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

    Ethan Mollick Recommends Pre-AI Pages on Complex Systems

    The Wharton professor discusses how systems survive flaws and the need for AI defenses.

    DP
    EM
    WI
    5 Sources, 28d ago, first seen 28d ago

    TLDR

    Ethan Mollick posted on X recommending three pages written before the AI era on complex systems. He explained that these systems run broken yet survive because their flaws rarely line up, giving humans a chance to intervene. Mollick stated that AI can find or make aligned flaws and called for a new philosophy of system defense that uses AI. The post includes attached images of the pages. Mollick is an associate professor at Wharton and author of Co-Intelligence.

    Combined views

    95.4K

    5 Sources, first seen 28d ago

    Combined views

    95.4K

    5 Sources, first seen 28d ago

    808 likes
    808 likes
    63 comments
    641 saves
    135 reposts
    63 comments
    641 saves
    135 reposts

    Sentiment

    Positive73.7%26.3%Negative

    Summary

    Many accounts welcomed integrating complex systems research and SFI scientists into AI for improved robustness, while others raised concerns about slow community engagement and risks of losing human oversight in automated systems.

    Based on 23 sentiment-bearing replies from 19 accounts across 2 conversations.

    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    Sentiment

    Positive73.7%26.3%Negative

    Summary

    Many accounts welcomed integrating complex systems research and SFI scientists into AI for improved robustness, while others raised concerns about slow community engagement and risks of losing human oversight in automated systems.

    Based on 23 sentiment-bearing replies from 19 accounts across 2 conversations.

    5 Sources

    @emollickThese 3 pages (from before AI) on complex systems are worth reading Complex systems run broken but survive because the flaws rarely line up, which allows humans a chance to intervene. AI can find or make aligned flaws. We need a similar new philosophy of system defense using AI
    @alexolegimasWe need more complex systems, SFI-pilled scientists involved directly in AI research. I spent 7 years at CMU in the same department as some of the best complex systems researchers around (aside: CMU's Social and Decision Sciences department was heaven on earth). It is a huge shame that scientists who have spent decades studying multi-agent systems and the emergent properties from those interactions are not more central to the conversation.
    @pfauRT @alexolegimas: We need more complex systems, SFI-pilled scientists involved directly in AI research. I spent 7 years at CMU in the sam…
    @wsisaacI cannot stress how much I agree with @alexolegimas on this point. I'm proud that DeepMind has partnered with SFI for years on bridging the worlds of complexity science and AI research. https://www.santafe.edu/news-center/news/measuring-ai-in-the-world

    5 Sources

    @emollickThese 3 pages (from before AI) on complex systems are worth reading Complex systems run broken but survive because the flaws rarely line up, which allows humans a chance to intervene. AI can find or make aligned flaws. We need a similar new philosophy of system defense using AI
    @alexolegimasWe need more complex systems, SFI-pilled scientists involved directly in AI research. I spent 7 years at CMU in the same department as some of the best complex systems researchers around (aside: CMU's Social and Decision Sciences department was heaven on earth). It is a huge shame that scientists who have spent decades studying multi-agent systems and the emergent properties from those interactions are not more central to the conversation.
    @pfauRT @alexolegimas: We need more complex systems, SFI-pilled scientists involved directly in AI research. I spent 7 years at CMU in the sam…
    @wsisaacI cannot stress how much I agree with @alexolegimas on this point. I'm proud that DeepMind has partnered with SFI for years on bridging the worlds of complexity science and AI research. https://www.santafe.edu/news-center/news/measuring-ai-in-the-world