# AI incidents at Bank of America

Source: https://www.bankingnewsai.com/banks/bank-of-america/ai-incidents
Last updated: Oct 21, 2026

## Has Bank of America had a publicly reported AI incident?

Yes: 2 publicly reported AI incidents on record since January 2022.

| Date | Institution | Incident | AI? | Role | Category | Severity | Status |
| --- | --- | --- | --- | --- | --- | --- | --- |
| 2023 | Bank of America | [AI clone of a client's voice calls a Bank of America banker to move money; attempt fails](https://www.bankingnewsai.com/ai-incidents/bank-of-america-2023-voice-clone-transfer-attempt) | ML (inferred) | AI used against the institution | Voice clone | 1/5 | Reported |
| Jul 14, 2022 | Bank of America | [Regulators fine Bank of America $225m over automated fraud filter that froze unemployment-benefit cards](https://www.bankingnewsai.com/ai-incidents/bank-of-america-2022-unemployment-card-fraud-filter) | Automation, not AI | Institution's own AI | Fraud model | 4/5 | Settled |

### AI clone of a client's voice calls a Bank of America banker to move money; attempt fails

The New York Times reported in August 2023 that in the spring of that year a Florida investor called his Bank of America representative about a large planned transfer, and that a second call, apparently in his voice, was actually generated by software that tried to get the banker to move the money elsewhere. According to the Times, the voice was repetitive and garbled, the banker hung up, and after two more such calls she reported the matter to Bank of America's security team. No loss was reported. Bank of America said it regularly trains staff to recognise scams but would not comment on specific customers, so the bank has not confirmed the incident; the Times called it a voice-deepfake scam attempt.

https://www.bankingnewsai.com/ai-incidents/bank-of-america-2023-voice-clone-transfer-attempt

### Regulators fine Bank of America $225m over automated fraud filter that froze unemployment-benefit cards

On 14 July 2022 the CFPB and the OCC fined Bank of America a combined $225 million ($100m and $125m) over its unemployment-benefit prepaid card program. Per the CFPB, from 28 September 2020 to 8 June 2021 the bank replaced reasonable investigations with an automated "Fraud Filter" using "a simple set of flags": it denied about 188,000 notices of error without further investigation, froze or blocked cardholders' accounts, and retroactively reversed more than 11,000 credits it had already paid. Both orders require redress, which the CFPB estimated at hundreds of millions of dollars. The bank neither admitted nor denied the findings and said states approved the payments amid unprecedented fraud. Regulators did not describe the filter as AI. The OCC terminated its consent order in June 2026.

https://www.bankingnewsai.com/ai-incidents/bank-of-america-2022-unemployment-card-fraud-filter

---

Canonical page: https://www.bankingnewsai.com/banks/bank-of-america/ai-incidents
Part of [BankingNewsAI](https://www.bankingnewsai.com/) — the register of AI incidents at banks: https://www.bankingnewsai.com/ai-incidents. Markdown versions of every reference page: append `.md` to the page URL; index at https://www.bankingnewsai.com/llms.txt.