https://www.scmp.com/rss/522943/feed The latest news and top stories on Liang Wenfeng. An alumnus of Zhejiang University with a background in electronic engineering and computer vision, he co-founded High-Flyer, a prominent quantitative hedge fund utilising artificial intelligence in finance. Pivoting to AI, he established DeepSeek in 2023. en https://assets.i-scmp.com/static/img/icons/scmp-meta-1200x630.png https://www.scmp.com Ke Meng Ke Meng In January, a top Chinese AI researcher told an industry summit in Beijing there was less than a 20 per cent chance of any Chinese company surpassing a leading US artificial intelligence firm in the next three to five years. The remark by Lin Junyang, until recently a technical leader working on Qwen, one of China’s most capable open-source AI models under Alibaba (which owns the South China Morning Post), made headlines. But much of the commentary missed a more important question Lin posed:... https://www.scmp.com/opinion/china-opinion/article/3347763/how-assess-chinas-real-chance-winning-ai-race-against-us?utm_source=rss_feed https://www.scmp.com/opinion/china-opinion/article/3347763/how-assess-chinas-real-chance-winning-ai-race-against-us?utm_source=rss_feed Sun, 29 Mar 2026 08:30:09 +0000 Mandy Zuo Mandy Zuo Ahead of China’s annual legislative meetings – typically a window into Beijing’s top-level policy agenda – this is the third entry in a series examining the complex economic recalibration driving China’s growth philosophy and its wide-ranging implications for local governments, financial investors and private enterprises. In China’s eastern province of Zhejiang, a sprawling laboratory for Beijing’s “common prosperity” campaign to reduce income inequality, statistics suggest success: between 2021... https://www.scmp.com/economy/china-economy/article/3344302/how-china-could-narrow-tech-hubs-income-gap-and-avoid-silicon-valleys-fate?utm_source=rss_feed https://www.scmp.com/economy/china-economy/article/3344302/how-china-could-narrow-tech-hubs-income-gap-and-avoid-silicon-valleys-fate?utm_source=rss_feed Sun, 01 Mar 2026 02:00:12 +0000 Zoey Zhang Zoey Zhang Not long ago, a humble fruit stall in Shanghai saw its sales skyrocket when Jensen Huang, the CEO of Nvidia, dropped by for some food. Meanwhile, a once-obscure village in southern China has become a tourist hotspot, drawing over 10,000 visitors daily, simply for being the hometown of Liang Wenfeng, the founder of the Chinese artificial intelligence (AI) company DeepSeek. During Tesla CEO Elon Musk’s 44-hour visit to China in 2023, the buzz surrounding his meals and itinerary on mainland social... https://www.scmp.com/news/people-culture/china-personalities/article/3341558/why-tech-billionaires-enjoy-rock-star-treatment-china-compared-us?utm_source=rss_feed https://www.scmp.com/news/people-culture/china-personalities/article/3341558/why-tech-billionaires-enjoy-rock-star-treatment-china-compared-us?utm_source=rss_feed Thu, 29 Jan 2026 06:00:08 +0000 Ni Tao Ni Tao Hangzhou’s space industry got off to a flying start in 2026. On January 7, China’s leading private rocket firm Space Epoch broke ground on a medium-to-large liquid rocket assembly, testing and reuse facility in Hangzhou’s Qiantang district. Basing this in the port city allows the rockets to be transported by sea to launch areas in the East China Sea and recovered the same way. Logistics alone do not explain the decision. Space Epoch is plugging into an industrial ecosystem Hangzhou has been... https://www.scmp.com/opinion/china-opinion/article/3341303/how-hangzhous-tech-ecosystem-nurtures-dragons-and-rockets?utm_source=rss_feed https://www.scmp.com/opinion/china-opinion/article/3341303/how-hangzhous-tech-ecosystem-nurtures-dragons-and-rockets?utm_source=rss_feed Tue, 27 Jan 2026 08:30:09 +0000 Wency Chen Wency Chen Baidu on Thursday unveiled Ernie 5.0, a multimodal artificial intelligence model with 2.4 trillion parameters, as use of the Chinese tech giant’s AI-powered namesake assistant climbed