Generative artificial intelligence company Boosted.ai announced this week the closing of a $15 million financing round. Part of the funding will go toward further developing its latest technology and expanding into the wealth management market.
While not a household name for most financial advisors, Boosted.ai launched in 2017 and is perhaps better known to asset and investment managers. It has offices in Toronto and New York, and the company states that it currently serves more than 300 active clients managing in excess of $3 trillion in assets across its institutional and wealth management segments.
“We started with institutions, hedge funds, sovereign funds,” said Joshua Pantony, co-founder and CEO of Boosted.ai, in an interview with WealthManagement.com earlier this year.
The latest financing includes funding from Fidelity Investments Canada ULC and Boosted.ai’s existing institutional shareholders, which include Ten Coves Capital, Spark Capital, Portage Ventures, Royal Bank of Canada and HarbourVest Partners.
According to a Boosted.ai spokesperson, this additional capital is an extension of the firm’s Series B round and brings its funding total to $61 million.
As a startup, Boosted.ai began building its own machine-learning algorithms for sophisticated users at hedge funds and institutions.
“Our users have used us for everything from idea generation to analyzing stocks, portfolios, and risk—really make the whole investment process more efficient and streamline and perform workflows on your behalf,” said Pantony.
The company is now developing its own agentic AI platform called Alfa, which is meant to be more autonomous than generative AI. It is meant to think for itself to a certain extent and act as an assistant to those employing it.
In other words, as Nvidia describes it, agentic AI “employs sophisticated reasoning and iterative planning to autonomously solve complex, multi-step problems.”
Specifically, Boosted.ai envisions Alfa as an AI coworker who can be trained to think like them and, in turn, monitor for things that might affect their portfolios or perform research and analysis.
“It will monitor everything happening in the world and notify me of anything that might affect my portfolio, and it can go to the next level and operate independently of the user,” said Pantony.
“We see the use case for the wealth side more about client communication; for example, we talk to a lot of RIAs that publish on a weekly basis about what went well in the portfolio and why and what went poorly—we want to automate that for the advisor,” he said.
The platform can generate the newsletter, the analysis and commentary within it or generate a podcast script.
“And let’s say the advisor wants commentary or a script using only hockey metaphors, no problem,” said Pantony.
He does not share the belief among some that human advisors are going to be replaced; too much is still stored in the human brain, especially when it comes to the experiential and emotional aspects of the job.
“A machine may know McDonald’s is going to roll out a new hamburger, but it won’t know how it tastes; someone will still need to taste it—huge amounts of text and analysis of it—humans won’t need to do that anymore,” he said.
Hallucinations, which are misleading or incorrect results generated by artificial intelligence, remain a major concern among all types of firms employing generative artificial intelligence.
“We cannot afford hallucinations, and our hallucination rate has to be very low,” Pantony said, adding that the company relies on the use of multiple large language models for the different types of processes the technology performs, including separate models for determining what needs to be fact-checked and for the actual fact-checking. The company even relies on a dedicated LLM committee of engineers and experts for oversight of those models.
“Behind the scenes, we have a lot of different models working together. Some are managed in-house, others outside,” he said.
While Boosted.ai does not publish its pricing information, Pantony said this is determined on a client-by-client basis. He did say he expects that for most wealth management firms his company’s technology will be considered a premium product, one that will most benefit large RIAs with large staffs, where improvements in efficiency offset the cost.