• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
Science
Reaction

AI’s potential to reshape scientific discovery

In an essay for the Google DeepMind Institute, the authors envision LLMs coordinating specialized tools like AlphaFold.

Alex ImasAI
星星之火可以燎原星星
2 Sources, 3h ago, first seen 3h ago

TLDR

An essay for the Google DeepMind Institute, written with James Manyika, argues that AI could do more than speed up existing research: it could change how discoveries are made. The authors envision LLMs calling specialized models, checking results and repeating the process until a step such as wet-lab testing requires people. A separate commenter argues that such workflows need reproducible evidence and independent checks, so successful predictions are not mistaken for understood mechanisms.

Combined views

14.2K

2 Sources, first seen 3h ago

123 likes11 comments68 saves30 reposts

Combined views

14.2K

2 Sources, first seen 3h ago

123 likes11 comments68 saves30 reposts

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Featured Source

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

2 Sources

Alex Imas@alexolegimasNew essay for the @GoogleDeepMind Institute with James Manyika on AI x Science, "Bending the Curve of Discovery". While accelerating existing scientific practices is certainly useful, the real promise of AI is its potential to act as an invention of a method of invention (IMI) a la Griliches--e.g., the microscope or statistical inference--which would unlock questions and whole modes of discovery that were previously beyond human reach. Today, LLMs and specialized models like AlphaFold act as economic complements. LLMs handle analysis, coding, and writing, while specialized tools handle domain-specific predictions. Most handoffs between them run through the scientist. What may the future of science look like? LLMs orchestrating those handoffs automatically--prompting specialized models, auditing outputs, and looping until either the question is answered or a non-automated stage is reached (e.g., wet lab testing). The scientist's role shifts from running each step to designing this loop. We saw a glimpse of this workflow with Anthropic’s enzyme discovery a few weeks ago, and we're seeing this in our own work too. This raises foundational epistemic questions: 1. What will scientific understanding look like when discoveries are made by black-box models whose output is increasingly difficult to interpret? 2. How do we extract underlying mechanisms, not just outputs, as science becomes more automated? What will theory look like? 3. If AI automates the writing and junior lab work, how do we train the next generation of scientists to push the frontier? 4. What will motivate scientists in this epistemic reality? There's an economics angle too. Hypothesis generation gets cheap, so bottlenecks move downstream to verification and to choosing which questions are worth asking. Realizing AI’s full potential will take investment in infrastructure and institutional reform--making this as much an organizational challenge as a technical one We don't have the answers yet, and I'd love to hear what others think. Paper here: https://bit.ly/ai-in-science3h
星星之火可以燎原@fireandstart这篇文章把 AI 对科学的影响从“更快完成既有步骤”推进到“改变发现问题与组织实验的方法”，我觉得关键判断是：生成和协调一旦变便宜，稀缺环节就会转向验证、选题，以及把结果转化为可解释的知识。让大模型调用 AlphaFold 一类专业工具、检查输出并继续迭代，确实可能把科学家从重复操作中解放出来；但自动化流程跑通，并不等于我们已经理解了发现为何成立。 因此，评价这类系统不能只看它是否给出正确答案，还要看证据链能否复现、失败时能否定位、模型之间的交接有没有保留假设与不确定性，以及湿实验或其他独立检验能否推翻它。对黑箱结果，机制解释不必成为每个探索阶段的前置门槛，但应明确区分“预测有效”“因果机制已知”和“在新条件下仍可靠”，不要把三者合并成一个成功率。否则，廉价生成反而会制造昂贵的验证债务。 研究训练也需要随工作流变化：初级研究者仍要学会提出好问题、设计对照、判断测量误差和识别伪相关，而不只是学会操作工具。自动化可以承担步骤，却不能替代共同体建立可信度的制度。真正值得投入的基础设施，除了模型和算力，也包括可复现的数据、共享评测、失败结果记录，以及让跨学科团队有时间核验的组织安排。1h
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI

    2 Sources

    Alex Imas@alexolegimasNew essay for the @GoogleDeepMind Institute with James Manyika on AI x Science, "Bending the Curve of Discovery". While accelerating existing scientific practices is certainly useful, the real promise of AI is its potential to act as an invention of a method of invention (IMI) a la Griliches--e.g., the microscope or statistical inference--which would unlock questions and whole modes of discovery that were previously beyond human reach. Today, LLMs and specialized models like AlphaFold act as economic complements. LLMs handle analysis, coding, and writing, while specialized tools handle domain-specific predictions. Most handoffs between them run through the scientist. What may the future of science look like? LLMs orchestrating those handoffs automatically--prompting specialized models, auditing outputs, and looping until either the question is answered or a non-automated stage is reached (e.g., wet lab testing). The scientist's role shifts from running each step to designing this loop. We saw a glimpse of this workflow with Anthropic’s enzyme discovery a few weeks ago, and we're seeing this in our own work too. This raises foundational epistemic questions: 1. What will scientific understanding look like when discoveries are made by black-box models whose output is increasingly difficult to interpret? 2. How do we extract underlying mechanisms, not just outputs, as science becomes more automated? What will theory look like? 3. If AI automates the writing and junior lab work, how do we train the next generation of scientists to push the frontier? 4. What will motivate scientists in this epistemic reality? There's an economics angle too. Hypothesis generation gets cheap, so bottlenecks move downstream to verification and to choosing which questions are worth asking. Realizing AI’s full potential will take investment in infrastructure and institutional reform--making this as much an organizational challenge as a technical one We don't have the answers yet, and I'd love to hear what others think. Paper here: https://bit.ly/ai-in-science3h
    星星之火可以燎原@fireandstart这篇文章把 AI 对科学的影响从“更快完成既有步骤”推进到“改变发现问题与组织实验的方法”，我觉得关键判断是：生成和协调一旦变便宜，稀缺环节就会转向验证、选题，以及把结果转化为可解释的知识。让大模型调用 AlphaFold 一类专业工具、检查输出并继续迭代，确实可能把科学家从重复操作中解放出来；但自动化流程跑通，并不等于我们已经理解了发现为何成立。 因此，评价这类系统不能只看它是否给出正确答案，还要看证据链能否复现、失败时能否定位、模型之间的交接有没有保留假设与不确定性，以及湿实验或其他独立检验能否推翻它。对黑箱结果，机制解释不必成为每个探索阶段的前置门槛，但应明确区分“预测有效”“因果机制已知”和“在新条件下仍可靠”，不要把三者合并成一个成功率。否则，廉价生成反而会制造昂贵的验证债务。 研究训练也需要随工作流变化：初级研究者仍要学会提出好问题、设计对照、判断测量误差和识别伪相关，而不只是学会操作工具。自动化可以承担步骤，却不能替代共同体建立可信度的制度。真正值得投入的基础设施，除了模型和算力，也包括可复现的数据、共享评测、失败结果记录，以及让跨学科团队有时间核验的组织安排。1h
    Today's Rank

    #11

    Today's Rank

    #11