The Grand Challengers Podcast Episode #58

Tadpoles, tolerance and why AI alignment is really a human problem

Guest: Juan Pablo Carbajal

March 3rd, 2026

Overview

We train AI systems to optimize for metrics — but what if the real alignment problem isn’t in the machine? What if it starts with us?

Juan Pablo Carbajal is a physicist, interdisciplinary researcher, and passionate educator at the Eastern Switzerland University of Applied Sciences (OST). Originally from Argentina, his career has spanned physics, robotics, biomechanics, agronomy, machine learning, and water research — driven not by any single discipline but by a fundamental need to understand the dynamical systems underlying everything he encounters. He holds a PhD in Informatics from the University of Zurich, has worked at Eawag (Swiss Federal Institute of Aquatic Science and Technology) and the University of Ghent, and currently chairs Dwengo Helvetica, a nonprofit advancing STEM education.

In this wide-ranging conversation, Juan Pablo joins host Peter Marcus Bach to reframe the AI alignment debate as fundamentally a human one. He argues that the same narrow, single-metric optimization we apply to AI systems — rewarding output over understanding — is already distorting science, education, and society at large. From the “publish or perish” crisis in academia to how we evaluate students through one-dimensional scores, the pattern is the same: pressure a system to hit a number, and unintended behaviours will emerge. Like Makoto Kern’s discussion of the AI “competence mirage” in enterprise software (Ep #52), Juan Pablo warns that optimizing for the appearance of intelligence is not the same as building genuine understanding.

The conversation traces Juan Pablo’s remarkable journey from collecting tadpoles in an Argentine river — complete with a memorable food-chain disaster involving crabs and a hungry bird — through studying embodied AI at the University of Zurich’s AI Lab, where intelligence is treated as inseparable from its physical body. Along the way, he shares a profound lesson from plant biology: how nitrogen-fixing bacteria and their “cheater” free-riders coexist through tolerance rather than punishment — a metaphor with deep implications for how we design both human institutions and AI systems.

Juan Pablo explains why physics-informed AI may be the necessary antidote to purely data-driven approaches. Using the analogy of a flat earth — perfectly valid at local scales but catastrophically wrong at global ones — he illustrates how data alone cannot reveal fundamental principles. True understanding requires integrating prior knowledge, physical constraints, and the kind of interdisciplinary thinking that our increasingly siloed institutions struggle to support.

Topics discussed include the limits of data-driven machine learning, Gestalt psychology and machine vision, the etymology of “robot” (from the Czech for “slave”), complex systems and emergent behaviour, the role of culture as humanity’s alignment mechanism, and why understanding our history remains our best defence against repeating its mistakes. Also featured: an expert recipe for homemade yerba mate and a childhood ecological massacre.

Episode Teaser

Juan Pablo Carbajal — physicist, interdisciplinary scientist, and self-described “principled troublemaker” — makes a provocative case: the AI alignment problem isn’t new and it isn’t about machines. It’s the same problem we’ve been failing to solve among ourselves. From Argentine tadpoles to embodied robotics to the publish-or-perish crisis, this episode reframes everything you think you know about AI risk.

Biography

“…A civilization that has forgotten history, is doomed to repeat it…”

Dr. Juan Pablo Carbajal is a researcher at the Institute for Energy Technology at OST – Eastern Switzerland University of Applied Sciences in Rapperswil, Switzerland. Originally from Argentina, he studied physics at the Universidad Nacional de Córdoba before moving to Switzerland to pursue his PhD in Informatics at the University of Zurich, where he worked in Rolf Pfeifer’s AI Lab on embodied artificial intelligence and robotics. His career has since taken him across an extraordinary range of disciplines: from magnetic flux leakage for pipeline crack detection in the oil and gas industry, to agronomy and plant-bacteria mutualism, biomechanics, machine learning, and water research. He has held positions at Eawag (the Swiss Federal Institute of Aquatic Science and Technology) and the University of Ghent, and his research today focuses on dynamical systems analysis, physics-informed machine learning, and complex systems applied to water treatment and recovery.

Beyond his research, Juan Pablo is a passionate advocate for interdisciplinary education and chairs Dwengo Helvetica, the Swiss chapter of the international nonprofit dedicated to making STEM education accessible to young people worldwide. A trained physicist at heart, he describes his driving force not as any single discipline but as a deep need to understand the systems underlying the world around him, a curiosity that has led him to work with artists, biologists, roboticists, doctors, and agricultural scientists alike. His current interests centre on complex systems, emergent behaviour, and the intersection of human and artificial intelligence alignment, arguing that the challenges we face with AI are reflections of problems we have yet to solve among ourselves.

