In just a few minutes, AI easily passed the CFA Level 3 exam
Researchers from NYU Stern and GoodFin tested 23 large language models and found that cutting-edge reasoning models like Gemini 2.5 Pro, Claude Opus, and o4-mini successfully passed the notoriously difficult CFA Level III mock exam using chain-of-thought prompting. While these models outperformed traditional ones on complex financial reasoning, experts note AI still falls short of human professionals in contextual understanding and client interaction.
Researchers from NYU Stern and GoodFin tested 23 large language models and found that cutting-edge reasoning models like Gemini 2.5 Pro, Claude Opus, and o4-mini successfully passed the notoriously difficult CFA Level III mock exam using chain-of-thought prompting. While these models outperformed traditional ones on complex financial reasoning, experts note AI still falls short of human professionals in contextual understanding and client interaction.
Read more
