Artificial intelligence agents have become table stakes in enterprise software, but in financial reporting, audit and compliance, speed counts for little if the output cannot be defended. That gap between capable and correct is what sets governance-first platforms, such as Workiva Inc., apart from the wider agent rush, because someone eventually signs their name to the work.

The company used its Amplify user conference to put controls, data lineage and human sign-off at the center of its product roadmap. Trusted data and traceable output are the preconditions for putting agents anywhere near a regulatory filing, according to Deepak Bharadwaj (pictured), executive vice president and chief product officer of Workiva.

“It’s really about the risk profile of our buyers and users,” Bharadwaj said. “And because of who we serve in financial reporting and GRC and sustainability, what they want is the work that they do can be trusted because ultimately in many of these cases somebody is signing their name to it.”

Bharadwaj spoke with theCUBE’s Krista Case, principal analyst and practice lead for cyber resilience and security, and co-host Alison Kosik at Amplify, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. (* Disclosure below.)

Workiva layers guardrails, context and human judgment

The approach stacks guardrails, plain-language instructions and human verification on top of data the platform has already certified, building on the specialized agents and intelligence layer the company launched in July. Where an agent can run unsupervised comes down to whether the task is repeatable and rules-based, Bharadwaj explained.

“But if you have tasks that require business judgment, or someone needs to look at it and approve and sign off, that’s where humans have to come in and make sure that what the AI is producing is correct,” he said. “And, oftentimes, that’s where multiple people will argue about one word in that disclosure.”

Agent Studio, announced at the event theCUBE covered this week, extends that model to customers, letting business users describe a process instead of writing Python scripts against application programming interfaces. Everything agents do inside Workiva is bounded by a defined set of capabilities and tied back to a person, Bharadwaj noted.

“AI is never doing things on its own,” he said. “It’s doing things on behalf of whoever is supervising it. And, ultimately, someone is going to be accountable. I like to say, ‘You cannot send an agent to prison.'”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of Amplify:

(* Disclosure: TheCUBE is a paid media partner for the Amplify event. Neither Workiva, the sponsor of theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE