AI Models Ace CFA Level III Exam, Challenging Human Financial Expertise

Would you trust AI for financial advice? That may not be as far off as you think

Less than half of (human) candidates passed the exam in February. Some tasks that demand Herculean cognitive effort from humans are trivially easy for AI systems, which are designed to detect and replicate complex patterns gleaned from enormous troves of data. The technology has already scored top marks in world-class competitions in mathematics and coding; soon -- according to the vision of some developers -- help human researchers make new scientific discoveries. Also: OpenAI tested GPT-5, Claude, and Gemini on real-world tasks - the results were surprising Now, AI's capabilities are also rapidly catching up with the most skilled human financial analysts. A new study conducted by the New York University Stern School of Business and AI-powered wealth management platform GoodFin found that some frontier AI models were able to successfully pass a mock version of the Chartered Financial Analyst (CFA) Level III exam, widely regarded as the world's most difficult and prestigious testing benchmark for aspiring investment management professionals. The study analyzed the capabilities of 23 industry-leading models, both proprietary and open-source, from developers like Google, OpenAI, Anthropic, Meta, xAI, and DeepSeek. Also: AI helps strong dev teams and hurts weak ones, according to Google's 2025 DORA report Previous studies had demonstrated that AI could pass the CFA Levels I and II exams, but it struggled with the third and final (and most difficult) phase. The Level III exam, designed to test candidates' ability to apply their knowledge of portfolio management and wealth planning to hypothetical real-world scenarios, includes a set of multiple-choice as well as essay questions. "This dual format comprehensively tests higher-order cognitive skills including analysis, synthesis, and professional judgment over rote memorization," the researchers behind the new study wrote in a paper that was originally posted to the preprint server site arXiv in June. "The exam's rigorous standards...make it an excellent benchmark for assessing advanced financial reasoning capabilities." Less than half (49%) of human test-takers passed the Level III exam in February, according to the CFA Institute. A handful of reasoning models, which specialize in breaking problems down into a series of sub-problems that are then tackled successively, were able to pass the mock exam. Also: This app will pay you $30/day to record your phone calls for AI - but is it worth it? OpenAI's o4-mini scored the top spot with an overall score of 79.1%, followed by Google's Gemini 2.5 Flash, which scored 77.3%. The passing threshold for the exam is 63%. The researchers note in their paper that while most of the models included in the study scored in the same range on the multiple-choice segment of the exam (around 71%-75%), their scores varied much more widely on the more rigorous and challenging essay segment. "This suggests that simpler, more straightforward tasks have become commoditized across models, whereas complex and nuanced reasoning tasks still differentiate frontier and reasoning-enhanced models from their peers," they write. Also: How AI can help you manage your finances (and what to watch out for) A recent report from Microsoft listed personal financial advisors as one of the forty job categories most likely to be taken over by AI. Still, Anna Joo Fee, founder and CEO of GoodFin, told CNBC that she's not concerned about immediate replacement. "There are things like context and intent that are hard for the machine to assess right now," Fee said. "That's where a human shines, in understanding your body language and cues."

AI can now pass the hardest level of the CFA exam in a matter of minutes

The new study was developed by researchers from New York University Stern School of Business and GoodFin, an AI-powered wealth-management platform. For humans to pass the prestigious, three-part chartered financial analyst exam, it typically takes around 1,000 hours of studying over the course of several years. New research found that the technology underpinning a slew of artificial intelligence models is now advanced enough to pass even the most difficult - Level III - mock exams in a matter of minutes. The new study - developed by researchers from New York University Stern School of Business and GoodFin, an AI-powered wealth-management platform - evaluated 23 large language models on their ability to answer multiple choice and essay questions on mock CFA Level III exams. They found frontier reasoning models, including o4-mini, Gemini 2.5 Pro, and Claude Opus, were able to use "chain-of-thought prompting" to successfully pass. Previous research, particularly from two years ago, had found AI could clear Levels I and II of the exam, but it struggled with Level III, due to the essay questions. However, the technology has evolved so rapidly that the researchers wanted to know whether the models could handle, "specialized, high-stakes analytical reasoning required for professional financial decision-making." The third CFA exam is primarily focused on portfolio management and wealth planning. "I think there's absolutely a future where this technology transforms the industry," said Anna Joo Fee, founder and CEO of GoodFin, which contributed to - but did not pay for - the research. Still, Fee said she does not think the AI will ultimately replace the CFA. "There are things like context and intent that are hard for the machine to assess right now," Fee said. "That's where a human shines, in understanding your body language and cues."

