DeepMind论文：AGI通往ASI的四条技术路径
Beautiful paper from Google DeepMind.
Beautiful paper from Google DeepMind.
来自 Google DeepMind 的精彩论文。
Explains the pathways from AGI to ASI, and why that jump could happen through several routes.
解释了从 AGI 到 ASI 的路径，以及为什么这种跳跃可能通过多种途径发生。
The authors frame the AGI-to-ASI transition around 4 technical pathways:
作者将 AGI 到 ASI 的转变围绕 4 条技术路径进行阐述：
- continued scaling of compute, model size, data, and test-time inference;
- algorithmic paradigm shifts beyond today’s transformer-based foundation-model stack;
- recursive self-improvement, where AI accelerates AI R&D and improves future systems; and
- multi-agent collective intelligence, where large populations of specialized agents coordinate into a superhuman group agent.
- 持续扩展计算、模型规模、数据和测试时推理；
- 超越当前基于 Transformer 的基础模型堆栈的算法范式转变；
- 递归自我改进，即 AI 加速 AI 研发并改进未来系统；以及
- 多智能体集体智能，即大量专业智能体协调成一个超人的群体智能体。
Scaling may work for a while, but it could hit limits in data, compute, energy, or weaker returns from making systems larger.
扩展可能在一段时间内有效，但可能会在数据、计算、能源或扩大系统规模带来的收益递减方面遇到限制。
Recursive improvement is the most uncertain path, because AI could speed up AI research, but that loop may also slow if hard research problems need real-world testing, scarce hardware, or new ideas.
递归改进是最不确定的路径，因为 AI 可能加速 AI 研究，但如果困难的研究问题需要现实世界测试、稀缺硬件或新想法，这个循环也可能放缓。
Multi-agent collectives may be the most underappreciated path, because a society of competent digital workers could outperform a brilliant individual model through specialization, speed, and coordination.
多智能体集体可能是最被低估的路径，因为一个由有能力的数字工作者组成的社会可以通过专业化、速度和协调超越一个杰出的个体模型。
The big point is that ASI may not arrive as 1 sudden event, but as a chain of faster changes as AI helps create better AI and stronger scientific tools.
重点在于，ASI 可能不会作为一个突发事件到来，而是作为一系列加速变化，因为 AI 帮助创造更好的 AI 和更强大的科学工具。
- arxiv. org/abs/2606.12683
- arxiv. org/abs/2606.12683
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