RI: Small: Characterizing the Meaning of Human Preferences for AI Alignment — NSF Award to University of Massachusetts Amherst (MA
Artificial Intelligence (AI) systems are becoming increasingly capable, yet deploying them to produce reliable, intended outcomes remains difficult. Large language models often fail to follow instructions, AI systems that govern important functions sometimes behave unpredictably, and autonomous systems, such as self-dr
| Award title | RI: Small: Characterizing the Meaning of Human Preferences for AI Alignment |
|---|---|
| Award ID | 2437426 |
| Awardee | University of Massachusetts Amherst |
| City | AMHERST |
| State | MA |
| Amount obligated | $599,969 |
| Principal investigator | Scott Niekum |
| Program | Robust Intelligence |
| Start date | 09/01/2025 |
| Abstract | Artificial Intelligence (AI) systems are becoming increasingly capable, yet deploying them to produce reliable, intended outcomes remains difficult. Large language models often fail to follow instructions, AI systems that govern important functions sometimes behave unpredictably, and autonomous systems, such as self-driving vehicles or robots, can act in ways that diverge from user expectations. To improve reliability and utility, future AI systems will need to demonstrate that their goals and b |
| Source | NSF Awards |
$799/mo
Try NSFGrants →