Study shows AI can pass the hardest CFA exam -- but is it ready to do your taxes?

AI is everywhere - from viral trends and topping the leaderboards for reasoning tests to now passing one of the toughest finance exams in the world. A new study from NYU Stern and AI wealth-management startup GoodFin put 23 AI models through their paces on Level III of the Chartered Financial Analyst (CFA) exam. For those unfamiliar, this is widely regarded as the most difficult stage. Shockingly, several models passed with ease, finishing the test in just minutes, using advanced prompting techniques. The researchers tested a mix of new and older large language models (LLMs). Using chain-of-thought prompting, meaning the AI explains its reasoning every step of the way. Here's how some of the top models performed: Two years ago, AI could only reliably handle Level I of the CFA exam. This jump to Level III shows just how quickly AI reasoning has advanced. In just a year, this test proves how much AI is catching up with human analysts. Passing Level III, the most challenging level, suggests that AI can now handle not just rote memorization, but complex scenario analysis, something financial analysts rely on daily. It's important to keep in mind that exams do not equal real-world judgment, and while AI is capable, it shouldn't replicate the trust, ethics and practical decision-making of a human CFA. Too much reliance on AI could lead to more workslop. But the success we've seen on these tests does mean we could see more hybrid workflows with analysts increasingly leaning on AI to support data analysis, report drafting, and researching, while certification and human oversight remain critical. AI's ability to ace the CFA exam underscores how reasoning-focused models are evolving to help humans in even more ways. Beyond finance, we could see similar breakthroughs happening in medicine, law and engineering, where exams highlight expertise. For aspiring CFA charter holders, this may be a preview of the tools you'll be using in your career as AI assistants speed up the grind, allowing you to focus on higher-level strategy and client trust. The bottom line here is that AI passing the hardest CFA level in minutes shows the raw potential of advanced AI like o4-mini, Gemini 2.5 Pro, and Claude Opus. But for now, human expertise still sets the gold standard in finance.

It Takes Humans 1,000 Hours of Study to Pass This Exam for Finance Professionals. It Took AI Mere Minutes.

Two years ago, AI passed the first two multiple-choice parts of the Chartered Financial Analyst (CFA) exam, but failed the third section due to the essay questions. Now, some AI models are advanced enough to ace the test in mere minutes (and sans studying), according to a new report released earlier this week by New York University's Stern School of Business and AI wealth management platform GoodFin. The test, which finance professionals take to obtain a certification that can lead to a better salary, takes humans, on average, 1,000 hours of studying to prepare for. The study found that nine AI reasoning models were able to pass all three parts of the CFA exam in minutes -- even the most difficult third level of the test, which focuses on portfolio management and wealth planning, and asks for answers to essay questions. Related: Microsoft Claims Its AI Is Better Than Doctors at Diagnosing Patients, But 'You Definitely Still Need Your Physician' The researchers evaluated 23 AI models using CFA mock exams. OpenAI's o4-mini (one of the models that powers ChatGPT), Anthropic's Claude Opus, DeepSeek's R1 model, and Google's Gemini 2.5 Flash were part of the group that passed, using "chain-of-thought prompting," a tactic that asks AI models to show how they arrived at an answer with step-by-step reasoning. Meta's Llama 4 Maverick, xAI's Grok 3, and OpenAI's GPT-4.1 nano were part of the 14 models that failed the test. GoodFin founder and CEO Anna Joo Fee told CNBC that AI carries the potential to "transform" the investment management industry -- but she doesn't think that AI will make the CFA credential obsolete. "There are things like context and intent that are hard for the machine to assess right now," Fee told the outlet. "That's where a human shines, in understanding your body language and cues." Related: ChatGPT Can Now Complete a Major Task That Would Take a Human Up to 30 Days. Here's How it Works. The CFA test is administered every few months to thousands of candidates and costs up to $4,600 for all three levels of the exam. Professionals who pass can see a 53% salary increase, according to research conducted by exam preparation site, 300Hours. The company that administers the exam, the CFA Institute, says the test is "one of the highest distinctions in the investment management profession." Despite now passing all three levels of the test, ChatGPT will likely never become a CFA charterholder or a professional who has earned the CFA credential. The CFA Institute requires 4,000 hours of work experience, two references, and the completion of practical skills training to fully obtain the CFA credential.