Synopsis: In this Techstrong.ai Leadership Insights interview, Mike Lynch, principal for artificial intelligence (AI) strategy and transformation for Auditoria.AI, explains how the role of finance teams is evolving as more rote tasks are automated.

Mike Lynch breaks down how artificial intelligence is reshaping work inside the office of the CFO — starting with the back office and radiating outward.

Rather than asking if AI will be adopted in finance, Lynch says the real questions now are how fast and where first. For most organizations, the journey begins in controllership functions such as accounts payable and accounts receivable, where highly repetitive, rules-based processes make strong candidates for automation. Lynch notes that some organizations are seeing up to an 85% reduction in manual work such as sorting shared inboxes, responding to supplier and customer inquiries, and hunting down invoices.

That shift is less about eliminating roles and more about redefining them. Lynch describes a “human in the loop” model where AI handles the rote work and finance professionals step into supervisory and strategic roles. AI becomes a kind of junior digital coworker, preparing transactions, coding invoices and pulling data together, while humans review, approve and focus on higher-value activities such as supplier relationships, cash optimization and discount capture.

Looking ahead, Lynch expects finance teams to benefit from specialized reasoning models and domain-trained tools rather than generic large language models. He points to emerging interest in causal AI — systems that don’t just describe what happened, but help explain why metrics moved — as a coming wave for CFOs and FP&A teams.

Ultimately, Lynch argues, finance is moving from transactional to transformational. As AI systems take on more of the heavy lifting, finance organizations will evolve into hybrid workforces made up of humans and digital agents, with teams spending less time “treading water” in spreadsheets and more time telling the business what the numbers actually mean.