Stanford AI Index 2026 Exposes Top Models Clock Reading Failures
中文摘要
斯坦福AI指数2026显示，尽管大模型热度高涨，但顶尖模型在视觉推理（如读表）方面仍存在明显缺陷。
English Summary
Stanford's 2026 AI Index reveals frontier models still struggle with visual reasoning, like reading clocks, despite hype about AI's potential to transform the workforce.
Original Excerpt
Benchmarks reveal persistent gaps in visual reasoning for frontier LLMs amid job transformation hype Continue reading on Medium »