[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fmz9u-2q3d6ZUJRTrraMm9x9Ct3XbbHpl666IjPrOle0":3},{"code":4,"msg":5,"data":6},200,"操作成功",{"id":7,"title":8,"content":9,"digest":10,"source":10,"coverPath":11,"thumbsCoverPath":12,"isTop":13,"isShow":14,"baseClick":13,"clickCount":15,"createTime":16,"typeId":17,"isNewest":18,"newsInfoTypeRespVo":19,"voiceUrl":22,"voiceSize":23,"taskId":24,"releaseTime":25,"titleEn":26,"contentEn":27,"voiceUrlEn":28,"taskIdEn":29,"voiceSizeEn":30},1166,"AI为核心的“虚拟实验室”创建，旨在解决复杂问题并提高科研效率","\u003Cp>\u003Cstrong class=\"ql-lineHeight-1-75\" style=\"font-size: 18px; color: rgb(255, 153, 0);\">美国斯坦福大学医学院团队开发出一种名为“虚拟实验室”的创新工具。该系统以人工智能（AI）为核心，结合跨学科科学家团队，旨在解决复杂问题并加速科学发现进程，可在多领域极大提高科研效率。相关论文29日发表于《自然》杂志。\u003C\u002Fstrong>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cimg alt=\"undefined\" src=\"https:\u002F\u002Fimage.51xinwei.com\u002F2025\u002F07\u002F244086c2deef431595eb38bd8a7aac59\u002F377353_f0ca27de-b3d4-4b49-ad5c-21933555296acopy.jpg\" width=\"528\" height=\"undefined\" style=\"display: block; margin: auto;\" class=\"ql-align-center\">\u003Cp class=\"ql-align-center\">\u003Cspan class=\"ql-lineHeight-1-75\" style=\"color: rgb(187, 187, 187);\">图片由AI生成\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">“虚拟实验室”的运作模式与现实实验室类似：人类科学家提出研究问题后，由人工智能首席研究员（AI PI）主导项目。AI PI再根据课题需求生成多领域代理团队，例如在疫苗研发中，系统生成了免疫学、计算生物学和机器学习等领域的代理。每个项目还配备专门的“评论家”代理，负责批判性评估和提醒潜在风险。团队为虚拟科学家配备了“阿尔法折叠”等蛋白质建模工具，支持其进行创造性分析，并允许代理提出工具需求以完善研究框架。\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">“虚拟实验室”的运行效率远超传统模式。其会议和讨论在秒级内完成，且多线程并行进行，无需休息或资源消耗。人类团队仅在预算限制和项目方向上进行宏观指导，干预率低于1%，确保AI的自主创造力。所有虚拟交互均通过记录系统存档，便于人类实时跟踪并在必要时调整方向。\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">该系统已在疫苗设计中展现出很大的潜力。例如，面对病毒新变种挑战，AI团队摒弃传统抗体方案，选择更小型的纳米抗体作为设计方向。实验验证显示，AI设计的纳米抗体不仅结构稳定，与病毒刺突蛋白的结合能力优于现有抗体，且具备跨变种有效性——既可针对原始株，亦能有效结合新变种。研究团队正将实验数据反馈至系统，以迭代优化分子设计。\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">除疫苗领域外，团队还开发了数据分析型代理，用于重新评估已发表论文的复杂数据集。团队表示，传统科研中不同背景专家的协作与当前AI代理技术的进步，共同催生了这一突破性尝试。而其表现出的深度挖掘能力，往往能揭示传统研究中未能触及的新发现，标志着“虚拟实验室”在多学科应用中的广阔前景。\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cstrong style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">总编辑圈点\u003C\u002Fstrong>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">一支人类组成的科研团队是这样开展工作的：核心人物把握战略方向，不同技术路线的小组在负责人带领下开展科研攻关，大家随时互相沟通，同步各自进展，核心人物再适时调整方向。在AI组成的虚拟实验室中，一切都加快了。AI科学家团队可以24小时无休，团队之间的沟通在秒级时间内就能完成，他们还能提供思维定势之外的解法。曾经，我们用AI作为辅助工具；如今，AI可以直接做科研。未来已至，科研范式将迎来新一轮革命。\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"color: rgb(187, 187, 187);\">【新闻来源】科技日报 记者 张梦然 \u003C\u002Fspan>\u003Ca href=\"https:\u002F\u002Fwww.stdaily.com\u002Fweb\u002Fgjxw\u002F2025-07\u002F29\u002Fcontent_377353.html\" rel=\"noopener noreferrer\" target=\"_blank\" style=\"color: rgb(187, 187, 187);\">https:\u002F\u002Fwww.stdaily.com\u002Fweb\u002Fgjxw\u002F2025-07\u002F29\u002Fcontent_377353.html\u003C\u002Fa>\u003C\u002Fp>\u003Cp class=\"ql-align-justify\">\u003Cspan style=\"color: rgb(187, 187, 187);\">（本网转发此文章，旨在为读者提供更多的信息资讯，所涉内容不构成投资、消费建议。文章事实如有疑问，请与有关方核实，文章观点非本网观点，仅供读者参考。）\u003C\u002Fspan>\u003C\u002Fp>","","https:\u002F\u002Fimage.51xinwei.com\u002F2025\u002F07\u002F367da6414bb54bca8b58554c3b96ee27\u002FAI领域.jpg","https:\u002F\u002Fimage.51xinwei.com\u002F2025\u002F07\u002Fthumbs\u002F367da6414bb54bca8b58554c3b96ee27\u002FAI领域.jpg",0,1,212,"2025-07-31 18:24",2,false,{"id":17,"name":20,"enName":21},"芯位视野","Xinwei Vision","https:\u002F\u002Fxinwei-dev-test.oss-cn-shenzhen.aliyuncs.com\u002Fintelligent\u002Faudio%3A60e76f6e-77b2-4783-ae67-de4d7a79a1bb%3A0.wav?Expires=1754012785&OSSAccessKeyId=LTAI5tNvY2RkKjZw4LLWsrPK&Signature=LD0Wi0E1rG7xk%2BeH7At4%2B5nGkVM%3D",5394012,"60e76f6e-77b2-4783-ae67-de4d7a79a1bb","2025-07-31 18:21","AI-driven \"Virtual Laboratory\" is created to solve complex problems and improve research efficiency.","\u003Cp>\u003Cstrong class=\"ql-lineHeight-1-75\" style=\"font-size: 18px; color: rgb(255, 153, 0);\">A team from the School of Medicine at Stanford University has developed an innovative tool called \"Virtual Laboratory.\" The system, centered on artificial intelligence (AI), combines a multidisciplinary team of scientists, aiming to solve complex problems and accelerate the process of scientific discovery, which can greatly improve research efficiency in multiple fields. The related paper was published in the journal Nature on the 29th.