以下文章来源于伯乐在线 ，作者伯小乐 伯乐在线 . 伯乐在线分享IT互联网职场和精选干货文章（原域名已不再维护）。组织维护10万+star的开源技术资源 库，包括：Python, Java, C/C++, Go, JS, CSS, Node.js, PHP, .NET 等 谈到热门的自主学习技术，他透露 2025 年就有团队在尝试用实时数据训练，但效果一般。核心问题不是技术 不行，而是 "自主学习的瓶颈是想象力缺失" ，大家不知道它落地后该做什么具体事，得先明确比如盈利交易 系统、科学难题工具这类目标才行。 对于产业落地，姚顺雨建议大公司要靠自己的场景拿真实数据，别再依赖外部标注。同时强调，" 现在人与人 的差距，关键在会不会用 AI 工具 "，中国得好好普及这方面教育。他还预测，To B 领域的 Agent 技术会快速 增长，要是能实现全球企业部署，可能让 GDP 多涨 5%-10%。 最后他提到，中国 AI 团队有望全球领先，但得突破算力、To B 市场成熟度和创新文化这三个关键点。 （参考：网易新闻、 21 世经，本文经由 AI 大模型优化） 2026 年 1 月 10 日，前 OpenAI 核心研究员 ...

Core Insights - The article discusses the emergence of agent technology in retail, focusing on how it addresses consumer problems and enhances the shopping experience through automation and AI [1][6][10] Group 1: Agent Technology and Consumer Experience - Agent technology aims to simplify tedious shopping tasks by automating processes such as price tracking and purchasing on behalf of consumers [2][7] - Consumers can track product prices and receive alerts when prices drop, allowing for automated purchases, which can help merchants retain customers who might otherwise abandon their carts due to high prices [7][11] - The technology has been positively received by both consumers and merchants, as it increases consumer confidence in product availability and facilitates smoother transactions [4][8] Group 2: Implementation and Development - The technology is still in its early stages, with ongoing developments in AI capabilities, such as the Duplex system that allows AI to make calls on behalf of users [9][10] - The company has built a shopping graph containing 50 billion product listings, which supports the agent technology by providing extensive data on available products [13] - Future developments may include enhanced interactions between different agents, potentially allowing for more complex transactions and communications between merchants and consumers [14]