Latest
  1. Report says AI is embedded across Home Office immigration tasks in near-total secrecy
  2. AI-generated ‘Xiaohai’ avatars allegedly used to cheat elderly viewers in China
  3. Stargate UAE’s first gigawatt would use triple the power of all UAE data centres
  4. UK government met 137 businesses on AI datacentres, no environmental groups
  5. SpaceX closes $40 billion in debt to buy GPUs, Musk says
  6. South Korean police bust ring accused of using AI to steal 8.2 million pieces of data
  7. HKU study ranks Chinese video AI models ahead of US rivals on visual quality
  8. IIT Madras whitepaper says AI adoption costs fall unevenly, with entry-level workers at risk

Latest / Frontier models and safety

ResearchFrontierMediaChina US Hong KongConfirmed

HKU study ranks Chinese video AI models ahead of US rivals on visual quality

See how this story connects

Source: South China Morning Post

Researchers at the University of Hong Kong ranked 12 leading AI video-generation models, finding ByteDance’s Seedance 2.0 Pro scored 3.84 out of 5 for visual quality, ahead of Google DeepMind’s Veo 3.1 at 3.47 and OpenAI’s Sora 2 Pro at 3.27. The study, led by Zhenhui Jack Jiang, found Seedance also led in visual consistency, physical realism, aesthetic quality and instruction compliance, and scored 3.86 against Veo’s 3.51 and Sora’s 3.35 when audio capabilities were included. Several Chinese models fell behind on safety safeguards, with Seedance placing sixth out of 11 for blocking harmful content.

Why it matters

The HKU ranking puts ByteDance’s Seedance 2.0 Pro ahead of Google DeepMind’s Veo 3.1 and OpenAI’s Sora 2 Pro on visual quality, while Seedance placed sixth out of 11 for blocking harmful content.

When
26 Oct 2026
ID
WTA-N-2026-0427
Who
University of Hong Kong, ByteDance, Google DeepMind, OpenAI; Zhenhui Jack Jiang

Summary by wheresthe.ai, from South China Morning Post. Checked against the source on 26 Oct 2026.

Read the original at South China Morning Post

Source copy: download the page as saved on 26 Oct 2026 (PDF). Also held by the Internet Archive.

Earlier in this Storyline: What led hereDeepSeek memory-saving technology may reduce model accuracy, researchers find

Cite this story

WTA-N-2026-0427: “HKU study ranks Chinese video AI models ahead of US rivals on visual quality.” wheresthe.ai, 26 Oct 2026. https://wheresthe.ai/d/cn-hku-study-ranks-chinese-video-ai-models-ahead-of-us-rival-2026-10/ (source: South China Morning Post).

Spotted an error? Email [email protected].