\u003C\u002Fstrong>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cimg alt=\"undefined\" src=\"https:\u002F\u002Fimage.51xinwei.com\u002F2025\u002F07\u002F244086c2deef431595eb38bd8a7aac59\u002F377353_f0ca27de-b3d4-4b49-ad5c-21933555296acopy.jpg\" width=\"528\" height=\"undefined\" style=\"display: block; margin: auto;\" class=\"ql-align-center\">\u003Cp class=\"ql-align-center\">\u003Cspan class=\"ql-lineHeight-1-75\" style=\"color: rgb(187, 187, 187);\">The image was generated by AI\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">The operation mode of the \"Virtual Laboratory\" is similar to that of a real laboratory: after human scientists propose research questions, the AI Chief Investigator (AI PI) leads the project. The AI PI then generates multi-disciplinary agent teams according to the needs of the topic, for example, in vaccine development, the system generated agents in fields such as immunology, computational biology, and machine learning. Each project also has a specialized \"critic\" agent responsible for critical evaluation and reminding of potential risks. The team equipped virtual scientists with protein modeling tools like \"AlphaFold,\" supporting creative analysis and allowing agents to request tool requirements to improve the research framework.\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">The operational efficiency of the \"Virtual Laboratory\" far exceeds traditional models. Its meetings and discussions are completed within seconds, and they run in parallel threads without needing rest or resource consumption. Human teams only provide macro-level guidance within budget constraints and project direction, with an intervention rate below 1%, ensuring the AI's autonomous creativity. All virtual interactions are archived through the recording system, making it easy for humans to track in real time and adjust directions when necessary.\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">The system has shown great potential in vaccine design. For example, facing the challenge of new virus variants, the AI team abandoned traditional antibody approaches and chose smaller nanobodies as the design direction. Experimental validation showed that the nanobodies designed by AI not only have stable structures but also better binding ability with viral spike proteins than existing antibodies, and they have cross-variant effectiveness—capable of targeting the original strain and effectively binding to new variants. The research team is feeding experimental data back into the system to iteratively optimize molecular design.\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">In addition to the field of vaccines, the team also developed data analysis agents used to reassess complex datasets from published papers. The team stated that the collaboration of experts with different backgrounds in traditional research and the advancement of current AI agent technology together led to this breakthrough attempt. Their ability to deeply mine information often reveals new discoveries that traditional research could not reach, marking the broad prospects of the \"Virtual Laboratory\" in multi-disciplinary applications.\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cstrong style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">Editor's Note\u003C\u002Fstrong>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">A research team composed of humans works like this: the core person sets the strategic direction, different technical route groups carry out scientific research under the leadership of their leaders, and everyone communicates at any time, synchronizing their progress, and the core person adjusts the direction in a timely manner. In the AI-based virtual laboratory, everything is accelerated. AI scientist teams can work 24\u002F7 without breaks, and communication between teams can be completed within seconds. They can also provide solutions outside of conventional thinking. Previously, we used AI as an auxiliary tool; now, AI can directly conduct research. The future has arrived, and the research paradigm will undergo a new round of revolution.\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"color: rgb(187, 187, 187);\">【News Source】Science and Technology Daily, Reporter Zhang Mengran \u003C\u002Fspan>\u003Ca href=\"https:\u002F\u002Fwww.stdaily.com\u002Fweb\u002Fgjxw\u002F2025-07\u002F29\u002Fcontent_377353.html\" rel=\"noopener noreferrer\" target=\"_blank\" style=\"color: rgb(187, 187, 187);\">https:\u002F\u002Fwww.stdaily.com\u002Fweb\u002Fgjxw\u002F2025-07\u002F29\u002Fcontent_377353.html\u003C\u002Fa>\u003C\u002Fp>\u003Cp class=\"ql-align-justify\">\u003Cspan style=\"color: rgb(187, 187, 187);\">（This article is reprinted by this website to provide readers with more information and news. The content does not constitute investment or consumer advice. If there are any doubts about the facts of the article, please verify with the relevant parties. The views expressed in the article are not the views of this website and are for reference only.）\u003C\u002Fspan>\u003C\u002Fp>","https:\u002F\u002Fxinwei-dev-test.oss-cn-shenzhen.aliyuncs.com\u002Fintelligent\u002Faudio%3A8671fc4d-af66-41b9-a089-76f122707ca0%3A0.wav?Expires=1774838507&OSSAccessKeyId=LTAI5tNvY2RkKjZw4LLWsrPK&Signature=nKlSZUfBjEqa36cuW1PRPQ1WUTw%3D","8671fc4d-af66-41b9-a089-76f122707ca0",7556202]