Share This

Showing posts with label Innovation. Show all posts
Showing posts with label Innovation. Show all posts

Sunday, February 16, 2025

DeepSeek sparks buying spree of Chinese equities

 DeepSeek Photo: VCG

DeepSeek Photo: VCG

While the artificial intelligent (AI) model developed by Chinese tech start-up DeepSeek has stunned the tech world, the company's breakthrough - a demonstration of China's innovation strength - has also triggered a buying spree among global financial institutions and hedge funds of Chinese equities.

To date, the benchmark Hang Seng Tech index has entered a bull market, rising more than 30 percent in the past month. Also in just the past month, China's onshore and offshore equity markets have added more than $1.3 trillion in total value, according to a Bloomberg report. 

The sweeping momentum highlights how the recent AI-driven development has been serving as a catalyst for global capitals to broadly reappraise Chinese assets, which have been largely undervalued in the past few years, analysts said, while expecting that foreign investors' confidence on Chinese equities will be further lifted up in the long-term along with tech-fueled industrial upgrade, more economic supportive measures as well as a robust economic growth streak.

According to the Bloomberg report on Sunday, hedge funds have been "piling into Chinese equities at the fastest pace in months," while another emerging market, India, in contrast is suffering a record exodus of cash, with market valuation shrinking by more than $720 billion in the past month. 

The MSCI China Index is on track to outperform its Indian counterpart for a third-straight month, the longest such streak in two years, the report said.

The bullish sentiment has led to an extended Chinese tech stock rally across the board. 

As of Friday's closing, Nasdaq-listed Chinese tech stocks have gained by approximately 24 percent since the beginning of the year, news website 21jingji.com reported. In A-share market, the price of AI-related equities in the Shanghai and Shenzhen bourses also skyrocketed, with a batch of them surging by over 50 percent since trading reopened in early February.

"DeepSeek's surging popularity has offered a window of investment opportunities for foreign capitals, which has once again placed the limelight into Chinese tech industries and the broader Chinese assets," Wang Peng, an associate research fellow at the Beijing Academy of Social Sciences, told the Global Times on Sunday. 

Integrating with real economy 

China's leading social media platform WeChat confirmed with the Global Times on Sunday that it is conducting small-scale testing to integrate DeepSeek's R1 large model. It added that multiple Tencent products, including Tencent Cloud's AI code assistant, are exploring integration with DeepSeek to offer users a richer experience and enhanced services.

In addition, multiple Chinese sectors, ranging from automakers, internet companies and industrial manufacturers to information service providers, have been integrating DeepSeek's large language models (LLMs) for enhanced AI capabilities. 

Dong Shaopeng, a senior research fellow at the Chongyang Institute for Financial Studies at Renmin University of China, told the Global Times on Sunday that this adaption signals a "critical historical phase" for China's AI models. 

There used to be some pessimist narratives surrounding China's homegrown tech innovations, but the stunning debut of DeepSeek marked a "watershed," an equity fund manager based in Shenzhen, South China's Guangdong Province, who prefers not to be identified, told the Global Times on Sunday. 

Analysts said the DeepSeek-triggered Chinese equities rally would not be a short-term speculative frenzy, but instead will be a "strategic repositioning" and a "long-term" bullish run" that not only takes into account the country's cutting-edge technologies, but also a package of pro-growth policies and sound economic fundamentals. 

Kuang Zheng, chief investment officer for HSBC Global Private Banking and Wealth in China, said that DeepSeek's success could serve as a catalyst for further technological innovation among China's private enterprises, potentially leading to more groundbreaking advancements in the tech sector and boosting investor confidence in the Chinese stock market, according to a Beijing Daily report. 

Yang Delong, chief economist at Shenzhen-based First Seafront Fund, told the Global Times over the weekend that he believes that Chinese companies' significant advantages in the AI software sector will give it a competitive edge in AI applications. And with the world's largest consumer market, China is poised to drive the global industrial shift from hardware infrastructure to software services. 

"Chinese tech sectors like AI, big data, and internet technologies are set to profoundly transform the country's asset structure. This transformation will optimize and drive growth in traditional industries, offering an extended boost to the Chinese real economy," Dong said, noting that is one of the primary reasons that Chinese assets have been captivating foreign investors in recent days.

A number of foreign financial institutions and equity funds have issued upbeat calls on Chinese equities since late January. 

"It would seem that we are less than halfway through the rally" driven by DeepSeek, UBS strategists including James Wang wrote in a note released in February, according to a Bloomberg report in February 12. 

Ample liquidity and lower interest rates should help AI-related names to further re-rate, they added.

"We think 2025 is the year the investing world realizes China is outcompeting the rest of the world," Deutsche Bank said in a February report entitled "China Eats the World" that went viral on Chinese social media platforms, according to a report by the South China Morning Post in early February.

