12月13日，在中国神经科学学会脑机接口与交互分会年会上，天桥脑科学研究院宣布成立尖峰智能实验 室（Spiking Intelligence Lab, SIL），致力于类脑大模型和脉冲神经网络的研发，探索人工智能与人类智 慧的深度融合。 第一财经记者从会议现场了解到，最新成立的尖峰智能实验室属于非营利研究机构，由中国科学院自动 化研究所研究员李国齐教授领衔，目标是为打造天桥脑科学研究院创始人陈天桥提出的"发现式智能"提 供关键能力。 李国齐对第一财经记者表示："发现式智能的一个关键能力是神经动力学。不同于当前依赖规模法则堆 叠参数的主流AI模式，尖峰智能实验室主张借鉴人脑这一自然界最精巧的智能载体，重点研发具有神 经动力学特性的类脑大模型，将脉冲通信、时空动态编码等计算特性与树突神经元的精细结构深度耦 合，构建一个既具备强大感知力，又拥有深刻记忆与思考能力的全脑架构。" 这一路径也是陈天桥此前提出的实现通用人工智能的"结构路径"。人脑以仅约20瓦的功耗支撑起千亿级 神经元的复杂运作，李国齐从事的脉冲神经网络和类脑大模型的研究，将为构建这样的全脑架构提供基 础研究及转化方面的支持。 陈天桥最近提出，仅靠数据和算 ...

Core Viewpoint - The EverMind team, founded by Chen Tianqiao, has launched EverMemOS, a world-class open-source long-term memory system for AI agents, aiming to establish a data infrastructure for future intelligent systems [3][4]. Group 1: Long-term Memory in AI - Long-term memory is seen as a strategic feature for AI applications, with Chen Tianqiao asserting that memory will be a core competitive advantage for AI, transitioning from tools to intelligent agents [3][4]. - The design of EverMemOS is inspired by human memory mechanisms, allowing AI to think, remember, and grow like humans [3][5]. Group 2: Discovering Intelligence - Chen Tianqiao emphasizes the concept of "discovery intelligence," which includes five core capabilities, with long-term memory being a crucial aspect [4][5]. - He argues that true discovery involves asking questions and understanding principles rather than merely predicting outcomes [4][5]. Group 3: Paths to Discovery Intelligence - Two paths to achieving discovery intelligence are proposed: the "scale path," which focuses on large models and data, and the "structure path," which emphasizes dynamic and continuous understanding akin to cognitive anatomy [5][6]. - The limitations of the scale path are acknowledged, with a call for new theories and interdisciplinary approaches to advance AI understanding [5][6]. Group 4: Investment in Research - The Tianqiao Brain Science Research Institute plans to invest over $1 billion in dedicated computing clusters to support young scientists in exploring memory mechanisms and new causal architectures [6][7]. - An incubator has been established to provide independent research opportunities for young scientists, allowing them to lead their own projects without waiting for graduation [7].

Core Insights - The concept of "memory" is viewed as a critical competitive advantage for future AI applications, marking a transition from tools to intelligent agents and from passive responses to proactive evolution [1][3] - EverMind, a team under Chen Tianqiao's Shanda Group, has launched EverMemOS, a world-class open-source long-term memory system aimed at becoming the data infrastructure for future intelligent agents [1][3] - Chen Tianqiao emphasizes the importance of "discovery-based intelligence," advocating for a research paradigm that endows AI with a persistent, coherent, and evolvable "soul" [1][3] AI Memory Systems - Current AI applications, including Claude and ChatGPT, have integrated long-term memory as a strategic feature, reflecting a broader trend in the industry [3] - EverMemOS is inspired by human memory mechanisms, mimicking the encoding, indexing, and long-term storage processes of the human brain [3][4] - The EverMind team aims to tackle one of the most profound challenges in AI: enabling machines to possess memory, which is seen as a gateway to higher levels of general intelligence [3][4] Discovery-Based Intelligence - Chen Tianqiao argues that true "discovery" in AI should involve posing questions and understanding principles rather than merely predicting outcomes [4] - He outlines two pathways to achieving "discovery-based intelligence": the "scale pathway," which focuses on the size of models and data, and the "structure pathway," which emphasizes dynamic and continuous cognitive processes [4][5] - The "scale pathway" has led to significant applications in AI, such as protein prediction and compound generation, but is viewed as limited in its ability to foster true understanding [5] Research and Development Initiatives - The Tianqiao Brain Science Research Institute plans to invest over $1 billion in dedicated computing clusters to support young scientists in exploring structural intelligence [6] - The institute has established a PI incubator, allowing young researchers to have independent budgets and lead their own experiments without waiting for graduation [6] - Chen Tianqiao stresses the need for interdisciplinary collaboration among neuroscience, information theory, physics, and cognitive psychology to drive innovation in AI [5][6]