to 200 million monthly active users. The omni-modal foundation model, capable of processing text, images, audio and video, is the Beijing-based company’s most advanced to date. The model was first previewed in November and has since climbed the rankings. A leaderboard published last week by LMArena showed... https://www.scmp.com/tech/tech-trends/article/3340866/baidu-launches-ernie-50-firms-ai-assistant-users-reach-200-million-month?utm_source=rss_feed https://www.scmp.com/tech/tech-trends/article/3340866/baidu-launches-ernie-50-firms-ai-assistant-users-reach-200-million-month?utm_source=rss_feed Thu, 22 Jan 2026 12:30:15 +0000 Vincent Chow,Xinyi Wu Vincent Chow,Xinyi Wu The CEO of Chinese artificial intelligence company MiniMax met Premier Li Qiang on Monday, becoming the second representative from a foundational AI model developer to meet with the national leader after Liang Wenfeng, founder of DeepSeek, a year ago. The appearance of 36-year-old Yan Junjie along with eight other leading figures from China’s business, culture and education spheres underscored the rapid ascent of the newly minted billionaire, who steered his company founded just four years ago... https://www.scmp.com/tech/policy/article/3340515/china-signals-global-ai-push-minimax-ceo-joins-meeting-premier-li-qiang?utm_source=rss_feed https://www.scmp.com/tech/policy/article/3340515/china-signals-global-ai-push-minimax-ceo-joins-meeting-premier-li-qiang?utm_source=rss_feed Tue, 20 Jan 2026 06:17:18 +0000 Ben Jiang Ben Jiang Chinese artificial intelligence firm DeepSeek continues to keep the world guessing on when its next major release – the much-anticipated updates to its V3 and R1 models – will be launched, according to analysts, amid its recent publication of technical papers. The papers underscored DeepSeek’s efforts to improve the underlying infrastructure of AI systems in China at a time when geopolitical tensions and domestic production hurdles restricted the country’s access to advanced semiconductors to... https://www.scmp.com/tech/tech-trends/article/3339769/deepseek-stays-mum-next-ai-model-release-technical-papers-show-frontier-innovation?utm_source=rss_feed https://www.scmp.com/tech/tech-trends/article/3339769/deepseek-stays-mum-next-ai-model-release-technical-papers-show-frontier-innovation?utm_source=rss_feed Wed, 14 Jan 2026 02:00:14 +0000 Vincent Chow Vincent Chow A technical paper co-authored by Liang Wenfeng, the founder of Chinese artificial intelligence start-up DeepSeek, and a group of Peking University researchers has proposed a new model training technique, which they say can facilitate “aggressive parameter expansion” by bypassing graphics processing unit (GPU) memory constraints. The development underscores the Hangzhou start-up’s continued focus on maximising cost efficiency amid a deficit in computational power relative to US industry leaders,... https://www.scmp.com/tech/tech-trends/article/3339740/deepseek-founders-latest-paper-proposes-new-ai-model-training-bypass-gpu-limits?utm_source=rss_feed https://www.scmp.com/tech/tech-trends/article/3339740/deepseek-founders-latest-paper-proposes-new-ai-model-training-bypass-gpu-limits?utm_source=rss_feed Tue, 13 Jan 2026 10:00:16 +0000 Coco Feng Coco Feng High-Flyer Quant, co-owned by DeepSeek founder and CEO Liang Wenfeng, recorded a return of 56.6 per cent last year to rank second among China’s 10 top-performing large hedge-fund firms. According to the ranking published on Monday by data provider Shenzhen PaiPaiWang Investment & Management, High-Flyer only trailed behind Lingjun Investment, which posted a return of 73.5 per cent in 2025. All the top performers were quantitative hedge funds, underscoring the popularity in China of using complex... https://www.scmp.com/tech/tech-trends/article/3339633/deepseek-founders-high-flyer-ranks-among-chinas-top-hedge-fund-firms-2025?utm_source=rss_feed https://www.scmp.com/tech/tech-trends/article/3339633/deepseek-founders-high-flyer-ranks-among-chinas-top-hedge-fund-firms-2025?utm_source=rss_feed