and AI service platform 'Volcano Engine' and chatbot 'Doubao' at the Beijing Capital International airport in Beijing. (Photo by Adek BERRY / AFP)
TT
TT

China’s ByteDance Releases Doubao 2.0 AI Model for 'Agent Era’

This picture taken on February 5, 2026 shows advertising promoting ByteDance's cloud and AI service platform 'Volcano Engine' and chatbot 'Doubao' at the Beijing Capital International airport in Beijing. (Photo by Adek BERRY / AFP)
This picture taken on February 5, 2026 shows advertising promoting ByteDance's cloud and AI service platform 'Volcano Engine' and chatbot 'Doubao' at the Beijing Capital International airport in Beijing. (Photo by Adek BERRY / AFP)

China's ByteDance has rolled out its Doubao 2.0 model, an upgrade of the country's most widely used artificial-intelligence app, the company said on Saturday.

ByteDance is one of several Chinese firms hoping to generate overseas and domestic buzz around its new AI models during the Lunar New Year holiday, which starts on Sunday, when hundreds of millions of Chinese partake in family gatherings in their hometowns.

The company, like rival Alibaba, was caught off-guard by DeepSeek's meteoric rise to global fame during last year's Spring Festival, when Silicon Valley and investors worldwide were ⁠shocked by how ⁠a Chinese firm had come up with a model comparable to OpenAI's best but seemingly developed at a fraction of the cost.

The release of Doubao 2.0, ahead of a highly anticipated new DeepSeek model, is likely aimed at preventing such a scenario from repeating itself, Reuters reported.

A video-generation AI model that ByteDance released on Thursday, Seedance 2.0, has already drawn comparisons with DeepSeek's success last year after going ⁠viral on Chinese social media and drawing praise overseas on platforms like X, including from its owner Elon Musk.

Doubao 2.0 is positioned for the "agent era", where AI models are expected to execute complex real-world tasks rather than only answer questions, ByteDance said in a statement.

The model's pro version includes complex reasoning and multi-step task execution capabilities that match OpenAI's GPT 5.2 and Google's Gemini 3 Pro, while reducing usage costs by roughly an order of magnitude, according to the company.

"This cost advantage will become even more crucial as real-world, complex tasks involve large-scale inference and multi-step generation that will expend a huge amount of ⁠tokens," ByteDance said, ⁠referring to the unit of data processed by an AI model.

Doubao leads all AI chatbot apps in China with 155 million weekly active users, with DeepSeek second at 81.6 million, according to information provider QuestMobile's most recent data, published in late December.

But Doubao 2.0's release could help ByteDance fend off recent pressure from domestic competitors. Alibaba on February 6 announced it was spending 3 billion yuan ($400 million) on a coupon giveaway campaign to attract more users to its Qwen AI app, allowing them to use the incentives to purchase food and drink directly in the chatbot.

This led daily active users on Qwen to skyrocket from 7 million to 58 million, just 23 million shy of Doubao's figures on the same day, according to QuestMobile.