Core Insights - The article discusses the AI Accelerated Science Symposium held in San Francisco, where the concept of "Discoverative Intelligence" was introduced as a new paradigm for general artificial intelligence [1][3][4] - The speaker, Chen Tianqiao, emphasizes that AI should not merely replace human jobs but should aid in human evolution by helping discover the unknown [5][10] Group 1: Human Evolution and AI - Human evolution has not stopped; instead, it has transformed through scientific discoveries and technological inventions, extending human capabilities beyond biological limits [3][4] - The concept of "Discoverative Intelligence" is presented as a true form of general artificial intelligence, which can actively construct testable theoretical models and propose falsifiable hypotheses [5][10] Group 2: Paths to Discoverative Intelligence - Two main paths to achieving "Discoverative Intelligence" are identified: the "Scale Path," which relies on large models and data, and the "Structure Path," which focuses on cognitive mechanisms akin to human brain functions [6][12] - The "Scale Path" has achieved significant results in AI applications, while the "Structure Path" is emerging as a necessary approach to overcome the limitations of current AI systems [13][14] Group 3: Time Structure and Core Capabilities - The article outlines five core capabilities essential for managing information over time, which are necessary for achieving "time structure" in AI: neural dynamics, long-term memory, causal reasoning, world modeling, and metacognition [8][9][12] - These capabilities form a continuous and active loop, enabling a system to evolve over time and engage in scientific discovery [12] Group 4: Opportunities for Young Researchers - The article highlights the need for new theories, algorithms, and interdisciplinary approaches, positioning young researchers as key players in redefining intelligence through the "Structure Path" [13][14] - The company is investing over $1 billion in dedicated computing clusters to support young scientists in exploring new structures and validating cognitive mechanisms [16]

Core Insights - The AI-driven scientific symposium held in San Francisco gathered top scholars and industry leaders to discuss how AI can drive scientific discovery [1][3] - Chen Tianqiao, founder of the Tianqiao Brain Science Research Institute, introduced the concept of "Discoverative Intelligence," which he argues represents true general artificial intelligence [1][4] Group 1: Discoverative Intelligence - Discoverative Intelligence is defined as the ability to ask questions and understand principles, rather than merely predicting outcomes [6][7] - This form of intelligence is seen as essential for human evolution, emphasizing AI's role in helping humans discover the unknown [6][8] Group 2: Paths to Discoverative Intelligence - Two main paths to achieving Discoverative Intelligence are identified: the "scale path," which focuses on the size of models and data, and the "structure path," which emphasizes cognitive mechanisms akin to human brain functions [8][10] - The scale path has led to significant advancements in AI applications, while the structure path is emerging as a necessary complement to achieve deeper understanding and discovery [8][13] Group 3: Time Structure Analysis - The concept of "time structure" refers to the brain's ability to process information dynamically over time, contrasting with the static nature of current AI models [9][10] - Five core capabilities are essential for managing information over time: neural dynamics, long-term memory, causal reasoning, world modeling, and metacognition [10][11] Group 4: Opportunities for Young Researchers - The company plans to invest over $1 billion in dedicated computing clusters to support young scientists in exploring new theories and algorithms [13][14] - A new benchmark will be established to measure AI's ability to "discover," focusing on the five core capabilities necessary for true intelligence [13][14]