Adam Bowen is the CEO & Cofounder of Tergle, a Y Combinator-backed company building AI agents for audit.
Back office work appears to be on the brink of extinction and AI agents are leading the charge. Industries like accounting, auditing, compliance, and payroll have been bogged down by repetitive, manual tasks for years. But with AI, the back office now stands at the cusp of automation.
Why AI Is Taking Over
An AI agent is a piece of software designed to automate work that would have normally been done by a subject-expert human. Nowhere are these AI agents more applicable than in the back office, a domain dominated by repetitive, rule-based tasks that follow clear procedures.
Consider a typical audit engagement at a Big Four accounting firm. Staff usually spend thousands of hours manually extracting data from client systems, reconciling transactions across multiple ledgers and preparing work papers. In my experience, a single audit might involve reviewing more than 50,000 transactions and teams can spend three months on a single client, with data entry and document processing consuming most of their billable hours.
Now, imagine this same engagement powered by AI agents. The data extraction process that once took weeks can happen in minutes as AI agents connect to client systems, identify relevant transactions and transform unstructured data into audit-ready formats. Verification agents can simultaneously analyze 100% of transactions continuously—not just samples—flagging anomalies and potential risks using pattern recognition far beyond human capacity.
What This Could Mean For Businesses And Employees
Role Of Humans
As these AI agents begin taking over knowledge work, the role of humans in the back office will shift from doers to supervisors. Human judgment will still matter critically, but it will be applied selectively in moments of ambiguity or high risk, rather than wasted on the monotonous tasks that currently consume most of the working day. We see this already underway in San Francisco: Waymo’s driverless cars, powered by machine learning and AI, now claim to outperform Uber drivers in major cities.
As I see it, the impact on employment will be substantial and swift. McKinsey & Co. estimates that by 2030, AI could automate up to 30% of work hours, with back-office functions among the most vulnerable. The economics are simply too compelling for businesses to ignore: A single AI system can process the workload of dozens of employees at a fraction of the usual labor cost.
Startup Advantage
I believe the companies that embrace AI agents now will gain the same kind of advantage as those that adopted the internet in the 1990s. In the early days of the web, businesses that dismissed it as a fad lost out to competitors who moved faster, operated leaner and used the internet to scale globally. We’re seeing the same pattern unfold today: Startups deploying AI agents are unlocking levels of efficiency that legacy systems aren’t able to match. Traditional incumbents struggle to keep pace, held back by the inertia of large, slow-moving teams. Nimble startups, however, can rapidly iterate and deploy rapidly, making best use of these agents.
Potential Cost Savings
In the near to medium term, many more service-heavy industries, like boutique law firms, private medical practices and wealth managers, may see profitability rise. These firms could be radically slimmed down, run by a small group of senior partners, with the back-office layers beneath them—assistants, junior staff, administrative coordinators—replaced by AI agents. Human touch will likely play a larger role with existing clients, particularly older ones who value face-to-face interaction, so senior professionals in these industries may remain insulated for another decade or two. But the actual bulk of the work will no longer require human hands. It’s the junior and mid-level staff, not the partners, who are the most at risk.
Regulation
Regulatory bodies will likely mandate human sign-off on some outputs—like a doctor authorizing a prescription written by AI. But even in these cases, the human role is increasingly ceremonial—more like a pilot monitoring an advanced autopilot system that can navigate, detect turbulence and adjust course far faster than human reflexes allow.
Final Thoughts
Organizations embracing this new reality will likely secure a long-lasting competitive advantage, while those who cling to legacy systems and bloated hierarchies could find themselves outpaced not just by cheaper labor but by software that doesn’t sleep, makes fewer mistakes and only gets better with time. As I see it, the question isn’t whether AI agents will replace traditional back-office functions—it’s how long companies can afford to pretend they won’t.
Forbes Business Council is the foremost growth and networking organization for business owners and leaders. Do I qualify?
Read the full article here
