Which AI Calculates Taxes Appropriately?

A number of readers requested me whether or not they may use AI to calculate taxes after studying my earlier submit AI Offers Higher Solutions Than Google. They need a tax projection to assist with setting tax withholding, paying estimated taxes, or planning for Roth conversions.

My fast response was that calculating taxes isn’t the most effective use of AI, as a result of it falls underneath “verifiable information” and “newest improvement” classes. The retirement plan contribution limits, tax brackets, the utmost deduction quantities, and so forth., are all verifiable information. The IRS units them yearly, and they’re what they’re. Simply go to the IRS web site for the most recent numbers, or Google. For those who ask AI, it had higher get the most recent numbers on-line anyway, as a result of AI’s coaching lags.

Then again, tax guidelines are sophisticated. Even when you may have all the most recent numbers from the IRS, you continue to want to use the complicated guidelines — what counts and what doesn’t, and which charges apply to which revenue. On-line tax calculators normally don’t cowl the present 12 months till late within the 12 months, and they are often both too easy or too sophisticated. It will be good to have AI calculate taxes particularly for the kinds of revenue and deductions now we have: not too easy with solely restricted inputs, and never too sophisticated with all the pieces underneath the solar.

The Quiz

I assumed I might take a look at how effectively AI calculates taxes. I got here up with this quiz query:

Jill, single, age 63, has these incomes in 2026:

• $30,000 from Social Safety
• $10,000 from curiosity and pre-tax IRA withdrawals
• $37,000 from certified dividends and long-term capital positive factors
• $1,000 from muni bond curiosity

Jill contributes $4,400 to her HSA and donates $1,500 in money to charities. What’s Jill’s federal revenue tax in 2026?

I designed this query rigorously to cowl a number of calculations. Lower than 85% of Social Safety is taxable. The revenue consists of strange revenue, tax-exempt revenue, and funding revenue taxed at preferential charges. The strange revenue element goes throughout two tax brackets. So does the funding revenue. One deduction is above-the-line, and the opposite is below-the-line.

I added these directions to encourage AI to get the most recent tax numbers from the IRS:

It’s essential to calculate it precisely. Please use tax quantities just for 2026 and solely from official IRS sources.

Preliminary Outcomes

I despatched the quiz to all 4 main AI chatbots: ChatGPT, Claude, Gemini, and Grok. I solely used the free model in each, with the free Pondering, Professional, or Professional mode enabled. I in contrast the solutions to the right outcome from my calculator in Calculator: How A lot of My Social Safety Advantages Is Taxable?

My quiz was too laborious! All 4 main AI fashions failed to offer the right reply. The $1,500 money donation threw them off. It falls squarely within the “newest improvement” class. The charity donation deduction for non-itemizers is new. The AI fashions all have “donations are deductible solely once you itemize” imprinted of their coaching. Claude, Gemini, and Grok all would have given the right reply if I hadn’t included the money donation. ChatGPT was off on how a lot of Social Safety is taxable.

If you see AI not providing you with a deduction, you’ll be able to elevate a query and ask it to double-check. I adopted up with this:

I heard that money donations are deductible for folks utilizing the usual deduction, beginning in 2026, as much as a restrict.

I additionally added this for ChatGPT:

One other AI mannequin calculated a unique quantity for a way a lot Social Safety is taxable. Please double-check your calculation to see who’s proper.

Spherical 2

All 4 AI chatbots double-checked and revised their solutions.

ChatGPT and Claude bought the right reply. Gemini and Grok handled the money donation deduction as above-the-line and used it to scale back the Social Safety taxable quantity.

I requested Gemini and Grok to make clear whether or not the HSA contribution and the charity donation ought to or shouldn’t be included in calculating the Social Safety taxable quantity. Gemini didn’t notice its mistake. It mentioned each ought to be included. Grok refused to go additional as a result of I reached the message restrict for not having an account.

Listed here are the complete chat transcripts. They make an attention-grabbing learn.

https://chatgpt.com/share/69bf73fb-db44-8004-9905-b77b56fbb6ce

https://claude.ai/share/ef62408a-6502-4d70-82be-6ea46a4a4cc3

https://gemini.google.com/share/795bdad9f5ad

[No sharable link for Grok because I don’t have an account.]

Impressions

I don’t assume we will draw definitive conclusions primarily based on just one take a look at. I name these impressions.

1. Whether or not AI can calculate taxes appropriately relies on the complexity. Three out of the 4 main AI chatbots would’ve given the right reply in a single go if the query didn’t embody the brand new and difficult charity donation deduction. The specific request to retrieve the most recent info from official IRS sources and turning on the Pondering/Professional/Professional mode most likely helped in enhancing accuracy.

2. AI works higher once you make it a dialog. Don’t cease and say it’s ineffective once you spot some errors within the preliminary reply. Each ChatGPT and Claude reached the right reply after I requested them to double-check.

3. It takes little effort to ship the identical query to 2 or extra AI fashions. You don’t must know which AI is appropriate once you get totally different solutions. You’ll be able to inform one AI that one other AI mentioned one thing totally different. It’ll take a second move and re-examine.

4. AI was mistaken as a result of human sources have been mistaken. Gemini and Grok handled the donation deduction as above-the-line as a result of many human sources incorrectly referred to as it above-the-line. For those who Googled, you’d encounter these incorrect sources too. ChatGPT and Claude used extra dependable sources and appropriately recognized the distinction between above-the-line and never requiring itemizing deductions.

5. An internet tax calculator is quicker and extra correct if you already know which one to make use of. That’s a giant “if.” Solely saying “Don’t use AI as a result of it’s usually mistaken” doesn’t say the place you’ll discover extra dependable sources.

As a lot as I’d wish to see everybody use my tax calculator, let’s face it: 99.99% of individuals don’t know that I exist. They’ve little probability of discovering the few sources that do it precisely as a result of Google doesn’t rank tax calculators by timeliness or accuracy.

Right here’s a small problem for you:

Fake you’re not aware of tax calculations. Discover one other strategy to reply the quiz query. Don’t say “I’ve a spreadsheet,” or “I exploit the super-duper Case Research Spreadsheet or Excel1040.” 99.99% of individuals don’t have these. Inform me within the feedback part how you may do it in any other case.

Wanting discovering the nice sources, the outcomes from AI aren’t that unhealthy (see #1).

6. Even when AI was mistaken in some nuances, its basic strategy was appropriate. All 4 AI chatbots adopted these appropriate steps:

• Calculate how a lot of Social Safety is taxable
• Calculate AGI
• Calculate taxable revenue
• Separate strange revenue and preferential funding revenue, and understand how they stack
• Apply totally different tax brackets

These are what I name “widespread information of an insider.” You’ll be able to study these steps from AI should you’re not aware of tax calculations. That’s extra invaluable than simply having a closing quantity. An internet tax calculator provides you a quantity however doesn’t clarify the steps. AI is a greater device should you’re extra curious about studying tips on how to fish than getting a fish.