Grant to "support a project to develop interactive AI algorithms for alignment that can uncover the causal features in human reward systems, and thereby help AI systems learn underlying human values that generalize to new situations."
$800K
Source
Notes
[AI safety] Grant to "support a project to develop interactive AI algorithms for alignment that can uncover the causal features in human reward systems, and thereby help AI systems learn underlying human values that generalize to new situations."
Other Grants by FTX Future Fund
74Showing 10 of 74 grants
Other Grants to University Of California, Berkeley
43Showing 10 of 43 grants