Serendipity with Generative AI: Repurposing knowledge components during polycrisis with a Viable Systems Model approach
This paper demonstrates how generative AI can function as a serendipity engine to extract and organize reusable knowledge components from existing literature into a Viable Systems Model-aligned repository, thereby offering a strategic framework for organizations to navigate polycrisis uncertainty through systematic knowledge repurposing rather than relying solely on breakthrough innovation.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
The Big Picture: The "Polycrisis" and the Lost Library
Imagine the world is currently facing a Polycrisis. Think of this not as a single storm, but as a perfect storm where a hurricane, a wildfire, and a flood are hitting at the exact same time. Climate change, pandemics, and economic shifts are all happening together.
In this chaos, organizations (companies, universities, governments) are trying to innovate to survive. But they are making a huge mistake: they are looking outside for new answers while ignoring the goldmine of knowledge they already have sitting in their own basements.
The Problem: Organizations have thousands of documents, research papers, and reports (their "knowledge"). But this knowledge is locked away. It's like having a library where every book is written in a different language, buried under piles of dust, and no one knows what's inside them. When a new crisis hits, they try to invent a solution from scratch, wasting time and money.
The Solution: The authors propose using Generative AI (like the smart chatbots we use today) not to write poetry or draw pictures, but to act as a super-librarian and a knowledge translator.
The Core Idea: Repurposing as "Upcycling"
Think of knowledge like furniture.
- The Old Way (Breakthroughs): When you need a chair, you try to build a brand new one from raw wood, even though you have a perfectly good, slightly damaged chair in the attic.
- The New Way (Repurposing): You take that old chair, sand it down, paint it, and turn it into a stylish side table. You didn't invent a new object; you repurposed an existing one.
The paper argues that organizations are full of "furniture" (models, frameworks, checklists, patterns) that were built for one specific problem but could be easily fixed up and used for a totally different problem.
The Engine: How the AI Works
The researchers built a tool using Generative AI to scan 206 academic papers. Here is how they describe the process using a metaphor:
The "Needle in a Haystack" Metaphor:
Imagine a massive haystack (the 206 papers). Hidden inside are 711 tiny, golden needles (reusable ideas like a "checklist for safety" or a "model for growth").
- Humans: If you asked a human to find all those needles, it would take them years. They might miss some because they are tired or biased.
- The AI: The AI is a super-powered magnet. It scans the haystack in hours, finds every single needle, and sorts them into neat boxes.
The Result: They found an average of 3.5 reusable ideas per paper. That means a single department has hundreds of "building blocks" waiting to be used.
The Theory: The Viable Systems Model (VSM)
To explain why this works, the authors use a theory called the Viable Systems Model (VSM). Let's break this down with a Human Body analogy:
- System 1 (The Hands/Feet): These are the workers doing the actual jobs. They create the papers and the data. They are busy and focused on their specific tasks.
- System 4 (The Brain/Strategist): This is the part of the organization that looks at the horizon, sees the storms coming, and plans how to adapt.
- The Problem (The Transduction Gap): Usually, the "Hands" (System 1) are too busy to talk to the "Brain" (System 4). The Brain doesn't know what the Hands have discovered. The information gets lost in translation. This is called a Transduction Cost.
- The AI as the "Nervous System": The Generative AI acts as the nervous system connecting the Hands to the Brain. It takes the raw, messy work of the "Hands," translates it into a clear, usable format, and instantly sends it to the "Brain."
The Magic: By lowering the cost of this translation, the AI allows the organization to react instantly to crises. Instead of waiting for a new invention, the "Brain" can say, "Hey, the 'Hands' already built a model for this last year! Let's use that!"
The "Serendipity" Engine
The paper introduces a cool concept: Planned Serendipity.
- Serendipity is usually a happy accident. You are looking for a sock, but you find a $20 bill.
- Planned Serendipity is setting up a machine that guarantees you find the $20 bill.
By using the AI to scan everything, organizations stop waiting for luck. They create a system where "happy accidents" (finding a solution in a totally different field) happen all the time.
- Example: A paper about "customer co-creation" in marketing might accidentally contain a perfect pattern for "community engagement" in a city's urban planning department. A human wouldn't see the link, but the AI spots the pattern immediately.
Real-World Impact: The "5-Day Sprint"
The paper gives a story (a vignette) about a mid-sized manufacturer whose supplier suddenly collapsed.
- Without AI: Panic. They spend weeks trying to figure out a new supply chain.
- With AI: The "Brain" (System 4) instantly queries the repository. The AI pulls out:
- A Checklist for risk assessment.
- A Framework for digital procurement.
- A Pattern for temporary staffing.
- A Heuristic (a quick rule of thumb) for calculating safety buffers.
In 5 days, they have a new supply chain plan. They didn't invent anything new; they just recombined existing knowledge blocks.
The Sustainability Angle: "Knowledge Upcycling"
Finally, the paper connects this to the environment.
- We know that recycling plastic bottles saves energy and reduces waste.
- Knowledge Upcycling is the same thing. If we reuse a framework or a model, we don't have to spend energy (research hours, money, carbon footprint) reinventing it.
By reusing knowledge, organizations become more sustainable. They stop "throwing away" smart ideas after they are used once.
Summary
The Paper in One Sentence:
In a world of constant chaos, organizations shouldn't keep trying to invent new solutions from scratch; instead, they should use AI as a super-librarian to find, translate, and reuse the thousands of smart ideas they already have hidden in their own documents.
The Takeaway:
Don't wait for a breakthrough. Start repurposing. Use AI to turn your "old furniture" into new, life-saving solutions.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.