Contemplative Superalignment

  • Ruben E. Laukkonen
  • , Fionn Inglis
  • , Shamil Chandaria
  • , Lars Sandved-Smith
  • , Edmundo Lopez-Sola
  • , Jakob Hohwy
  • , Jonathan Gold
  • , Adam Elwood

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations

Abstract

As artificial intelligence (AI) improves, current alignment strategies may falter in the face of unpredictable self-improvement and the sheer complexity of AI. Rather than trying to control behavior, we show how four principles from contemplative traditions can help intrinsically align (super) intelligence. First, mindfulness enables self-monitoring and recalibration of emergent subgoals. Second, emptiness forestalls dogmatic goal fixation and relaxes rigid priors. Third, non-duality dissolves adversarial self–other boundaries. Fourth, boundless care motivates the universal reduction of suffering. We find that prompting AI to reflect on these principles improves performance on the AILuminate Benchmark (d = .96) and boosts cooperation and joint-reward on the Iterated Prisoner’s Dilemma task (d = 7 +). We also show how active inference offers parameters for integrating contemplative wisdom deeper into the architecture and world models of AI. This interdisciplinary approach offers a resilient alternative to brittle control schemes and may be the first empirical test of ‘ancient wisdom’.

Original languageEnglish (US)
Title of host publicationArtificial General Intelligence - 18th International Conference, AGI 2025, Proceedings
EditorsMatthew Iklé, Anton Kolonin, Michael Bennett
PublisherSpringer Science and Business Media Deutschland GmbH
Pages346-361
Number of pages16
ISBN (Print)9783032006851
DOIs
StatePublished - 2026
Event18th International Conference on Artificial General Intelligence, AGI 2025 - Reykjavic, Iceland
Duration: Aug 10 2025Aug 13 2025

Publication series

NameLecture Notes in Computer Science
Volume16057 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference18th International Conference on Artificial General Intelligence, AGI 2025
Country/TerritoryIceland
CityReykjavic
Period8/10/258/13/25

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science

Keywords

  • Alignment
  • Artificial Intelligence
  • Buddhism
  • Compassion
  • Contemplative Science
  • Large Language Models
  • Machine Learning
  • Meditation
  • Mindfulness
  • Neural Networks
  • Neurophenomenology
  • Non-duality

Fingerprint

Dive into the research topics of 'Contemplative Superalignment'. Together they form a unique fingerprint.

Cite this