Mon, 12 Jan 2026 13:02:30 +0000 Vincent Chow Vincent Chow Chinese artificial intelligence firm DeepSeek has updated the technical paper of its landmark R1 model, showcasing the contributions of the 18 core scientists who powered its AI ambitions and capabilities, as anticipation heightened about a potential new major model release. The paper suggested that DeepSeek retained all 18 scientists behind its AI model development efforts, as well as many of the R1 project’s 176 contributors, despite fierce competition for talent in China’s AI industry. The... https://www.scmp.com/tech/tech-trends/article/3339215/core-chinese-research-team-behind-cutting-edge-ai-model-r1-remains-intact-deepseek?utm_source=rss_feed https://www.scmp.com/tech/tech-trends/article/3339215/core-chinese-research-team-behind-cutting-edge-ai-model-r1-remains-intact-deepseek?utm_source=rss_feed Thu, 08 Jan 2026 13:00:12 +0000 Eunice Xu Eunice Xu DeepSeek’s proposed “mHC” architecture could transform the training of large language models (LLMs) – the technology behind artificial intelligence chatbots – as developers look for ways to scale models without simply adding more computing power. However, experts cautioned that while the approach could prove far-reaching, it might still prove difficult to put into practice. In a technical paper released last week, co-authored by DeepSeek founder and CEO Liang Wenfeng, the company proposed... https://www.scmp.com/tech/big-tech/article/3338763/deepseek-pitches-new-route-scale-ai-researchers-call-more-testing?utm_source=rss_feed https://www.scmp.com/tech/big-tech/article/3338763/deepseek-pitches-new-route-scale-ai-researchers-call-more-testing?utm_source=rss_feed Mon, 05 Jan 2026 11:00:10 +0000 Ben Jiang Ben Jiang On the last day of 2025, DeepSeek published a new technical paper, with founder and CEO Liang Wenfeng among the 19 co-authors, about “manifold-constrained hyper-connections” – a general framework for training artificial intelligence systems at scale, which suggested “promising directions for the evolution of foundational models”. That release was a fitting reminder to the world, especially during the peak of the Christmas holiday season, about Chinese AI companies’ sharpened focus on innovation... https://www.scmp.com/tech/tech-war/article/3338528/tech-war-china-takes-confident-strides-develop-more-ai-innovation-2026?utm_source=rss_feed https://www.scmp.com/tech/tech-war/article/3338528/tech-war-china-takes-confident-strides-develop-more-ai-innovation-2026?utm_source=rss_feed Sat, 03 Jan 2026 03:00:22 +0000 Eunice Xu Eunice Xu DeepSeek’s latest technical paper, co-authored by the firm’s founder and CEO Liang Wenfeng, has been cited as a potential game changer in developing artificial intelligence models, as it could translate into improvements in the fundamental architecture of machine learning. The paper’s theme of Manifold-Constrained Hyper-Connections (mHC) marks an improvement to conventional hyper-connections and residual networks (ResNet), a fundamental mechanism underlying large language models (LLMs),... https://www.scmp.com/tech/tech-trends/article/3338535/deepseek-proposes-shift-ai-model-development-mhc-architecture-upgrade-resnet?utm_source=rss_feed https://www.scmp.com/tech/tech-trends/article/3338535/deepseek-proposes-shift-ai-model-development-mhc-architecture-upgrade-resnet?utm_source=rss_feed Fri, 02 Jan 2026 10:30:15 +0000 Vincent Chow Vincent Chow Chinese artificial intelligence start-up DeepSeek has ushered in 2026 with a new technical paper, co-authored by founder Liang Wenfeng, that proposes a rethink of the fundamental architecture used to train foundational AI models. The method – dubbed Manifold-Constrained Hyper-Connections (mHC) – forms part of the Hangzhou firm’s push to make its models more cost-effective as it strives to keep pace with better-funded US rivals with deeper access to computing power. It also reflected the... https://www.scmp.com/tech/big-tech/article/3338427/deepseek-kicks-2026-paper-signalling-push-train-bigger-models-less?utm_source=rss_feed