OpenAI Trains AI to Build Financial Models with Ex-Bankers in Project Mercury
OpenAI has enlisted more than 100 former investment bankers to help train its artificial intelligence to build financial models, according to Bloomberg. The work is part of a confidential effort known internally as Project Mercury, aimed at automating the routine modeling tasks typically handled by junior bankers.
Participants in the program, including former employees of JPMorgan Chase, Morgan Stanley, and Goldman Sachs, are paid $150 per hour to write prompts and build models for various transaction types such as restructurings and initial public offerings. They are required to submit one model per week, following standard industry formatting conventions in Microsoft Excel.
The contractors receive early access to the AI system and provide feedback to improve its output. The application process for the project is conducted largely through automation, beginning with an interview via an AI chatbot followed by tests on financial knowledge and modeling skills.
A spokesperson for OpenAI said the company collaborates with domain experts to enhance model capabilities across different fields. The initiative reflects OpenAI’s push to make its technology more practical for business use, with Project Mercury focusing specifically on financial-sector applications.
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
42AI exits stealth with $62.5 million in seed funding
London's Model ML raises $250M at a $1.5B valuation for AI agents for bankers and consultants
Goldman Sachs Launches Small Business AI Institute
OpenAI Wins Anthropic Customers With Steep Discounts
Goldman Sachs Restricts AI Coding Tools for Traders
Daily AI Brief: the AI news that matters, in your inbox.