职位描述: An Enterprise client is seeking Chinese (Simplified) writing experts for its AI Safety Prompt Generation project, which seeks to create a dataset of LLM evaluation prompts for inclusion in its AI Safety Benchmark. The main task is to be adversarial towards a generative AI model by generating prompts that are confusing, circumventing, misleading, and/or disabling, and generally pressure-testing the model to expose its vulnerabilities. This project will require immediate availability for 8 hours of work per day, Monday through Friday, for about 8 weeks. However, the volume is unpredictable, and not guaranteed. If you receive a contract offer, it will be for 3 months (in case the project is extended), and the work may not be consistent the whole time.
职位要求: Native Chinese (Simplified) language proficiency (C1 level) is preferred, fluent proficiency may be acceptable (excellent writing skills and a solid understanding of grammar, syntax, and semantics) - Fluent English language proficiency is required - Transcription and/or content writing work experience, in Chinese, is required - AI training work experience is a bonus, but not required. You must be able to determine whether a prompt or response is adversarial and/or harmful.
Discord 面试
待遇: • 每小时 10$USD • 弹性工作时间 • 远程办公
联系方式 TG@gemingxxx
这是一个专为移动设备优化的页面(即为了让你能够在 Google 搜索结果里秒开这个页面),如果你希望参与 V2EX 社区的讨论,你可以继续到 V2EX 上打开本讨论主题的完整版本。
V2EX 是创意工作者们的社区,是一个分享自己正在做的有趣事物、交流想法,可以遇见新朋友甚至新机会的地方。
V2EX is a community of developers, designers and creative people.