https://www.scmp.com/tech/big-tech/article/3338427/deepseek-kicks-2026-paper-signalling-push-train-bigger-models-less?utm_source=rss_feed Thu, 01 Jan 2026 13:51:48 +0000 Danielle Popov Danielle Popov DeepSeek’s founder and CEO, Liang Wenfeng, has been listed among the top 10 “people who shaped science in 2025” by the British journal Nature, which hailed the 40-year-old entrepreneur as a “Chinese finance whizz” whose breakthrough artificial intelligence models had stunned the world. The Nature’s 10 profile of Liang recognised the disruption caused in January by the release of DeepSeek-R1, a reasoning model that showed how “the United States was not as far ahead in AI as many experts had... https://www.scmp.com/tech/tech-trends/article/3335748/deepseeks-liang-wenfeng-makes-cut-top-10-people-who-shaped-science-2025?utm_source=rss_feed https://www.scmp.com/tech/tech-trends/article/3335748/deepseeks-liang-wenfeng-makes-cut-top-10-people-who-shaped-science-2025?utm_source=rss_feed Tue, 09 Dec 2025 10:00:16 +0000 Wency Chen Wency Chen Chinese entrepreneur Liang Wenfeng, the founder and CEO of DeepSeek, appears to have made a big score from the initial public offering of chip designer Moore Threads Technology, whose shares surged more than fivefold in its trading debut in Shanghai. High-Flyer Quantitative Fund, a hedge fund co-founded by Liang, had subscribed to a total of 82,244 shares of Moore Threads, at 114.28 yuan per share, via two entities: Zhejiang High-Flyer Asset Management and Ningbo High-Flyer Quant Investment... https://www.scmp.com/tech/tech-trends/article/3335645/deepseek-founder-liang-wenfeng-cleans-chip-firm-moore-threads-shanghai-ipo?utm_source=rss_feed https://www.scmp.com/tech/tech-trends/article/3335645/deepseek-founder-liang-wenfeng-cleans-chip-firm-moore-threads-shanghai-ipo?utm_source=rss_feed Mon, 08 Dec 2025 13:00:21 +0000 Ben Jiang Ben Jiang Former DeepSeek researcher Luo Fuli says she has joined Xiaomi, months after the smartphone and electric vehicle giant reportedly approached the renowned Chinese artificial intelligence prodigy. In a post on her WeChat account on Wednesday, Luo said she had joined Beijing-based Xiaomi, a move that is expected to burnish the company’s AI credentials. “Intelligence will step beyond language into the physical world,” Luo wrote. “I’m at Xiaomi MiMo, with a group of creative, talented and sincere... https://www.scmp.com/tech/big-tech/article/3332502/chinese-ai-prodigy-luo-fuli-joins-xiaomi-industry-competition-talent-heats?utm_source=rss_feed https://www.scmp.com/tech/big-tech/article/3332502/chinese-ai-prodigy-luo-fuli-joins-xiaomi-industry-competition-talent-heats?utm_source=rss_feed Wed, 12 Nov 2025 10:30:13 +0000 Vincent Chow Vincent Chow Chinese artificial intelligence start-up DeepSeek made a rare appearance at a state-backed industry event on Friday as a senior researcher reaffirmed the AI lab’s commitment to developing artificial general intelligence (AGI) despite its potentially “dangerous” impacts on society. Chen Deli spoke during a panel discussion alongside the heads of five other companies collectively known as China’s “six little dragons” of AI at the World Internet Conference in Wuzhen, eastern Zhejiang province. It... https://www.scmp.com/tech/big-tech/article/3332086/chinas-deepseek-makes-rare-public-comment-calls-ai-whistle-blower-job-losses?utm_source=rss_feed https://www.scmp.com/tech/big-tech/article/3332086/chinas-deepseek-makes-rare-public-comment-calls-ai-whistle-blower-job-losses?utm_source=rss_feed Sun, 09 Nov 2025 04:37:07 +0000 Vincent Chow Vincent Chow DeepSeek’s landmark peer-reviewed article featured in the British journal Nature could encourage other Chinese artificial intelligence companies to submit their work to major scientific publications, according to AI experts. Hangzhou-based DeepSeek set a “very good example” for future AI model releases from both Chinese and US firms, according to Huan