Asif Mohamed, head of digital product for global research, and Sam Azzarello, head of content strategy for global research at JP Morgan, said the bank uses artificial intelligence to provide faster, more targeted insights, enabling clients to enhance their alpha generation.

The research department has been using AI to find thematic information in the news and to better understand changes in sentiment from central bankers.

Economists use AI to analyze central bank speeches to determine whether they are hawkish or dovish. Analysts can use AI to identify thematic information in the news that impacts companies, sectors, and countries, and for sentiment analysis to help clients generate alpha. AI can also generate insights by analyzing sources such as earnings calls and reports.

Mohamed said: “We have accelerated using AI to generate faster insights. Looking for a needle in the haystack in a very structured way is the focus of some of the modern products in global research.”

Keyword search has historically been a challenge for the research industry because of contextual limitations, according to Mohamed. For example, a standard query on ‘DC’ produces mostly information about the city of Washington, DC and companies with DC in their names.

Artificial intelligence facilitates a more advanced semantic search, in which a DC query would yield more relevant and meaningful information about US government policy and White House news.

“AI can scan across decades of written research and data and synthesize that information relatively quickly,” added Mohamed. “We think this is a game changer, as we are able to generate faster insights for clients.”

Mohamed and Azzarello spoke about JP Morgan’s research suite, the evolving needs of client, and AI applications. A transcript of the conversation was provided to Markets Media.

Investment Mosaic

JP Morgan’s global research is one input to the mosaic that clients build for their investment thesis, so AI helps analysts with their own deep expertise and knowledge to adapt to news and changes in markets.

“We help clients discover and integrate data, information and insights that they may not necessarily know that they needed,” added Mohamed. “Our platform has tremendous scale and coverage, so we are able to generate applicable insights for most clients throughout the globe.”

Azzarello noted cross-asset research is becoming more relevant because many market events ripple around the globe and impact across equities, fixed income, FX and derivatives.

“The ability for us to pull together the breadth and depth of our expertise and showcase it in a cross-asset format is going to become more powerful,” she added. “Cross-asset is a fantastic way to light up key themes for clients and help them find the needle in the haystack which drives alpha.”

Technology enhances clients’ ability to find research that matches their interests — for example, AI can generate insights by making recommendations based on their reading. Technology is also critical for efficient distribution, as Azzarello described global research as a content powerhouse, with more than 1,000 JPM analysts producing 130,000 pieces of content per year, augmented with data sets and analysis.

Azzarello said the distribution of global research is omnichannel. The bank uses AI, the cloud and data to provide research via email; through JP Morgan Markets, the bank’s research portal; through a podcast on Apple or Spotify; on LinkedIn, or in a live call where clients can ask questions. She said: “At the core, it’s about meeting clients where they already are.”

For instance, when news broke in April that the US would be imposing tariffs on other countries, the bank hosted a cross-asset webinar for clients. The research department pulled together its best insights from around the world, recorded a podcast, and produced a thematic page on JP Morgan Markets.

Delivering at Scale

Mohamed said the research department sends about 1.5 million emails to clients daily, and 90,000 clients access the JP Morgan Markets portal every month, doing thousands of daily searches. In addition, the bank uses aggregators to distribute content to clients, has mobile platforms and provides electronic or machine-readable research to clients who want to integrate the data into their platforms.

“Delivering at scale in the most innovative ways is really important,” he said.

Electronic delivery of machine-readable research has become increasingly important as clients make investment decisions using more and more data.

Mohamed said: “Data is exploding. Clients would like to search JP Morgan’s research as well as their own proprietary data sets and content to generate insights.”

JP Morgan Markets’ DataQuery platform has 13 million time-series of data that investors use to make decisions. The bank has also developed JP MAQS, a macroeconomic quant data set that traders can use to back-test strategies.

“That’s a powerful combination,” said Azzarello. “Augmenting our analysts’ insights with new and proprietary data leads to faster and smarter insights.”

The focus on the research product will continue. Mohamed said: “We want to skate to where the puck is going to be, and that requires ongoing investments over a long period of time. We need to continue learning, being dynamic and delivering as the markets and our clients’ needs are changing.”