The Coming AI Shift in Accounting

Artificial intelligence is already reshaping how adults work in nearly every field.

Not long ago, computer science professionals enjoyed high demand, strong salaries, and job security. Today, however, many experienced programmers are struggling to find new positions or are holding tightly to the ones they have. AI didn’t eliminate programming, but it dramatically reduced how many people are needed to do the same work.

Now, that same shift is coming to accounting and tax preparation.

AI excels at structured tasks. It can follow rules, analyze data, and prepare documents at incredible speed. Tools like Claude or ChatGPT allow one professional to accomplish what once required several staff members. From a business perspective, that’s a huge gain in productivity.

Tax preparation seems like a perfect match for AI. Tax laws are complex but clearly written. The forms are standardized, and the math is formula-driven. On paper, this looks like work that a machine could handle easily.

But here’s the problem: risk.

Tax filing leaves almost no room for error. A single oversight can lead to penalties, audits, or long-term IRS issues. And when a return is wrong, the taxpayer—not the software—is responsible.

AI also has a well-documented weakness: it can sound confident while being completely wrong. In other professions, people have already relied on AI-generated information that turned out to be false or misleading. In tax work, such mistakes can be costly.

That’s why most taxpayers don’t yet trust AI to manage their returns entirely on its own. But that trust may soon change.

In a recent Wall Street Journal, mathematician and RAND researcher Mr. Price laid out a realistic timeline for this transition. By 2027, he expects professionals to use large language models to draft tax filings as a first pass, followed by human review and final submission. By 2028, he predicts that enough people will trust general-purpose AI to prepare their individual taxes that it will start to disrupt major companies like H&R Block and Intuit.

This doesn’t mean accountants will disappear—it means the work will evolve.

Just as in software development, the first roles affected will be routine, entry-level tasks. Simple tax returns, data entry, and basic reviews are exactly the areas where AI performs best.

The CPA of the future will spend less time filling out forms and more time exercising judgment, offering strategy, and taking responsibility.

AI can draft a tax return.
AI can organize information.
AI can even suggest answers.

But AI cannot explain strategy to a client.
It cannot represent someone in an audit.
And it cannot accept legal responsibility.

The lesson from computer science is clear: high pay and strong demand today don’t guarantee safety tomorrow. Jobs built primarily on routine work are the most vulnerable to automation.

AI won’t replace every professional. But it will reduce how many are needed.

Computer science felt it first.
Accounting is next.