Chinese Tech Giants Fuel Costly AI Model Race, Rejecting Acquisitions Amid Fierce Domestic Competition

China's technology giants are investing heavily in proprietary artificial intelligence models rather than acquiring existing platforms, fueling an intensely competitive race where widespread model distillation and rapid iteration are preventing any single company from establishing market dominance. The strategy contrasts sharply with earlier patterns where major technology companies might license or acquire proven AI systems. Instead, Baidu, Alibaba, Tencent, and ByteDance have each committed substantial resources to building foundation models from scratch, with recent launches including DeepSeek's V4 series (released April 2026), Alibaba's Qwen 3.6 Plus featuring a 1-million-token default context window, and ByteDance's upgraded Doubao lineup alongside new image and video generation models. ERNIE 5.0, Baidu's 2.4-trillion-parameter unified multimodal model trained to handle text, image, video, and audio in a single framework, represents the scale of investment driving this wave.