Tencent's SilentMind model faces scrutiny over built-in content filters
Shortly after its announcement, Tencent's SilentMind model became a top search trend, largely because its built-in filters block content the company deems sensitive or political. The filters reportedly act on both prompts and outputs, returning generic refusals or truncating responses that contain restricted material. The article outlines categories of blocked terms, including topics tied to Chinese politics, recent events, and sexuality and religion. It also describes developer workarounds such as comparing outputs across several models and rephrasing prompts, while warning that these methods may breach Tencent's terms of service.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in