Research on AI systems that learn and internalize human values through interaction, observation, or inference.
Organizations
3
Key Papers
2
Grants
7
Total Funding
$5.6M
First Proposed: 2016 (Hadfield-Menell et al.)
Cluster: Scalable Oversight
Parent Area: Scalable Oversight
Grants7
| Name | Recipient | Amount | Funder | Date |
|---|---|---|---|---|
| Machine Intelligence Research Institute — General Support (2017) | Machine Intelligence Research Institute (MIRI) | $3.8M | Coefficient Giving | 2017-10 |
| UC Berkeley — AI Safety Research (2019) | University of California, Berkeley | $1.1M | Coefficient Giving | 2019-12 |
| Grant to "support research on value alignment in AI systems, practical algorithms for efficient value alignment verification, and user studies and experiments to test these algorithms." | University of Utah | $280K | FTX Future Fund | 2022-05 |
| Funding towards a 2 year postdoctoral stint to work on Safety in AI, with a focus on developing value aligned systems | Kush Bhatia | $275K | Long-Term Future Fund (LTFF) | 2022-01 |
| Create a value learning benchmark with contextualized scenarios by leveraging a recent breakthrough in natural language processing | - | $55K | Long-Term Future Fund (LTFF) | 2020-01 |
| 12-month salary for researching value learning | Charlie Steiner | $50K | Long-Term Future Fund (LTFF) | 2022-01 |
| Berkeley Existential Risk Initiative — Algorithmic Alignment Group | Berkeley Existential Risk Initiative | $30K | Coefficient Giving | 2024-09 |
Funding by Funder
| Funder | Grants | Total Amount |
|---|---|---|
| Coefficient Giving | 3 | $4.9M |
| Long-Term Future Fund (LTFF) | 3 | $380K |
| FTX Future Fund | 1 | $280K |
Tags
value-learningalignmentinverse-rl