Developmental AI is a bootstrapping approach where embodied AIs develop abilities along a bio-inspired trajectory. They start with innate competences and learn more by interacting with the world including people.
Aspirationally, developmental AIs would learn, share what they learn, and collaborate to achieve high standards. They would learn to communicate, establish common ground, read critically, consider the provenance of information, test hypotheses, and collaborate. The approach would make the training of AIs more democratic.
This project has produced three working papers.
The “Why” Paper (“What AIs are not Learning”) — 15 pages
- Why do robotic service applications require so much knowledge?
- Why is it better to create them using deep learning and foundation models?
- Why is a developmental approach right for creating experiential (“robotic”) foundation models?
The “How” Paper (“Bootstrapping Developmental AIs”) — 106 pages
- How do children learn so much so quickly?
- How do people (and animals) acquire competences?
- How does multi-model information fusion work?
- How do early-acquired competences prepare the way for later ones?
- How does a trajectory for acquiring competences work (for humans, animals, machines)?
The Collaborative AI Paper (“Roots and Requirements for Collaborative AIs”) — 24 pages
- Who needs collaborative AI?
- What competences are needed for (human-compatible) collaboration?
- What is the relationship and history of AI (artificial intelligence) and IA (intelligence augmentation)?
Publications
Stefik, M. (2024) What AIs are not Learning (and Why). (2024) arXiv https://arxiv.org/abs/2404.04267
Stefik, M., Price, R. (2023) Bootstrapping Developmental AIs. arXiv https://arxiv.org/abs/2308.04586
Stefik, M. (2023) Roots and Requirements for Collaborative AI. arXiv https://arxiv.org/abs/2303.12040