JPMorgan strategists including Rajiv Batra noted that fund flows into Chinese internet names have been positive this year, with a surge following the DeepSeek shock, according to the Bloomberg report on February 12. "We see a window of opportunity in Asia over the coming months, with another tactical rally in China driving upside," they were quoted as saying in the report.

Leading US bank Goldman Sachs forecast potential gains of 14 percent year-on-year for the MSCI China index by the end of 2025 in its base case, while up to 28 percent in its bull case outlook, according to a research report published by Goldman Sachs in early February.


RELATED ARTICLES

Wednesday, February 12, 2025

AI Action Summit kicks off in Paris with aim of harnessing potential while improving governance

A man looks at his mobile phone as he walks past flags during Artificial Intelligence Action Summit at the Grand Palais in Paris on February 10, 2025. Photo: VCG

As the 2025 AI Action Summit begins in Paris, discussions surrounding AI are set against a backdrop of geopolitical tensions and high stakes for the future of global AI governance.


With the aim of addressing how to harness artificial intelligence's (AI) potential while enhancing AI governance, political and industry leaders from more than 100 countries gathered at the Grand Palais in Paris, France for the two-day AI Action Summit, which kicked off on Monday. 

The summit focuses on five major themes: Public Interest AI, Future of Work, Innovation and Culture, Trust in AI, and Global AI Governance. 

Chinese Vice Premier of the State Council Zhang Guoqing, US Vice President JD Vance, Canadian Prime Minister Justin Trudeau and European Commission President Ursula von der Leyen are among dignitaries attending the summit.

Indian Prime Minister Narendra Modi on Monday left for France to co-chair the AI Action Summit on Tuesday, along with French President Emmanuel Macron, India TV reported on Monday.

On the first day of the Summit, workshops and panels on a wide range of topics are scheduled, including "harnessing AI for the future of work", "privacy, cybersecurity and information integrity," "towards safe and trustworthy AI," and "reinforcing efficient, effective and inclusive global AI governance," according to the organizer's official website.

The Summit of Heads of State and Government will take place on Tuesday with a plenary session to be held at the Grand Palais with international participants to discuss the key common actions to take on AI. 

The first and second editions of the summit were held in UK in 2023 and South Korea in 2024, and were both named AI Safety Summit. Compared with previous editions of the summit, the title of the third edition has evolved beyond focusing merely on safety and on a more comprehensive action sphere, so that more attention has and should be given to the global development of AI, Zhu Rongsheng, an assistant researcher at the Center for Strategic and Security Studies, Tsinghua University, told the Global Times on Monday.

As US President Donald Trump tears up his predecessor's AI guardrails to promote US competitiveness, pressure has built on the European Union to pursue a lighter-touch approach to AI to help keep European firms in the tech race, Reuters reported on Monday. Some EU leaders including the summit's host, French President Macron, also hope flexibility will be applied to the bloc's new AI Act to help homegrown startups, said the report.


DeepSeek buzz

An article published on Monday by Associated Press titled "Trump's AI ambition and China's DeepSeek overshadow an AI summit in Paris," which claimed that geopolitics of AI would be in focus in the summit. 

Addressing concerns over China's DeepSeek, Macron said in a televised interview on public television channel France 2 that France has no plan to ban it at the moment. "I do not think that it's appropriate to ban a technology because of its country of origin," Macron said, adding that France does not share the US approach of restricting technologies due to its nationality while accepting others.

Indian media platform Policy Circle wrote on Monday that the AI Action Summit is particularly important for India, given recent developments in China. India cannot afford to lag behind and must take lessons from China's AI advancements, particularly in cost-effective model training, it said.    

China is embracing the AI transformation and is working hard to advance AI, said Chinese Foreign Ministry spokesperson Guo Jiakun on Monday, in response to extensive attention and heated discussion in the world over DeepSeek, noting that we have helped developing countries enhance capacity building, advocating that AI technologies should be open sourced and there should be greater accessibility to AI services so that the benefits of AI can be shared by all countries. 

However, Guo emphasized that we are against drawing lines along ideological difference, overstretching the concept of national security, or politicizing trade and tech issues. 


Different approach

The AP report said that "organizers are working on getting countries to sign a joint political declaration gathering commitments for more ethical, democratic and environmentally sustainable AI," adding that it is unclear whether the US would agree to such a measure. 

The US position might undermine any joint communique, said Nick Reiners, senior geotechnology analyst at the Eurasia Group. "Trump is against the very idea of global governance," Reiners said, per AP report.

French President Macron rejected an outright ban on Chinese AI, emphasizing careful evaluation based on sovereignty rather than origin. France will closely examine non-European technologies, ensuring they do not compromise security or sovereignty in critical sectors, Macron was quoted by Indian news outlet FirstPost as saying. Macron's stance reflects a desire to avoid isolationist policies like those seen in the US, promoting a more nuanced view of global technology, the report said. 

