通用人工智能(AGI)

Search documents
蚂蚁集团联合中国人工智能学会发布AGI科研专项基金
Zheng Quan Ri Bao Wang· 2025-08-06 05:16
据了解,本次发布的基金,聚焦通用人工智能相关的关键技术和前沿方向,从底层发力,提升模型整体 的智能能力。例如,方向一AGI数据与评测,共开放3项课题,围绕AIGC视频评测、大模型高效数据蒸 馏、大模型动态评测和反污染检测展开,通过多种方式的数据生产处理与模型效果评测,提高AGI技术 的准确度、智能度和稳定性。方向二AGI基础模型,共开放18项课题,围绕多模态大模型交互体验、多 模态生成与理解一体化、高效注意力机制等课题展开。方向三AGIInfra共开放6项课题,围绕RL训推一 体、高性能AgenticRL、面向RL大模型推理加速等课题展开。 近年来,蚂蚁集团以AIFirst为战略,持续投入AGI,在基础研究攻关、产学研共建、开源开放等层面取 得了一定的进展。由蚂蚁发起的InclusionAI开源社区,持续开源了百灵基础大模型、强化学习推理框架 AReaL、多智能体框架AWorld等工作,登上了著名模型开源社区HuggingFace发布的中国开源热点地 图,其中百灵多模态大模型还获得了模型趋势(anytoany类型)榜第一。去年以来,蚂蚁集团接连与上 海交通大学、浙江大学、南京大学等知名高校成立了联合实验室,聚 ...
六年来首次开源,OpenAI放出两款o4-mini级的推理模型
Jin Shi Shu Ju· 2025-08-06 03:47
Core Insights - OpenAI has launched two open-source AI inference models, GPT-oss-120b and GPT-oss-20b, which are comparable in capability to its existing models [1][2] - The release marks OpenAI's return to the open-source language model space after six years, aiming to attract both developers and policymakers [2][3] Model Performance - In the Codeforces programming competition, GPT-oss-120b and GPT-oss-20b scored 2622 and 2516, respectively, outperforming DeepSeek's R1 model but slightly below OpenAI's own o3 and o4-mini models [2] - In the Human-Level Exam (HLE), the models achieved scores of 19% and 17.3%, surpassing DeepSeek and Qwen but still lower than o3 [3] - The "hallucination" rates for the GPT-oss models were significantly higher than those of OpenAI's latest models, with rates of 49% and 53% compared to 16% and 36% for o1 and o4-mini [3] Model Training Methodology - The GPT-oss models utilize a "Mixture-of-Experts" architecture, activating only a portion of their parameters for efficiency [5] - Despite having 117 billion parameters, GPT-oss-120b activates only 510 million per token, and both models underwent high-computational reinforcement learning [5] - Currently, the models only support text input and output, lacking multi-modal processing capabilities [5] Licensing and Data Transparency - GPT-oss-120b and GPT-oss-20b are released under the Apache 2.0 license, allowing commercial use without authorization [5] - OpenAI has chosen not to disclose the training data sources, a decision influenced by ongoing copyright litigation in the AI sector [6] Competitive Landscape - OpenAI faces increasing competition from Chinese AI labs like DeepSeek and Alibaba's Tongyi (Qwen), which have released leading open-source models [2] - The focus in the industry is shifting towards upcoming models from DeepSeek and Meta's Superintelligence Lab, indicating a rapidly evolving competitive environment [6]
赛博沙盒:如何与AI共创未来
3 6 Ke· 2025-08-06 03:01
Key Insights - The relationship between AI and gaming is historical, with AI development closely tied to early gaming, such as chess, which served as a foundational environment for AI research [4][5][6] - Gaming is viewed as a "womb" for AI, providing a safe space for AI to evolve and iterate, which is essential for its future development [5][6] - The integration of AI into gaming is expected to enhance game development processes, improve NPC interactions, and create new business models [11][12][14] - The low-code revolution is making game creation accessible to more individuals, allowing anyone to become a game creator [25][26] - Games are being recognized as valuable tools for social science research, providing a sandbox for testing theories and models in a controlled environment [27][28][30] Group 