Sun, an associate professor of computer science and engineering at Ohio State University. “I hope frontier model developers could follow suit and... https://www.scmp.com/tech/tech-trends/article/3326415/deepseek-sets-example-ai-firms-have-work-peer-reviewed-major-journals-experts?utm_source=rss_feed https://www.scmp.com/tech/tech-trends/article/3326415/deepseek-sets-example-ai-firms-have-work-peer-reviewed-major-journals-experts?utm_source=rss_feed Mon, 22 Sep 2025 23:00:10 +0000 Vincent Chow Vincent Chow DeepSeek has revealed details about the risks posed by its artificial intelligence models for the first time, noting that open-sourced models are particularly susceptible to being “jailbroken” by malicious actors. The Hangzhou-based start-up said it evaluated its models using industry benchmarks as well as its own tests in a peer-reviewed article published in the academic journal Nature. American AI companies often publicise research about the risks of their rapidly improving models and have... https://www.scmp.com/tech/big-tech/article/3326214/deepseek-warns-jailbreak-risks-its-open-source-models?utm_source=rss_feed https://www.scmp.com/tech/big-tech/article/3326214/deepseek-warns-jailbreak-risks-its-open-source-models?utm_source=rss_feed Sat, 20 Sep 2025 23:00:17 +0000 Vincent Chow Vincent Chow Chinese artificial intelligence start-up DeepSeek has lifted the veil on how it filters data to train its models, raising red flags about “hallucination” and “abuse” risks. In a document published on Monday, the Hangzhou-based start-up said it “has always prioritised AI security” and decided to make its disclosure to help people use its models, at a time when Beijing is ramping up oversight over the industry. The company said data in the pre-training stage was “mainly” collected from publicly... https://www.scmp.com/tech/tech-trends/article/3324149/deepseek-sheds-light-data-collection-ai-training-and-warns-hallucination-risks?utm_source=rss_feed https://www.scmp.com/tech/tech-trends/article/3324149/deepseek-sheds-light-data-collection-ai-training-and-warns-hallucination-risks?utm_source=rss_feed Wed, 03 Sep 2025 10:00:13 +0000 Zhou Xin Zhou Xin DeepSeek, the Hangzhou-based artificial intelligence start-up, may have made another breakthrough that could disrupt the way AI models are trained in China. Its innovative data format, UE8M0 FP8, could pave the way for home-grown graphics processing units (GPUs) to be deployed in training powerful models even though local GPUs are not as powerful as Nvidia’s. China’s stock market investors were excited about the breakthrough, pouring money into local GPU developers such as Cambricon... https://www.scmp.com/opinion/china-opinion/article/3323051/deepseeks-data-format-innovation-local-ai-chips-fans-fresh-hopes-more-disruption?utm_source=rss_feed https://www.scmp.com/opinion/china-opinion/article/3323051/deepseeks-data-format-innovation-local-ai-chips-fans-fresh-hopes-more-disruption?utm_source=rss_feed Mon, 25 Aug 2025 23:00:11 +0000 Ben Jiang Ben Jiang Chinese artificial intelligence start-up DeepSeek said on Thursday that its newly released V3.1 model supported both “think” and “non-think” modes, marking the firm’s “first step towards the agent era” – a shift that suggests a change in its research focus and and the possibility it would forgo the highly anticipated R2 reasoning model. The “think” mode on DeepSeek’s namesake chatbot was previously powered by its R1 reasoning model that garnered global attention after its release in January,... https://www.scmp.com/tech/tech-trends/article/3322669/deepseeks-upgraded-ai-model-absorbs-reasoning-feature-move-towards-agent-era?utm_source=rss_feed https://www.scmp.com/tech/tech-trends/article/3322669/deepseeks-upgraded-ai-model-absorbs-reasoning-feature-move-towards-agent-era?utm_source=rss_feed Thu, 21 Aug 2025 11:00:08 +0000 Ben Jiang Ben Jiang Chinese artificial intelligence start-up DeepSeek has updated its foundational V3 model and removed references to its reasoning model