Morgan Stanley

Morgan Stanley Wealth Management sits on a proprietary knowledge base of over 100,000 research reports, investment strategy notes, market commentary, and internal analyses. Financial advisors could not efficiently search or synthesize this content during live client conversations — finding the right insight required keyword search across fragmented systems, deep institutional knowledge, and time advisors did not have. The intellectual capital of the firm was effectively locked away from the people who needed it most.

Morgan Stanley partnered with OpenAI to build the 'AI @ Morgan Stanley Assistant' — a GPT-4 powered retrieval system trained and evaluated exclusively on the firm's internal content. The assistant uses retrieval-augmented generation (RAG) over the vetted knowledge base so answers are grounded in Morgan Stanley's own research rather than the open internet. Rigorous evaluation, prompt engineering, and human expert review by the firm's research and knowledge management teams ensured accuracy and compliance before rollout. A second product, AI @ Morgan Stanley Debrief, was later added to summarize and take notes on client meetings.

Total timeline: 2022–2024: from OpenAI partnership and internal evaluation to 98% advisor-team adoption across the wealth management division

By late 2024, the assistant was adopted by 98% of advisor teams, with over 16,000 financial advisors gaining instant access to the collective intelligence of the firm. Advisors reported dramatically faster answers to complex client questions and more time spent on relationship-building rather than document search. Morgan Stanley became the flagship enterprise reference case for OpenAI, demonstrating that a large regulated financial institution could deploy generative AI at scale with governance controls intact.