H&R Block's AI Tax Assist gave wrong or unhelpful answers to 30%+ of tax questions tested
Washington Post columnist Geoffrey A. Fowler tested H&R Block's AI Tax Assist with tax professionals and found it gave wrong or unhelpful answers to more than 30 percent of questions. Specific errors included advising a single parent to file as Single instead of Head of Household and incorrectly stating the IRS had not addressed cryptocurrency wash sale rules. H&R Block defended the tool by saying the test questions lacked specificity and the bot was curated for common tax scenarios from the prior year.
A tax chatbot confidently told a single parent to file as Single instead of Head of Household and insisted the IRS had not addressed crypto wash sale rules when it had.
Key facts
- What
- Washington Post columnist Geoffrey A.
- Incident date
- Mar 1, 2024
- Who
- H&R Block
- Failure mode
- Hallucination
- AI surface
- Chatbot
- Severity
- High
What happened
Washington Post technology columnist Geoffrey A. Fowler tested H&R Block's AI Tax Assist chatbot with assistance from tax professionals at EP Wealth Advisors. The chatbot gave wrong or unhelpful answers to more than 30 percent of questions, including advising a single parent to file as Single instead of Head of Household and incorrectly stating the IRS had not addressed cryptocurrency wash sale rules. Other errors included misleading advice about multi-state filing requirements for college students. H&R Block spokesperson Teri Daley defended the tool by saying the test questions lacked specificity and that the bot was curated for the most common tax scenarios from the prior year.
What broke inside the model
- 01 · TriggerA user asks for a fact, a citation, or a figure.
- 02 · Model stepThe model writes a fluent, confident answer.
- 03 · Control gapNothing ties the claim back to a real source.
- 04 · FailureA fabricated fact ships as if it were verified.
- 05 · ConsequenceThe false claim reaches a customer, a court, or the public.
Confidence holds, and even spikes, as the claim detaches from any source.
The generative AI chatbot produced confident but factually incorrect answers about tax filing statuses and IRS rules, failing to correctly apply Head of Household eligibility criteria and misrepresenting the IRS position on cryptocurrency wash sale rules. The system lacked sufficient guardrails to prevent authoritative-sounding hallucinations on tax law questions where precision is legally required. H&R Block acknowledged the tool was not perfect and relied on disclaimers urging users to confirm answers rather than implementing rigorous factual checks before output.
What it cost
Sources
- PressTurboTax and H&R Block now use AI for tax advice. It's awful.washingtonpost.com
- PressTech Columnist: TurboTax and H&R Block Chatbots Are Unhelpful or Wrong Much of the Timenysscpa.org
Cite this entry
https://failureindex.ai/failures/h-r-block-ai-tax-assistAI Failure Index. "H&R Block's AI Tax Assist gave wrong or unhelpful answers to 30%+ of tax questions tested" (FI-0085). Realm Labs. https://failureindex.ai/failures/h-r-block-ai-tax-assist (indexed Jun 4, 2026).Data fields CC-BY 4.0, prose citation permitted. Incident ID FI-0085. Full dataset at /data.
Note from Realm Labs, the Index steward
How Realm would have caught this
- Prism
- OmniGuard
- AI Detection & Response (AIDR)
A runtime layer that watches the model's internal state can flag the moment a model commits to a claim it has no support for, and hold or reroute the response before it reaches a user. Realm reads those signals in real time rather than grading the transcript after the fact.