NEW YORK — Years ago, T. Rowe Price dipped its toe into AI. Now it’s jumping all the way in.
The Baltimore-based investment firm recently partnered with Anthropic to offer Claude, an AI assistant, to its workforce, and it is expanding its artificial intelligence lab, too, beyond its center in New York. In September, the company converted a small wing of its new Harbor Point headquarters into an AI hub.
“AI is moving into the core of how we invest,” the firm’s CEO, Rob Sharps, told analysts on a recent earnings call.
At its labs, data scientists and engineers tweak and customize tools. The hope, put simply, is to offload rote tasks and streamline work for employees.
Claude can help associates cover the movement of significantly more stocks, for example, said Christopher LePre, head of AI engineering and application development. Or it could read the news for an associate, synthesize it and share how recent events might impact portfolios they’re managing.
“We’re trying to get all of our associates to have access, to help them do their daily tasks,” he said, “to help them remove some of the toil in their daily work, partly so they’re more engaged and happier but also so they can focus more on our clients.”
Read More
Positioned either as a portent of apocalypse or as a pragmatic embrace of the next wave of innovative technology, depending upon one’s perspective, the use of AI models like Claude has increased in recent years. Anthropic is expected to be valued at over $1 trillion in its upcoming initial public offering, which would make it one of the largest ever.
About half of Americans said this year they’d used a chatbot, and employers across industries are welcoming the technology, too. Last week, Under Armour founder and CEO Kevin Plank described his company as being in the midst of an “AI transformation.”
A survey this year from Acuity Analytics found that four-fifths of asset management companies are using AI.
Cheryl Pipia, T. Rowe’s head of global sales management, said the company wants employees to realize “it’s OK for you to build a skill that might automate part of what you’re doing, so you can do those more important things.”
The advent of publicly available generative AI in 2022 initially gave pause to some companies that feared employees might share private data with chatbots. Many firms eventually created protective layers around existing models.
T. Rowe Price built its own tool, dubbed “ChatTRP.”
Company officials have stressed that AI can give employees more time for the human-made decisions.
Plus, everyone is doing it.
“I think investing in AI is a competitive necessity,” Sharps said.
The firm is retiring ChatTRP, and although some employees have access to other models, such as OpenAI’s ChatGPT and Microsoft’s Copilot, LePre said the company believes Claude offers “the best of those tools.”
In 2017, T. Rowe launched a technology development center with an eye toward data science in New York — an intentional location to attract tech talent in the days before the AI boom.
That center later rebranded to “The Lab,” and about 68 employees work there, with another eight now located at the new Baltimore expansion. Two others within the department will soon operate out of London.
The two-floor New York office, near Gramercy Park in Manhattan, doesn’t look much different than any other modern workspace. There are computers, desks and conference rooms with floor-to-ceiling glass doors and walls. Collaboration is encouraged. It was within this space that ChatTRP was created, and it is where engineers and data scientists might wrap intelligence layers around an AI model to yield higher-quality results.
One of the worries of the AI age, in addition to environmental concerns, is job loss — that the technology will soon replace a chunk of the workforce, causing economic upheaval. That has mostly not played out thus far, research shows, though the concern looms for many.
And, although T. Rowe Price reduced its roughly 8,000 global employees by about 6% from last summer to this one, according to documents filed with the Securities and Exchange Commission, the company has said that was unrelated to its AI push.
“Our AI strategy is focused on enabling our workforce and expanding our capabilities,” company spokesperson Arminta Plater said in a statement.
Another potential issue with AI adoption is runaway costs, which have plagued some firms. AI companies generally charge on a per-use basis, and T. Rowe employees are given a certain amount of tokens to use.
“We are very shareholder-driven and very client-driven, so we will not let the expenses get out of control,” LePre said.
A few years ago, LePre was content to work the way he always had and reluctant to use new AI models. That didn’t last long.
“Now I’m a little bit in love with it,” he said.




Comments
Welcome to The Banner's subscriber-only commenting community. Please review our community guidelines.