Morgan Stanley has 16,000 financial advisors serving clients across dozens of countries. Each of them needs to know what the firm knows — research reports, investment theses, product documentation, compliance guidelines, market analyses. The volume: over 350,000 documents. The problem: finding the right one, when needed, in the middle of a client conversation.
Before the AI deployment, a financial advisor searching for a specific piece of institutional knowledge might spend 30 minutes or more navigating the firm’s internal systems. By the time they found it, the conversation had moved on.
The original idea: make AI the librarian
In spring 2023, Morgan Stanley partnered with OpenAI as its exclusive wealth management strategic partner. The question the team asked was specific: what if every advisor had instant access to everything the firm knows, in the middle of a conversation, in seconds?
The answer was the AI @ Morgan Stanley Assistant — a GPT-4 powered chatbot embedded into advisor workflows, trained on 350,000+ internal documents: research notes, analyst reports, product fact sheets, compliance materials. The assistant went live in September 2023.
The idea was precise: AI as institutional memory. The assistant did not replace advisor judgment — it answered the question the advisor would have spent 30 minutes researching.
The results — from the official Morgan Stanley press release
By June 2024, when Morgan Stanley announced the AI @ Morgan Stanley Debrief — a second tool that auto-generates meeting notes and drafts follow-up emails — the company disclosed the adoption rate for the original assistant: 98% of Financial Advisor teams were actively using it [Morgan Stanley Press Release, 26 June 2024].
The Debrief tool extended the same approach into a different moment — the client meeting itself. With client consent, it records the session, generates structured notes, surfaces action items, drafts a follow-up email, and saves everything to Salesforce. The advisor reviews and edits — the AI handles the administrative layer.
- 98% of FA teams actively using AI @ Morgan Stanley Assistant [Morgan Stanley, June 2024]
- 350,000+ internal documents searchable in real time
- Answer time: from 30+ minutes to seconds
- AI @ Morgan Stanley Debrief: auto-generates notes, email drafts, Salesforce sync
- Partnership with OpenAI as exclusive wealth management strategic partner, announced March 2023 [CNBC, September 2023]
Why 98% adoption matters
Enterprise software adoption rates of 98% are exceptional. The industry standard for voluntary adoption of new internal tools rarely exceeds 40-60% within the first year. The Morgan Stanley result points to something the design team got right: the tool solved a problem advisors already felt, delivered value immediately, and required no behavior change in the way advisors worked with clients.
The assistant integrated into existing workflows. It did not require advisors to change how they structured conversations — it gave them an answer when they needed one. The Debrief tool operated the same way: it captured what already happened in a meeting, and generated the output the advisor would have spent 30 minutes producing afterward.
What you can take from this
The Morgan Stanley deployment is instructive for any organization with a large, distributed knowledge base — research, documentation, policy, guidelines — that employees struggle to access in real time.
The design question Morgan Stanley answered is worth asking in any organization: where do your people spend time searching for knowledge your organization already has? When the answer is “frequently, and during high-value moments,” the AI librarian model becomes the right architecture.
The 98% adoption rate was the market’s answer to how well the tool solved that problem.
SAGA — Success Stories & Real Cases · Curating real AI implementations: the original idea, the decision, the verified result.