default prescriptions

A couple years after our first conversation on artificial intelligence and ecological restoration, I sat back down with Timothy Pape and Sam Woodrich to ask: what’s actually happening now? Their new research looks at how mainstream AI chatbots generate restoration plans across North American ecosystems. The results are familiar; almost too familiar. Plant native species. Remove invasive plants. Repeat. A longer show title might be - Default Prescriptions: AI, Ecology, and the Stories We Repeat. 

In this follow-up conversation, I reconnect with Timothy Pape and Sam Woodrich to explore what’s changed—and what hasn’t. Their recent study examines how AI chatbots generate restoration prescriptions across different ecosystems, and what emerges is a kind of pattern recognition loop: vegetation-first, context-light, and strikingly similar across places that should demand very different approaches.

Pape, T., & Woodrich, S. T. (2026).  Examining preference in chatbot answers related to ecological restoration prescriptions.  Restoration Ecology. https://doi.org/10.1111/rec.14264 

We talk about why that happens, what it says about the knowledge systems AI is trained on, and how these tools may be reinforcing the dominant narratives already present in restoration ecology. Along the way, we get into:

• why AI defaults to “plant natives and remove invasives”

• the absence of social, cultural, and economic context in restoration plans

• the limits of chatbots when it comes to asking deeper questions

• how practitioners are actually using AI in the field (for better and worse)

• the risk of “convincingly shallow” answers

• and the paradox of using resource-intensive technology to plan ecological repair

This episode sits at the intersection of ecology, technology, and values—and asks what happens when we let machines reflect our field back to us.

Guests

Timothy Pape
Assistant Professor, Bowling Green State University
Focus: ecological restoration, environmental studies, systems thinking

Sam Woodrich
PhD Candidate, Oregon State University
Focus: predator ecology, riparian systems, restoration science

Woodrich, S. T., & Pape, T. (2024). Ecological restoration and artificial intelligence: Whose values inform a project?  Restoration Ecology, 32(4), e14128. https://doi.org/10.1111/rec.14128

Work With Me

Interested in restoration strategy, climate adaptation, or ecological storytelling?
Reach out through Madrone Grove Adaptation & Restoration - treehuggerpod@gmail.com

Read more reflections: Grove & Grit (Substack)

Music from this episode is from YouTube Audio Library: True Coockoo, Xander Jones, The Grey Room