- Real-world impact is more important than benchmark scores, says Meta AI VP
- Song emphasizes security, trust and real-world benefits
- Meta’s latest model prioritizes people, she says
Meta’s newly appointed head of AI research, Dawn Song, is betting on agent AI for the future of artificial intelligence, stressing that they should augment humans rather than replace humans entirely.
Song sees agents performing “economically valuable” tasks as repetitive and time-consuming work that ultimately frees people to do more creative work.
As companies struggle to quantify AI’s impacts and deliver a meaningful ROI, Song believes the focus should be on the real world rather than benchmark scores.
AI agents should augment – not replace – humans
“The goal is not to replace people,” Song said South China Morning Post. She confirmed that she would be joining Meta Superintelligence Labs in a LinkedIn post along with other team members from Virtue AI.
“[AI] must be safe, trustworthy and beneficial,” she added.
Song is also a professor of computer science at the University of California, Berkeley—a university that recently introduced the Agents’ Last Exam (ALE), a new type of benchmark that assesses whether AI agents can perform more than 1,500 economically valuable tasks across 55 different industries.
Meta itself launched its first new model, the Muse, in April, which it says is designed to “put people first.”
As MSL’s Vice President of AI Research, Song will focus on AI safety, security and research, and will likely continue to emphasize the role of humans in an AI-first era.
Model capabilities are no longer a disadvantage for AI developers, with human, socio-economic and geopolitical impacts now emerging as a major focus. Anthropic was recently forced by the White House to pull its latest Frontier models over jailbreaking issues.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews and opinions in your feeds.
