Helping employees conduct easier Wall Street data searches was the initial purpose behind Morgan Stanley’s AI chatbot, AskResearch. D’Arcy Carr, the global head of research, editorial, and publishing, says that AskResearch’s range is expanding in 2020 to include external research clients.
And, with easier access to Wall Street information, courtesy of the AI chatbot, clients stand poised to make better and quicker decisions that could save them up to $8 billion annually.
Training the AI Chatbot
Before chatbots like AskResearch can effectively answer questions, they must know how to interpret requests. Morgan Stanley spent a year training the chatbot to respond as expected.
The process began with teaching AskResearch about Wall Street-related terminology. After all, most non-investors don’t know terms like “bull case,” but the chatbot must understand them to fulfill its purpose. After the basics, came in-depth training for more complex terms.
How Does AskResearch Work?
AskResearch provides information that’s stored inside analyst reports. Even unstructured data comes into play when a user asks the AI chatbot a question. For example, users can ask anything from “what is our U.S. GDP forecast” to “what is Morgan Stanley’s standpoint on gold?” AskResearch then delves through the necessary data before offering an appropriate response.
Carr indicated that one of the best aspects of this chatbot is its ability to pull details from hard to find places. She told Business Insider that this feature solves a long-standing problem,
“You have this historical archive sitting in a library and there is so much value embedded in it, but traditionally it has been hard to unlock that value because insights and data are fixed in monolithic PDFs.”
Chatbots have become all the rage on Wall Street lately. I spoke to @MorganStanley about the virtual assistant it created for its research and the plans it has to spin it out to external customers. https://t.co/0tAmbNiaVF via @BIPrime
— Dan DeFrancesco (@dandefrancesco) July 10, 2019
Since implementing AskResearch, Morgan Stanley has seen a significant increase in their internal readership. In other words, AskResearch is directing people to research they would have otherwise missed.
The Future of AskResearch
Aside from rolling out an app for external research clients, Morgan Stanley sees a big future for AskResearch. Carr believes the AI chatbot can learn many other skills. A mastery of which would assist analysts and free up enough time for users to get out in the field more often.
Now that AskResearch understands how to handle the more subtle nuances of research-related questions, Carr believes it’s ready for a much bigger user base. External research clients are the first target group. Meanwhile, Morgan Stanley’s eventual goal includes rolling the chatbot out to large investors as well.
AI Chatbots Continue to Pop-Up in the Financial World
The entire financial sector is banking on chatbots to provide results similar to Morgan Stanley’s. For instance, PayPal has begun rolling out chatbot’s to cover certain customer service requests. Now, if you reach out to PayPal to inquire about the source of a particular charge, the odds are high that a chatbot will field your call.
PayPal estimates that replacing humans for this one task alone will cut more than 6 million annual calls and $25 million in payroll expenses. Consumers also benefit because they can get the information they seek without spending a lot of time on hold.
Are Chatbots as Good as Advertised?
Some members of Wall Street question whether or not AI chatbots need more refining before they can become true time savers. Since the overall goal is quickly providing useful information, it may be necessary to curate data more carefully before it’s fed to a chatbot. Otherwise, some fear becoming inundated with even more data they don’t need.
Solving AI Chatbot Issues
Anyone who has used a chatbot knows they do have limitations. Therefore, Morgan Stanley and all other chatbot adopters will need to keep moving forward to achieve truly impressive results. It’s a virtual certainty that Morgan Stanley’s research technology experts are continuously seeking better results from AskResearch.
Perhaps one of the participants of Morgan Stanley’s Girls Who Code summer immersive program will end up solving some of the AI’s existing problems. This year marks the 7th annual effort to train and embolden the next generation of female coders.
We've taught 185,000 girls to code. Every minute that goes by, one of our students is learning to code.
— Girls Who Code (@GirlsWhoCode) July 10, 2019
No matter who perfects the technology, it’s clear that Wall Street has no plans of going back to an old school approach. AI chatbots are here to stay, and everyone from Wall Street to Main Street can expect to encounter them with increasing frequency.
Are you interested in finding out more about AI Chatbots? If so, be sure to check out how this cutting edge technology is helping make the travel industry more efficient here.