Gary Marcus：神经符号混合已成主流，纯神经网络路线被边缘化
Gary Marcus 直接点名 Hinton：纯神经网络路线输了。他用 Lean 检查 LLM、Claude Code 里的符号构件当证据，解释为什么现在的 AI 系统都离不开经典符号工具。做 agent 或编程助手的开发者读一下，能看懂自己的 harness 里到底装了什么。
Gary Marcus 在 X 上发文重申其长期观点：理解当代 AI 必须区分“作为组件的模型”和“作为整体的系统”，而包含神经网络与符号的系统在定义上就是神经符号系统。他用引擎与汽车作类比，指出仅靠神经网络难以构建可靠、健壮的 AI 系统。
他以数学领域为例：Lean 定理证明器会检查甚至拒绝 LLM 的输出，Claude Code 也大量依赖规则、代码解释器、函数、条件与循环等源自经典符号 AI 的构件。数学系统使用的自动定理证明器直接承袭 Newell 和 Simon 在 1950 年代的符号主义工作。
Marcus 指出，Geoffrey Hinton 曾将符号比作“以太”、主张纯神经网络路线，这一立场被证明是错误的。他引用 François Chollet 的观点称，扩展（scaling）确有帮助但不充分，符号工具、代码解释器和 harness 已成为多数实用系统的核心组件，神经符号混合路线已全面胜出。
The 🗝️ fact: You can’t understand contemporary AI unless you can appreciate the difference between a model (that is part of a system) and a system as a whole. If a system as a whole includes both neural networks and symbols, it is by definition neurosymbolic. To think otherwise is like thinking that an engine IS a car, rather than a component of a car. You can’t make a car out of engines alone. and you can’t (in most cases) make robust, reliable AI systems out of neural networks alone. Instead, in many domains you get better results by *integrating* both neural networks and symbols. (Math, where e.g., Lean checks and sometimes rejects the work of LLMs is a good example of this; so is Claude Code). That was literally the point of my 2001 book The Algebraic Mind and has been a central point throughout my career. For a long time, people like @GeoffreyHinton argued otherwise, suggesting that all we needed were neural networks by themselves, and that symbols were an archaic waste of time. Hinton likened symbols to “aether”, as if they were complete scientific mistake. He was wrong. Systems like Claude Code depend heavily on combining neural networks with things like rules, code interpreters, functions, conditionals and loops - all symbolic constructs drawn from classical AI. Math systems use automated theorem provers, which flow directly from Newell and Simon’s symbolic work in the 1950s. Symbols weren’t a mistake; they were vital, just as neural networks are. Scaling helped, but by itself it wasn’t enough; symbolic tools and code interpreters and harnesses are now a central part of most working systems. Neurosymbolic AI has won, overwhelmingly; the pure neural networks that the dogmatists like Hinton used to argue for have been almost entirely displaced by neurosymbolic hybrids. As @fchollet has also noted, to pretend otherwise is just silly.📊 2 ⚡
本条情报归属 AITOP「行业」核心栏目，重点聚焦北美前沿实验室与硅谷科技巨头。在跨区域产业观察坐标中，该动态直接关联「中美大模型技术代差与商业化对照」「AI 开发者工具链、Coding Agent 与工程化实践」研究主线，反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括：Gary Marcus、GeoffreyHinton、Claude Code、神经符号，读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明：本文原始线索采编自「Gary Marcus」（@GaryMarcus），经 AITOP 多维评分管线结构化提炼。