1 - AI's creative limitations and the need for breakthroughs in underlying algorithms are highlighted [3][4][6] - The concept of "algorithmic womb" suggests that games have historically nurtured AI development [5][6] - The future of AI in gaming includes the potential for intelligent NPCs and enhanced player experiences [11][12][14] Group 2 - The low-code movement is democratizing game development, enabling broader participation [25][26] - Games are being utilized as experimental platforms for social science, allowing for the simulation of real-world scenarios [27][28][30] - The integration of AI into gaming is expected to lead to innovative gameplay and narrative experiences [11][12][14]
LangChain CEO 再聊 Agent:chat 模式只是起点,Ambient Agents 才是未来
Founder Park· 2025-08-05 12:15
「现实中最靠谱的路径是 Agent+workflow 这种组合的优化。」 「workflow 本质上是工具,只是工具中用到了 AI 能力,所有被定义成 Work Flow 的就应该被做成工 具。」 一边是 Agent 遍地开花,另一边,创业者们还在争论到底 Agent 和 Workflow 孰优孰劣。 在 LangChain CEO Harrison Chase 看来,Agent 并不是「非黑即白」,而是像一个光谱。引用吴恩达的 观点,与其讨论一个东西是不是智能体,不如讨论它的「智能体化程度」(agenticness)。LLM 决定下 一步的程度越高,应用的「智能体化程度」就越高。 那 chatbot 是 Agent 最佳模式吗?未来到底是一个 Agent 还是很多 Agent?大家都在做 Agent,Agent 的 下一步会是什么方向? Harrison Chase,以及 企业 Agent 平台 Dust 的 CEO Stanislas Polu,在这个话题上还是有一些发言权的。 在最近的一期播客中,两人对 Agent 的下一步会怎么走进行了深入探讨。 以下为两人的对谈内容,Founder Park 编译了 ...
我们扒完了 GPT-5 全网爆料,奥特曼和 OpenAI 这次的饼真不好画了
3 6 Ke· 2025-08-05 10:39
Core Insights - OpenAI's marketing strategy for GPT-5 has faced criticism for being overly hyped without substantial product information [3][5][7] - The anticipated release of GPT-5 has been delayed, with various rumors about its launch date circulating since last year [8][23][26] - GPT-5 is expected to feature significant upgrades in multi-modal capabilities, software engineering, and AI agent functionalities [9][10][12][15] Group 1: Marketing and Anticipation - OpenAI has been criticized for its marketing tactics, which include frequent but vague updates about GPT-5, leading to public skepticism [3][5][7] - The excitement around GPT-5 has diminished despite OpenAI's continuous benchmark improvements, indicating a shift in marketing strategy to maintain interest [7][8] - Speculation about the release date has intensified, with predictions pointing towards a launch in early August 2025 [23][26][31] Group 2: Technical Advancements - GPT-5's core upgrades include a unified foundational and reasoning model, enhancing its performance in practical applications [9][12] - The model is expected to achieve "complete multi-modal" capabilities, allowing it to process and generate various types of media more effectively [10][11] - Significant improvements in software engineering capabilities will enable GPT-5 to modify and maintain complex enterprise-level codebases, challenging competitors like Anthropic [12][14] Group 3: AI Agent and Reasoning - GPT-5 is designed to execute complex multi-step tasks with reduced human supervision, marking a step towards autonomous AI agents [15][18] - The introduction of a "universal verifier" technology aims to enhance the model's ability to evaluate responses in subjective domains, improving its performance in creative writing and strategy analysis [16][18] - The model's architecture may include a smart routing system that dynamically selects the most suitable model for user queries based on complexity [22] Group 4: Development Challenges - The development of GPT-5 has faced challenges, including the limitations of traditional