PulseAugur
EN
LIVE 15:10:11
中文(ZH) 字节辟谣"群演数据喂大模型"：严重失实！

ByteDance denies using Hengdian extras' data for LLM training

By PulseAugur Editorial · [1 sources] ·

ByteDance has refuted claims that it used data from 150,000 extras from Hengdian World Studios to train its large language models. A spokesperson clarified that the project, aimed at building an AI service platform for the film industry, involves data assets owned by Hengdian or its licensors. Any use of personal likenesses or voices requires explicit consent, and the collected data will not be used for training ByteDance's large models. AI

IMPACT Clarifies data usage practices in AI development, addressing concerns about unauthorized data collection for model training.

RANK_REASON The item is a denial of a rumor about data usage for LLM training, not a direct announcement of a new model or product release.

Read on 雷峰网 (Leiphone) →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

ByteDance denies using Hengdian extras' data for LLM training

How we ranked this

Signal score
2 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Commentary
The item is a denial of a rumor about data usage for LLM training, not a direct announcement of a new model or product release.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. 雷峰网 (Leiphone) TIER_1 中文(ZH) ·

    ByteDance refutes 'large models fed by extras data': Seriously untrue!

    <p>近日，有网络传言称&quot;横店与字节合作，将15万名群演的肖像和声音喂给字节大模型&quot;。字节豆包公关负责人刘星今日通过微博作出回应，表示该传言严重失实。</p><p>据介绍，此次引发讨论的项目为横店影视AI服务平台建设项目。刘星表示，该项目的初衷是希望建设一套影视行业的公共数字基础设施，把影视基地实景等资产做成高质量数据集，帮助剧组更高效地选景、筹备和拍摄，同时降低中小影视企业使用AI的门槛，让更多创作者有条件把作品做好。</p><p>针对传言中的关键疑问，刘星澄清了三点：</p><p>第一，火山引擎在该项目中仅是技术支持方，所有数据资…

RELATED ENTITIES

RELATED TOPICS