Former Intel CEO’s New AI Benchmark Focuses on Human Flourishing
SMRTR summary
Pat Gelsinger, former Intel CEO, introduced the Flourishing AI Benchmark through Gloo, evaluating large language models on their alignment with human well-being across seven dimensions. Based on the Global Flourishing Study, the benchmark uses over 1,200 questions to assess models' promotion of human flourishing. OpenAI's o3 currently leads with 72 points, but no model has reached the 90-point threshold for robust alignment. Gelsinger aims to improve AI models' performance in areas like faith and meaning, hoping to spark discussions about values and ethics in AI.
SMRTR provides this summary for quick context. The original article belongs to Daily.dev.
Read the original article