Dr. Juan Pablo Carbajal headshot – physicist, interdisciplinary researcher, and AI alignment expert at OST Eastern Switzerland University of Applied Sciences

Resources Related to the Episode

Key Research & Publications

  • Schneider, M.Y., Torfs, E. & Carbajal, J.P. – “The dissolved oxygen ramp is immoral: facing global water challenges with mathematical analysis” (2024, preprint) [Open Access Link]
  • Juan Pablo Carbajal on ESMH – “Neural computing, paradigm shifts, and Europe” (interview, Jan 2025) [Link]
  • Carbajal, J. and Bellos, V., 2018. An overview of the role of Machine Learning in hydraulic and hydrological modeling. [Link]
  • Marco, D.E., Carbajal, J.P., Cannas, S., Pérez-Arnedo, R., Hidalgo-Perea, Á., Olivares, J., Ruiz-Sainz, J.E. and Sanjuán, J., 2009. An experimental and modelling exploration of the host-sanction hypothesis in legume–rhizobia mutualism. Journal of Theoretical Biology, 259(3), pp.423-433. [Link]

Books & Influential Works

  • Steven Strogatz – Nonlinear Dynamics and Chaos: With Applications to Physics, Biology, Chemistry, and Engineering [View on Amazon]
  • Ken Robinson – The Element: How Finding Your Passion Changes Everything [View on Amazon]
  • Rutger Bregman – Humankind: A Hopeful History [View on Amazon]
  • Rutger Bregman – Moral Ambition: Stop Wasting Your Talent and Start Making a Difference (2025) [View on Amazon]
  • Mihaly Csikszentmihalyi – Flow: The Psychology of Optimal Experience [View on Amazon]

Concepts & References

  • The Sokal Affair – physicist Alan Sokal’s 1996 hoax exposing misuse of scientific terminology in postmodern academic journals [Link]
  • Gestalt Psychology – Wikipedia article referenced during episode for visual perception examples [Link]
  • Origin of the word “robot” – coined in Karel Čapek’s 1920 play R.U.R. (Rossum’s Universal Robots), from Czech “robota” (forced labour) [Link]
  • Embodied Cognition – the study of intelligence as inseparable from its physical body [Link]
  • Physics-informed machine learning – integrating physical laws and prior knowledge into ML models [Link]
  • Rhizobium bacteria – nitrogen-fixing soil bacteria with mutualistic plant relationships and “cheater” strains [Link]
  • Magnetic Flux Leakage – non-destructive testing technique for detecting cracks in metal pipes [Link]
  • Pipe Inspection Gauge (PIG) – in-line inspection tool used for pipeline integrity in the oil and gas industry [Link]

Cultural / Fun References

  • Yerba Mate – Argentine tradition, Juan Pablo’s homemade recipe (boiled method with overnight rest and lemon) [Link]
  • Daniel Córdoba – Juan Pablo’s mentor and physics teacher in Argentina, advocate for motivation-driven education
  • Argentinian mate culture – shared drinking ritual as social bonding — [Link]

Organizations & Institutions

  • OST – Eastern Switzerland University of Applied Sciences (Juan Pablo’s profile) [Link]
  • Dwengo – international nonprofit for STEM education (Juan Pablo chairs Dwengo Helvetica) [Link]
  • University of Zurich AI Lab (Rolf Pfeifer) – where Juan Pablo pursued embodied AI research [Link]

Episode Chapters

(Chapters are embedded in the episode for quick access, click this to expand and view all chapters and time stamps)
  • 0:00 Intro
  • 1:59 Guest Intro & Yerba Mate
  • 6:47 JP’s Secret Mate Recipe
  • 11:13 Summarizing JP’s passions into a keyword
  • 13:08 The Tadpole Story
  • 17:24 JP’s dilemma of what to study and his lifelong mentor
  • 22:57 Versatility of a physicist and the Sokal affair
  • 25:58 Critique of the current education systems
  • 31:23 From physics to magnetic flux leakage detection
  • 36:51 A life lesson on tolerance from bacteria
  • 46:12 Embodied AI, Robotics and Studying Intelligence
  • 54:07 Physics-informed AI
  • 59:21 The challenge with prior knowledge – example from Gestalt Psychology
  • 1:05:08 The Real Alignment Problem
  • 1:22:49 JP’s current exciting next steps
  • 1:26:12 Q&A Start
  • 1:26:43 What does innovation mean to you?
  • 1:28:38 Key moment, book, person
  • 1:31:10 Time Management
  • 1:33:32 Favourite childhood memory
  • 1:33:56 Biggest challenge to date
  • 1:35:58 Advice for young professionals
  • 1:36:35 What would you most like to be remembered for?
  • 1:37:28 Where can people find you?
  • 1:39:02 Final Message
  • 1:39:40 Outro

Connect with Juan Pablo Carbajal

Related Episodes

Episode #11 – Anas Ghadouani
A limnologist, filmmaker, foodie, water diplomat, and passionate teacher walk into a
kitchen…
(Related Topics: education, science)
Episode #25 – Shubber Ali
Your garden, the ‘field of dreams’ for tackling the twin crises of climate change and artificial intelligence
(Related Topics: AI, community, robots)
Episode #52 – Makoto Kern
Feature-centric to user-centric – UX Design to navigate global challenges and the “AI Competence Mirage”
(Related Topics: AI, human-computer interfaces)
Hiten Sonpal headshot – CEO of RISE Robotics and robotics industry leaderEpisode #57 – Hiten Sonpal
A LEGO crane, a belt and disrupting the 600 billion dollar robotics industry
(Related Topics: robotics, AI)

Credits