Avalara CEO Hugo Sarrazin says speed without accountability creates new risk. A new survey backs that up with real numbers.
Australian finance leaders are moving fast on AI agents. Their governance frameworks aren’t keeping pace.
That’s the core finding of new research from Avalara, a company specialising in tax and compliance software, which surveyed CFOs and senior finance leaders across Australia who have deployed, piloted or evaluated AI agents in financial processes over the past year. The report, titled “Agents of Change: How the Race to Deploy AI Agents is Outrunning Financial Governance,” paints a picture of a function under pressure to prove value quickly, sometimes faster than its internal controls can support.
The pressure to show results
According to the survey, 88 percent of respondents feel moderate or significant career pressure to demonstrate that AI agent investments are delivering return on investment, with half describing that pressure as significant. Ninety percent say their AI agent initiatives are delivering at least some measurable ROI so far. Fifty-nine percent say the pressure they’re under is focused primarily on deployment speed rather than other priorities.
Hugo Sarrazin, Chief Executive Officer at Avalara, said the drive to adopt AI quickly is understandable, but comes with risk if governance is left behind. “Australian finance leaders are right to move quickly to capitalise on agentic AI opportunities, but speed without accountability creates new forms of risk, and speed without rethinking workflows limits ROI,” Sarrazin said. “The organisations that realise the greatest value from AI won’t simply deploy more agents. They’ll leverage agents with trusted data, governed workflows, and clear controls that enable automation with confidence.”
Where governance is falling behind
The numbers on governance readiness are where the gap becomes clearest. Just 12 percent of respondents say their organisation prioritises governance over deployment speed. When evaluating AI vendors, only 17 percent say governance and control documentation is the top proof they look for, and 18 percent point to auditability and explainability evidence, well behind the 23 percent who prioritise ROI evidence instead.
Perhaps most telling, 59 percent say they’re only somewhat confident they could explain an AI agent’s actions to an auditor or regulator if asked.
Frank Cirone, VP Commercial Strategy at cloud data platform Snowflake, said the skills required to govern AI agents properly go beyond what most finance teams currently have in-house. “Finance leaders are being asked to move quickly with AI, but governing agents requires a new combination of domain, AI, IT, and data governance expertise,” Cirone said. “As AI agents gain access to financial and compliance workflows, organisations need to know what those agents can see, what they can do, and when human approval is required. That kind of control has to be built into the architecture, not added after the fact.”
The survey found 75 percent of respondents lack dedicated in-house expertise to understand how their AI agents actually work, relying instead on IT teams or external vendors to fill that gap.
Who’s accountable when something goes wrong
The research also probed a question many businesses haven’t fully worked out: who’s responsible when an AI agent makes a significant error. Eighteen percent of respondents said accountability would be unclear or would sit with no one. A separate 18 percent believe the executive who approved the AI investment would ultimately be held personally accountable.
Jim Lundy, Founder, CEO and Lead Analyst at Aragon Research, said this uncertainty reflects a broader shift happening as AI moves deeper into core business processes. “As enterprises scale agentic AI, the question becomes less about whether the technology can act and more about whether organisations can understand, control, and explain those actions,” Lundy said. “In finance, where workflows are auditable and outcomes carry real business consequences, governance and explainability will become essential requirements for adoption.”
What would build more confidence
The survey suggests finance leaders aren’t opposed to using AI agents, they simply want more structure around how it’s done. When asked what would most increase their confidence in expanding AI agent use, respondents pointed to AI agents operating within existing systems of record (32 percent), audit trails documenting every AI action (31 percent), and outputs grounded in verified tax, compliance and financial data (30 percent).
The single capability rated most valuable for financial operations was audit-ready documentation for every AI-driven action, selected by 37 percent of respondents.
Keep up to date with our stories on LinkedIn, Twitter, Facebook and Instagram.