pre-training methods and the need for a new approach to scaling [40][41] - Internal projects like "Orion" did not meet expectations, leading to a pivot in strategy towards enhancing reasoning capabilities [40][41] - The successful development of the Q* technology has significantly improved the model's reasoning abilities, allowing it to tackle previously unseen problems [41][42] Group 5: Competitive Landscape - OpenAI's upcoming release of GPT-5 is crucial for regaining its competitive edge in areas like programming, where it has lost ground to competitors [50] - The partnership with Microsoft is aimed at leveraging OpenAI's technology to enhance Microsoft's own products while maintaining flexibility in their collaboration [32][35] - The competitive landscape is intensifying, with other companies like Anthropic and Google DeepMind also preparing to launch their advanced models [48][50]
扎克伯格15亿美元挖不动的男人
投中网· 2025-08-05 06:37
以下文章来源于智东西 ,作者王 涵 智东西 . 智能产业新媒体!智东西专注报道人工智能主导的前沿技术发展,和技术应用带来的千行百业产业升 级。聚焦智能变革,服务产业升级。 将投中网设为"星标⭐",第一时间收获最新推送 安德鲁·塔洛克,现任Thinking Machines Lab联合创始人兼首席研究员。 OpenAI前CTO带队"团拒"。 作者丨 王涵 编辑丨 漠影 来源丨 智东西 智东西8月4日消息,据外媒《华尔街日报》近日报道,知情人士透露,与OpenAI前首席技术官米拉 ·穆拉蒂(Mira Murati)联合创办Thinking Machines Lab的安德鲁·塔洛克(Andrew Tulloch),拒绝了扎克伯格的可能高达 15亿美元(约合人民币108.2亿元) 的薪酬。 据报道,试图收购Thinking Machines Lab失败后,扎克伯格在随后数周内接触了该公司 约十几名 员工,最终成功 带走0个人 。 《华尔街日报》报道称,Meta发言人安迪·斯通(Andy Stone)回应称该薪酬描述"失实且荒谬", 强调任何薪酬方案都需以股价上涨为前提,并声明Meta无意收购Thinking Mach ...
OpenAI引信任危机:AI教父与诺奖得主等联名,要求回应八大问题
3 6 Ke· 2025-08-05 05:30
Core Viewpoint - A coalition of NGOs has expressed concerns regarding OpenAI's planned restructuring, fearing it may undermine its non-profit mission in favor of profit maximization [1][4]. Group 1: Signatories and Concerns - The open letter has been signed by notable figures, including four Nobel laureates, who are worried about OpenAI potentially prioritizing profits over its original charitable mission [1][2]. - The letter includes former OpenAI employees who accuse the organization of planning to transition into a public benefit corporation, which could weaken safeguards like profit caps and non-profit oversight [3][4]. Group 2: Transparency Issues - OpenAI has been criticized for failing to meet its transparency obligations, including changes to its core commitments and suppressing internal dissent through confidentiality agreements [4][6]. - The restructuring plan has raised further transparency concerns, as it aims to shift from a mixed structure of non-profit and for-profit to a conventional for-profit model, potentially diminishing its legal obligations to prioritize public interest [4][5]. Group 3: Key Questions Raised - The coalition has posed eight critical questions to OpenAI, seeking clarification on its commitment to its charitable mission, non-profit control, and the implications of its restructuring on profit distribution and governance [5][8]. - Specific inquiries include whether OpenAI will maintain its profit cap and how it plans to handle excess profits in relation to its original mission [7][8]. Group 4: Broader Implications - The coalition emphasizes that OpenAI's governance decisions will not only shape its future but also have significant societal implications, particularly in the context of AGI's impact on humanity [6][9]. - The call for transparency is framed as essential for restoring public trust and ensuring that OpenAI remains accountable to its founding principles [6][9].