R1 from its chatbot, prompting speculation about a shift in the company’s research focus. DeepSeek announced on Tuesday the release of the V3.1 model in a brief message to one of its WeChat user groups. The update expands the context window to 128k, allowing the model to hold more information – equivalent to a roughly 300-page book – during user interactions. The... https://www.scmp.com/tech/big-tech/article/3322481/deepseeks-v31-update-and-missing-r1-label-spark-speculation-over-fate-r2-ai-model?utm_source=rss_feed https://www.scmp.com/tech/big-tech/article/3322481/deepseeks-v31-update-and-missing-r1-label-spark-speculation-over-fate-r2-ai-model?utm_source=rss_feed Wed, 20 Aug 2025 09:00:09 +0000 Brian Rhoads,Raymond Ma Brian Rhoads,Raymond Ma Mainland Chinese shares hit the highest in a decade amid appealing valuations and meagre interest rates driving savers into the stock market. The Shanghai Composite Index has climbed 11 per cent this year and 36 per cent from a low in September. The gauge is on pace for four straight monthly gains for the first time since the end of 2021, even with a slight retreat on Tuesday. Traditionally cautious Chinese savers have turned to stocks after the interest rate on one-year bank deposits fell below... https://www.scmp.com/plus/article/3322350/china-stocks-reach-10-year-high-savers-seek-escape-low-rates?utm_source=rss_feed https://www.scmp.com/plus/article/3322350/china-stocks-reach-10-year-high-savers-seek-escape-low-rates?utm_source=rss_feed Tue, 19 Aug 2025 09:03:33 +0000 Ben Jiang Ben Jiang China’s leading artificial intelligence start-up, DeepSeek, finds itself at the centre of speculation about the release of its next-generation R2 reasoning model, following OpenAI’s launch of its GPT-5 model last week. DeepSeek – which made waves in the industry with its V3 model in December and the R1 model in January – has not introduced major updates to its products in the past months, aside from two minor revisions. While the market had expected DeepSeek to introduce a new foundation model... https://www.scmp.com/tech/tech-trends/article/3321855/where-deepseeks-next-ai-model-speculation-rises-after-openai-unveils-gpt-5?utm_source=rss_feed https://www.scmp.com/tech/tech-trends/article/3321855/where-deepseeks-next-ai-model-speculation-rises-after-openai-unveils-gpt-5?utm_source=rss_feed Thu, 14 Aug 2025 10:00:23 +0000 Hannah Wang Hannah Wang A research paper co-authored by Liang Wenfeng, founder of Chinese artificial intelligence start-up DeepSeek, was honoured with the best paper award at the Association for Computational Linguistics (ACL) conference in Vienna, Austria, widely recognised as the premier global conference for AI researchers. The paper, titled “Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention,” was published on February 27, with Liang listed as one of 15 authors. The “native sparse... https://www.scmp.com/tech/big-tech/article/3320255/deepseek-founder-shares-best-paper-award-top-global-ai-research-conference?utm_source=rss_feed https://www.scmp.com/tech/big-tech/article/3320255/deepseek-founder-shares-best-paper-award-top-global-ai-research-conference?utm_source=rss_feed Thu, 31 Jul 2025 10:00:17 +0000 Nora Mankel Nora Mankel This year, Beijing’s Tsinghua University is up two spots to be just outside the top 10. Peking University and Zhejiang University have also moved up the list, climbing to 25th and 45th respectively. The rankings are the latest list of the Best Global Universities compiled by American media company US News and World Report, which looks at 2,250 top institutions from 105 countries. The assessment focuses exclusively on the overall academic research and reputations of the universities, weighing up... https://www.scmp.com/news/china/science/article/3318178/talent-gambit-how-us-brain-drain-chinas-brain-gain?utm_source=rss_feed https://www.scmp.com/news/china/science/article/3318178/talent-gambit-how-us-brain-drain-chinas-brain-gain?utm_source=rss_feed Mon, 14 Jul 2025 22:00:07 +0000