# LLM
Latest news and articles about LLM
Total: 26 articles found
The 14-Million-Dollar Brain: China’s Tech Giants Ignite a Cutthroat War for AI Talent
ByteDance’s high-profile hiring of a DeepSeek researcher highlights a frantic and expensive talent war in China's AI sector. As tech giants compete for 'Post-95' scientists, astronomical compensation packages and frequent cross-firm poaching are becoming the new industry standard.
DeepSeek V4: A Trillion-Parameter Challenge to the Global AI Status Quo
DeepSeek is set to launch its V4 flagship model in late April, featuring a trillion-parameter architecture and a million-token context window. Most significantly, the model is deeply optimized for Huawei's Ascend chips, marking a major step in China's efforts to achieve AI independence from Nvidia's CUDA ecosystem.
DeepSeek’s ‘Expert’ Gambit: China’s AI Underdog Signals Strategic Bifurcation
DeepSeek has introduced a dual-mode interface featuring 'Fast' and 'Expert' tracks, both powered by a 671-billion-parameter architecture. The update signals a strategic move toward specialized reasoning models and has already driven significant gains in Chinese technology-focused ETFs.
Alibaba’s FIPO Breakthrough: A New Frontier in Reinforcement Learning and Model Reasoning
Alibaba's Tongyi Lab has launched FIPO, an algorithm that overcomes 'reasoning length stagnation' in AI models. By outperforming OpenAI’s o1-mini and DeepSeek in logic-heavy benchmarks, Alibaba is positioning itself as a leader in the next generation of reasoning-centric artificial intelligence.
Beijing’s Spy Agency Sounds the Alarm on China’s AI ‘Token’ Gold Rush
China’s Ministry of State Security has issued a warning against speculative trading and 'hoarding' of AI tokens, citing risks ranging from financial fraud to foreign espionage. As AI token usage in China surges to 140 trillion units daily, the state is moving to define these computational units as critical infrastructure rather than investment vehicles.
Poisoning the Well: Why China’s AI Marketing Industry is Thriving After a Major Scandal
China's '3.15' gala exposed a gray market for manipulating AI recommendations, but the scandal has unexpectedly fueled a surge in demand for Generative Engine Optimization (GEO) services. As search habits shift from browsers to AI bots, businesses are racing to influence LLM outputs, sparking a new 'arms race' between marketers and AI developers.
Chasing Anthropic: Zhipu AI’s Financial Debut Reveals the High Cost of China’s AGI Ambitions
Zhipu AI's 2025 financial report shows 131% revenue growth alongside a massive 4.7 billion RMB loss, highlighting the extreme capital intensity of the Chinese LLM market. While the company is successfully pivoting toward an API-first 'MaaS' model to emulate Anthropic, it remains heavily dependent on low-margin localized deployments for the bulk of its income.
DeepSeek’s Strategic Silence: Outages Hint at China’s Next AI Shockwave
DeepSeek experienced multiple service outages in late March 2026, including a massive 12-hour disruption, sparking widespread speculation that the company is preparing for the imminent launch of its highly anticipated V4 model.
DeepSeek’s Growing Pains: Outage Highlights the Strain on China’s AI Champion
DeepSeek has restored its web and app services after a major overnight outage that sparked concerns over its infrastructure capacity. The disruption occurred amid high anticipation for the upcoming V4 model, highlighting the operational challenges facing China's leading AI contender.
The Price of Popularity: DeepSeek’s Prolonged Outage Signals Infrastructure Growing Pains in China’s AI Sector
China's DeepSeek AI platform experienced a major, multi-hour outage starting March 29, 2026, leaving both its app and website non-functional. The failure highlights the significant infrastructure and scaling challenges facing independent AI labs as they struggle to keep pace with explosive user demand.
Scaling Shadows: DeepSeek’s Service Outage Highlights the Infrastructure Strain on China’s AI Giants
DeepSeek experienced a widespread service disruption on March 29, 2026, leaving users unable to access new chat sessions. The outage underscores the significant infrastructure and compute challenges facing China's leading AI developers amidst rapid user adoption.
Gaming the Oracle: How ‘Marketing Pollution’ is Distorting China’s AI-Driven Fund Advice
Chinese marketing firms are using 'content feeding' tactics to manipulate AI fund recommendations, creating a trend of 'marketing pollution.' This shift from search visibility to algorithmic influence masks promotional content as neutral advice, posing new risks for retail investors and regulatory compliance.