China's and Europe's starting points are to limit the disruptive impact of AI on human rules, while the US is focused on limiting challenges from China, which is a different approach, Wang Yiwei, a professor at the School of International Relations at Renmin University of China, told the Global Times on Monday. 

Zhu Rongsheng said that fierce competition between powers could undermine global cooperation, and as the US continues its zero-sum mentality in its pursuit of a bigger share of the market, cooperation would be very challenging in global efforts to jointly develop AI.

DeepSeek provides an opportunity to a broader range of countries and regions and advanced AI can be obtained with relatively low cost, and the model of open-source AI, with proper safeguards, is a practical approach on AI capacity building for good and for all, Zeng Yi, Professor of AI at Chinese Academy of Sciences and member of the United Nations AI Advisory Body, told the Global Times.

The world is big and inclusive enough to have more countries contributing fundamental and pioneering research, as well as industrial applications of AI for global public good, Zeng said.  

The integration of AI with industries can unleash tremendous productivity, providing the foundational support and effective empowerment necessary for the liberation of productive forces. In an ideal scenario, this would drive a global effort toward this direction. Otherwise, it could exacerbate the wealth gap, lead to AI exploitation, create new technological oligarchs and power elites, and undermine the general safety and development of countries, widening the gap between countries and between different groups within a country, leading to a new intelligence divide, Shen Yi, a professor at Fudan University, told the Global Times. 

Diving into DeepSeek and AI for education; OpenAI targets higher education in the U.S. with ChatGPT rollout at California State University




Sunday, June 16, 2024

The classic course on Generative AI by Martin Musiol; Can Generative AI unlock productivity and growth?

     

The classic course on Generative AI by Martin Musiol | Udemy


In 2022, it seemed as though the much-anticipated AI revolution had finally arrived


无需编程基础和相关经验,零基础AI实习就业班,$10万年薪不是梦。 零...


28 Feb 2024 — MARTIN MUSIOL is the founder of generativeAI.net and the publisher of Generative AI: Short & Sweet, a popular artificial intelligence newsletter ...




If you want the economy to change, appoint business leaders who understand how to manage institutional change that remains business-friendly. — Reuters

Andrew Sheng

Recent and archived articles by Andrew Sheng

Can Generative AI unlock productivity and growth?

IF you watched Nvidia chief executive officer Jensen Huang’s remarkable presentation at Taipei Computex last month, you would be convinced that artificial intelligence (AI) has ushered in a new Industrial Revolution, in which accelerated computing with the...

Users will have control over generative AI in Windows



Copilot should be central to Windows 11 24H2. — AFP Relaxnews

Central to the next major Windows update, generative artificial intelligence promises to make its way into most Microsoft programs, in the aim of boosting user productivity. Users should, however, be able to decide which applications can and can't make use of the technology.

Faced with concerns that generative AI could be too invasive, Microsoft is reportedly set to give users a say in how applications access these artificial intelligent tools. According to the XDA Developers website, the incoming major update to Windows 11 (24H2), expected by the end of the year, will offer the possibility of defining individual permissions for each application.

This will enable users to disable the use of generative AI for some or all applications. On a larger scale, companies will be able to disable access to this AI for all their employees if they deem it inappropriate or unnecessary.

The integration of generative artificial intelligence into Windows should simplify system management, as well as the day-to-day use of most of its accompanying programs. At the core of this update are the latest developments for Microsoft's Copilot, provided it finally complies with European legislation on digital markets (DMA). 

Indeed, until further notice, Europeans will be left without Copilot, the AI-powered intelligent assistant with which it is possible to interact or customize a computer's operating system. The assistant can be useful for working on various documents (rewriting, summarising or simply explaining them) and can answer practical questions. It can be accessed directly from the taskbar, and soon via a dedicated button on future PCs.

Meanwhile, Microsoft has sought to reassure users after its new Recall feature sparked controversy. In fact, the firm has said that Recall will now be an opt-in feature rather than activated by default. Considered to be particularly intrusive, but promising to facilitate PC searches, Recall is designed to take a series of screenshots of the computer at regular intervals and then save them locally, raising questions about privacy. Initially, however, this feature will only be available on the new Copilot+ PCs, which are due to go on sale this summer. – AFP Relaxnews

Related stories:

AI will be at the heart of the next Windows update

Martin Musiol - Generative AI: Navigating the Course to the ...
AI Course for Beginners - AI From Zero To Hero
Dive into theory and complete numerous practice exercises to master your coding skills. Open up new career options in tech! "Intro to AI" Self-Paced Online Course.
AI技术学习,零基础轻松掌握 - artificial intelligence course
无需编程基础和相关经验,零基础AI实习就业班,$10万年薪不是梦。 零...

Generative AI - Martin Musiol

28 Feb 2024 — MARTIN MUSIOL is the founder of generativeAI.net and the publisher of Generative AI: Short & Sweet, a popular artificial intelligence newsletter ...Related stories:

AI will be at the heart of the next Windows update