AI Just Scored 100 Percent on Accounting Tasks That Trained CPAs Only Got 37 Percent Right

-

TL;DR: A new study from Mercor found that Claude Opus 5 scored a perfect 100% on four month-end close accounting tasks that licensed CPAs averaged just 37% on. With AI compute costs falling roughly 50% per quarter, this result is a preview of where professional work is heading fast.

I've been watching AI benchmarks closely for a while. Most of them feel like progress in a lab. This one felt different.

Mercor, an AI hiring platform, ran a head-to-head comparison between Claude Opus 5 and licensed CPAs on real month-end close accounting tasks. Four tasks. Real accounting work. The kind that takes trained professionals hours to complete.

Claude Opus 5: 100%. Licensed CPAs: 37% average.

That's not a slight edge. That's not AI catching up. That's AI running laps around credentialed professionals on their home turf.

What the Study Actually Tested

Month-end close tasks are the real work of accounting. Reconciling accounts, verifying transactions, producing financial statements. This isn't simplified test content designed to flatter an AI. It's the structured, high-stakes workflow that companies depend on every month.

The CPAs who took these tasks are licensed. They passed one of the hardest professional exams in finance. And they averaged 37% accuracy.

One study, one company, one set of tasks. I'm not saying this settles every debate about AI in professional work. But I am saying this result deserves to be taken seriously, not dismissed as a fluke.

The Cost Side Changes Everything

Here's what makes this more than an interesting benchmark: AI compute costs are falling at roughly 50% per quarter. What costs $10 to run today costs around $5 in three months, $2.50 in six, and so on.

That trajectory fundamentally changes the economics of AI in professional work. Even if you had concerns about accuracy today, the accuracy is clearly improving and the cost is collapsing at the same time.

We're not talking about replacing an accountant's salary with a tool that costs the same. We're talking about a capability curve that's moving faster than most industries are ready for.

What This Means for Finance Professionals

I want to be precise here because the "AI replaces jobs" framing usually misses what's actually happening.

Month-end close is structured, repeatable, rule-based work. That's exactly the kind of task AI handles well. The judgment calls, client relationships, complex tax strategy, fraud detection with real context — those are different. A CPA does a lot more than close the books.

But the structured parts of the job are being automated faster than the industry expected. And those structured parts are often a significant portion of junior accountants' work.

The professionals who adapt well will be the ones who treat AI as a collaborator: let it handle the volume, focus on review, exception handling, and client-facing judgment. That combination is much more powerful than either alone.

The Pattern Is Bigger Than Accounting

This fits a broader trend across professional domains. A year ago it was coding. Now it's accounting. Medical image reading, legal research, and financial analysis are in the same trajectory.

AI isn't good at everything. Novel situations, emotional intelligence, long-range strategic thinking — still genuinely hard. But in structured domains with clear right and wrong answers? The progress has been faster and more durable than most experts predicted.

A 100% score on accounting tasks isn't magic. It's the result of models getting very good at pattern matching, rule application, and systematic verification — exactly what month-end close requires.

The Mercor study is one data point. But it's a loud one. And the direction it points is unmistakable.

What's your experience? Drop a comment below! 👇

Are you in finance or accounting? Has AI changed how you approach your work yet, or does this feel more like a distant headline?

#AI #Accounting #CPA #ClaudeAI #Anthropic #FutureOfWork #TechNews #Automation #FinTech #WorkplaceAI

Comments

Post a Comment

Popular posts from this blog

Plank You App Support and Privacy

-
Plank You is a simple, ad-free interval timer for planks and core workouts. Set your work and rest times, hold your plank, and let the countdown beeps do the counting. Everything runs on your device. no ads, no accounts, no tracking. How to use Start / Pause / Resume: tap the main button to run or pause the timer. Reset returns to Rep 1 and the start of your work interval. Rep counter — counts up automatically after each full work + rest cycle. Beeps — the last 5 seconds of every phase beep down (5-4-3-2-1), with a longer tone when the phase changes. Changing intervals Tap the settings icon (top right) to choose an interval: 45 / 15 default (45s work, 15s rest) 30 / 30 50 / 10 Custom — set any work and rest time in 5-second steps You can also toggle Beep sounds and Haptics in Settings. FAQ I don't hear any sound. Sound plays even when your ringer is on silent, but make sure the media volume is up and that "Beep sounds" is turned on in Settings. If another app is ...
Read more

WWDC 2026 Preview: Apple Needs to Fix Siri or It's Game Over for Apple Intelligence

-
Apple's WWDC 2026 kicks off on June 8, and the stakes couldn't be higher. While Google announced 900 million Gemini users at I/O last week, Apple is heading into its biggest developer conference with one uncomfortable question hanging over the whole event: what happened to Siri? I've been using Apple devices daily since the original iPhone, and I'll say it plainly — Apple Intelligence has been the most underwhelming software launch in Apple's recent history. WWDC 2026 is Apple's chance to fix that. Here's what they need to announce, what we're expecting, and what a failure would mean. The Problem Apple Needs to Solve When Apple announced Apple Intelligence at WWDC 2024, the demos were impressive. When the features started rolling out through 2025, the reality was a different story. The "new Siri" was slower, less reliable, and often less capable than the old one for basic tasks. The "Personal Context" features that were supposed ...
Read more

This AI Startup Is Worth $26 Billion and Writes 90% of Its Own Code — Should Software Engineers Be Worried?

-
I've been watching the AI coding space very closely for the past year, and every few months something happens that forces me to recalibrate my sense of how fast things are moving. This week, that moment came from Cognition AI — the startup behind Devin, the AI software engineer — which just announced a $1 billion funding round at a $26 billion valuation . But the number that actually stopped me cold wasn't the valuation. It was this: 90% of Cognition's own code is now written by Devin. From $10B to $26B in Eight Months — The Numbers Are Staggering Let's put this in perspective. Just eight months ago, in September 2025, Cognition closed a $400 million round at a $10.2 billion valuation. Today, that valuation has more than doubled to $26 billion. The new round was co-led by Lux Capital, General Catalyst, and 8VC, with Ribbit Capital, Atreides Management, and Peter Thiel's Founders Fund also participating. But what's really driving this isn't hype — it...
Read more