高德地图2025正式发布,全球首个基于地图的AI原生智能体
Guo Ji Jin Rong Bao· 2025-08-04 12:35
近日,阿里巴巴集团旗下的高德地图宣布全面AI化,基于超二十年的数据、技术和场景研发经验积累,推出了全球首个AI原生地图应用:高德地图 2025,即日起用户升级高德地图APP至最新版,搜索"空间智能",即可体验。 该旗舰版本旨在融合AI技术,把地图导航服务体验推升至另一次元,演进为具备深度时空理解和自主推理决策能力的出行生活智能体,将空间智能普及 至瞬息万变的现实场景。 "不同于语言智能,空间智能是在三维空间和时间中感知、推理、和行动的能力,也意味着我们对'连接真实世界'的使命演绎,将进一步跃迁至'理解'真实 世界。"高德地图CEO郭宁表示,希望从高德地图2025开始,推动AI从"对话工具"蜕变为"行动伙伴"。 高德空间智能:从"连接"到"理解",地图里的另一个世界 空间智能,是一套构建虚拟世界并深度理解、推理决策、总结输出的AI技术分支,以实现虚拟数字世界与真实物理世界的无缝互联,被行业普遍认为是 通向AGI(通用人工智能)的关键。 据悉,支持这一整套智能体架构的高德和通义共建大模型簇,由多模态空间感知、行为认知、时空意图理解、总结输出交互和反馈观测评估等多个专业模 型构成。 而在其背后,考虑到多场景智能体的 ...
揭秘:OpenAI是如何发展出推理模型的?
硬AI· 2025-08-04 09:46
硬·AI 作者 | 龙 玥 编辑 | 硬 AI 当全世界都在为ChatGPT的横空出世而狂欢时,你可能不知道,这只是OpenAI一次"无心插柳"的惊喜。科 技媒体Techcrunch一篇最新的深度文章揭示了, OpenAI从数学竞赛走向"通用AI智能体"(AI Agents) 的宏大愿景 。这背后,是一个长达数年的深思熟虑的布局,以及其对AI"推理"能力的终极探索。 01 意外的起点:数学 很多人以为OpenAI的成功故事是从ChatGPT开始的,但真正的颠覆性力量,却源于一个看似与大众应用 相去较远的地方——数学。 2022年,当研究员亨特·莱特曼(Hunter Lightman)加入OpenAI时,他的同事们正在为ChatGPT的发布 而忙碌。这款产品后来火遍全球,成为现象级的消费应用。但与此同时,莱特曼却在一个不起眼的团 队"MathGen"里,默默地教AI模型如何解答高中数学竞赛题。 让OpenAI名声大噪的ChatGPT,可能只是一次"美丽的意外"。在其内部,一个始于数学、代号"草莓"的宏大计划,已悄 然掀起一场"推理"革命。其终极目标是创造出能自主处理复杂任务的通用AI智能体。"最终,你只需告诉计 ...
奥特曼:ChatGPT只是意外,全能AI智能体才是真爱,Karpathy:7年前就想到了
3 6 Ke· 2025-08-04 09:37
Core Insights - The article highlights the evolution of OpenAI's MathGen team, which has been pivotal in enhancing AI's mathematical reasoning capabilities, leading to significant advancements in AI agents [2][6][9] - OpenAI's CEO, Altman, emphasizes the transformative potential of AI agents, which are designed to autonomously complete tasks assigned by users, marking a strategic shift in AI development [11][28] - The competition for top talent in AI has intensified, with major companies like Meta aggressively recruiting from OpenAI, indicating a fierce race in the AI sector [13][15][36] Group 1: Development of AI Capabilities - The MathGen team, initially overlooked, is now recognized as a key contributor to OpenAI's success in the AI industry, particularly in mathematical reasoning [2][4] - OpenAI's recent breakthroughs in AI reasoning have led to its model winning a gold medal at the International Mathematical Olympiad (IMO), showcasing its advanced capabilities [6][20] - The integration of reinforcement learning and innovative techniques has significantly improved AI's problem-solving abilities, allowing it to tackle complex tasks more effectively [17][21][25] Group 2: Strategic Vision and Market Position - OpenAI's long-term vision is to create a general AI agent capable of performing a wide range of tasks, which is seen as the culmination of years of strategic planning [8][9][11] - The upcoming release of the GPT-5 model is expected to further solidify OpenAI's leadership in the AI agent space, with ambitions to create an intuitive assistant that understands user intent [35][39] - The competitive landscape is becoming increasingly crowded, with various companies vying for dominance in AI technology, raising questions about OpenAI's ability to maintain its edge [36][38]