Fidelity Has Big Ambitions for Its Machine-Learning Compliance Offering

Fidelity Investments has unveiled a new technology that aims to automate much of the compliance work that goes into producing marketing materials, a major pain point for many advisory firms that operate in a highly regulated industry.

Fidelity this week announced Saifr, an artificial-intelligence and machine-learning tool that can comb through a company’s existing public-facing communications scanning for red flags. It also aids in drafting new materials.

Fidelity is considering taking Saifr beyond financial services to other highly regulated industries.

Dreamstime

“We are filling a gap in the regulatory landscape in the sense that this is a very manual process at the moment,” said Vall Herard, managing director of Saifr at Fidelity Labs, the company’s technology incubator.

Fidelity is launching Saifr while in negotiations with a handful of potential clients, although no deals have been finalized. Still, Fidelity’s ambitions for the product are broad, as Herard explained that it will be available as an annual subscription service to any financial services firm whose marketing content is regulated by the Securities and Exchange Commission, the Financial Industry Regulatory Authority, or their European counterparts.

“We consider any company that falls under those rules or similar types of rules in the case of Europe to be potential clients of Saifr, and obviously that number is in the thousands,” Herard told Barron’s Advisor.

At launch, the Saifr is focused on firms’ marketing materials, but Herard said that an expansion into preparing regulatory filings such as advisors’ Form ADV is “on the roadmap.”

Fidelity is also considering taking the product beyond the financial services industry. Pharmaceutical companies and medical-device makers, for instance, operate under similarly regulated advertising frameworks, answering to the Food and Drug Administration and Federal Trade Commission instead of the SEC and Finra. Food and beverage companies and telecommunications providers also have advertising compliance rules that Saifr could potentially help navigate.

“There are a number of industries where you have rules that govern the standards that companies have to follow when they’re communicating with the public,” Herard said.

But for now, Saifr resides in the financial services world.

Saifr hits the market with two products. Companies can use SaifrScan to analyze their existing marketing materials for potentially misleading statements or other content regulators might take issue with.

Then the primary platform, SaifrReview, helps firms as they’re drafting new materials, whether in written format, images, audio, or video. The product offers a risk score, helps identify where disclosures might be missing or incomplete, and offers suggestions before the compliance team finally signs off. It can also help with submitting those materials to regulators.

Saifr emerges from a massive data-aggregation and analysis effort, coupled with machine-learning and artificial-intelligence capabilities that enable the software to  gather insights and adapt its compliance review to match the nuances of a given firm.

Saifr used a set of proprietary natural-language-processing systems to comb through tens of millions data records, producing a system that aims to automate some of the tedious elements of compliance review while easing the friction that arises when the marketing, compliance, and legal departments all have a hand in developing and reviewing marketing materials.

“There’s a lot of back and forth that happens in that process,” Herard said. “People had a desire to use artificial intelligence in this area, but the lack of data is really a blocker, so we think that we’ve solved that.”

While Saifr aims to automate certain processes that could be described fairly as grunt work, it promises to free up compliance and marketing personnel to work on more complex problems—not to replace them altogether.

“I think the human touch is still required,” Herard said. “We’re not at a point yet where robots can take over.”