Constructive AI Alignment | Caryn Thiên Ngân Tran

This research agenda argues for alignment methods that account for interaction, adaptation, and feedback between humans and AI systems rather than treating behavior as static or one-shot. It is a long-running project I’ve been developing with Max Kanwal since our UC Berkeley days.

AAAI Workshop Constructive Alignment

Constructive Alignment

Max Kanwal^* and Caryn Tran^*

In AAAI-26 Workshop on Machine Ethics: from formal methods to emergent machine ethics, Singapore, Jan 2026

Bib

@inproceedings{kanwal2026constructive,
  project = {constructive_alignment},
  publication_sort = {2026-02},
  title = {Constructive Alignment},
  author = {Kanwal, Max and Tran, Caryn},
  booktitle = {AAAI-26 Workshop on Machine Ethics: from formal methods to emergent machine ethics},
  location = {Singapore},
  month = jan,
  url = {https://ceur-ws.org/Vol-4189/paper1.pdf},
  year = {2026},
}

AAAI Workshop Bounded Morality

Bounded Morality

Max Kanwal^*, Caryn Tran^*, and Patrick Mineault

In AAAI-26 Workshop on Machine Ethics: from formal methods to emergent machine ethics, Singapore, Jan 2026

Bib

@inproceedings{kanwal2026bounded,
  project = {constructive_alignment},
  publication_sort = {2026-01},
  title = {Bounded Morality},
  author = {Kanwal, Max and Tran, Caryn and Mineault, Patrick},
  booktitle = {AAAI-26 Workshop on Machine Ethics: from formal methods to emergent machine ethics},
  location = {Singapore},
  month = jan,
  url = {https://ceur-ws.org/Vol-4189/paper2.pdf},
  year = {2026},
}