Goodfin's AI Models Surpass CFA Level III Exam Standards
Goodfin, in collaboration with NYU Stern, has released a study showing that general-purpose AI models can now pass the CFA Level III exam, announced in a press release. This achievement highlights AI's potential to provide expert-level financial guidance.
The study, led by Goodfin's Co-Founder & CTO Shilpi Nayak and NYU Stern's Professor Srikanth Jagabathula, evaluated 23 leading AI models, including OpenAI's GPT-4 and Google's Gemini 2.5, against the CFA Level III exam. This exam is considered the most advanced credential in investment management, testing portfolio construction, ethics, and scenario analysis.
Key findings revealed that certain AI models, such as Claude Opus 4 and Gemini 2.5 Pro, passed the exam without domain-specific training. The study also noted that while AI models excelled in multiple-choice questions, only a few performed well on complex essay prompts. Additionally, the study highlighted the trade-offs between model accuracy and compute costs, with Gemini 2.5 Flash offering a strong performance-cost balance.
Goodfin is already leveraging these insights to enhance its platform, which uses advanced AI to provide high-growth private market investment opportunities. This research underscores the potential for AI to transform wealth management by delivering sophisticated financial expertise more broadly and affordably.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Finance AI Weekly or Daily AI Brief.
Also, consider following us on social media:
More from Finance
Nov 19 Arizona opens criminal investigation into OpenAI over what it calls rogue AI agents in banks Nov 19 Citi tells customers they are on their own if third-party AI agents misuse their accounts Nov 18 Klarna launches an AI concierge that books, shops and drafts emails Nov 18 Singapore sets up sandbox for AI agent payments under new national AI missions for finance Nov 18 Payhawk raises $280 million at a $2 billion valuation to build an AI-agent finance platformFinance AI Weekly
Weekly newsletter about AI in finance. Covers AI-driven trading, fintech innovations, and data analytics transforming markets
Whitepaper
Tensordyne Napier: What If One Rack Could Do the Work of Nine?
Tensordyne
This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.
Read moreYou may also like
Capgemini Says 59% Faced Higher AI Costs Than Expected
Free chatbots miss key points in financial advice for low-income families, benchmark finds
Goodfire trains AI models to avoid misleading shortcuts
Artificial Analysis Index v5: open-weight models fall 12 points behind
Hong Kong SFC finds 30% of surveyed firms use agentic AI
Daily AI Brief: the AI news that matters, in your inbox.