AI Doesn't Have to be Perfect to Accelerate Productivity
Trent Green
Corporate Tax Specialist | Technical Author | CPE Instructor | Founder of Tax Director Services
News articles pointing out AI's flaws are coming out almost daily (see recent examples below). Unfortunately, such media coverage can create noise that makes it easy to miss the positive impact that AI, even imperfect AI, can have on our work as tax professionals right now.
Since football is still fresh in the minds of many, allow me to use a related analogy to drive home the point. A good running back doesn't need everyone on defense to be blocked perfectly to make good yardage. If they're running hard when they hit the hole, all they need is a small crease and about a half second to make a move to break off a big run (click here for an entertaining example).
Similarly, a skilled tax professional doesn't need AI to answer every question perfectly or completely to accelerate productivity. As CPAs, we've long been trained not to rely solely on Internet searches for answers. Instead, all we need in many cases is for AI responses to be directionally correct, to trigger ideas or additional questions, or to focus our thoughts to quickly put us on the path to successfully getting answers and resolving issues.
In summary, does AI still need refinement? Yes. Will it continue to improve? Again, yes. And do we need to wait for it to be perfect to meaningfully impact our productivity? Absolutely not! We will see tangible benefits as we increasingly integrate AI into our day-to-day work, even as the technology continues to evolve.
Recommended by LinkedIn
Recent Articles on the Imperfections of AI