谷歌DeepMind首次测试双盲AI基准评估，防篡改新标准
AI benchmarks have a trust problem and Google wants to fix it
Google Deepmind is testing a double-blind evaluation of a frontier AI model for the first time. Cryptographic protection through Confidential Space is meant to keep Google from seeing the test questions and keep evaluators from seeing the model weights. The pilot project with the Singapore AI Safety Institute uses a Gemini Flash Lite and could set a new standard for tamper-proof AI benchmarks.
Google Deepmind 首次对前沿 AI 模型进行双盲评估测试。通过机密空间进行的加密保护旨在防止 Google 看到测试问题，并防止评估者看到模型权重。与新加坡 AI 安全研究所的试点项目使用了 Gemini Flash Lite，这可能为防篡改 AI 基准测试树立新标准。
The article AI benchmarks have a trust problem and Google wants to fix it appeared first on The Decoder.
文章《AI 基准测试存在信任问题，Google 希望解决它》最初发表于 The Decoder。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力