Cases › leaked private data › LR-2026-0605
Community bank discloses shadow AI data exposure
Preliminary Why: Publisher not yet named · subject to change
Developing: a major or severe case captured in the last 72 hours. Later changes are logged on the corrections page.
Status: fix reported
No source states: loss or records, developer and model, agent tool, could do, deployer.
Severity major: Test: a regulator involved. The filing states the company has been in communication with relevant banking and financial regulators.
An employee at CB Financial Services' Community Bank in Pennsylvania used an unapproved AI application while mocking up a presentation and entered customer information. The data included customer names, Social Security numbers and dates of birth. The upload was made from a personal device through a fully authenticated channel, and the bank says no threat actor was involved. The bank said it found out on May 5, decided the matter was material, and filed a Form 8-K on May 11. It brought in outside cybersecurity advisors and communicated with regulators, and it says the file and surrounding prompts were deleted after confirmation of upload. Afterward the bank blocked the web domains of unapproved AI tools, tightened employee access to customer data and expanded its data-security policies.
What happened
What was done
“Since the incident, the bank has blocked the web domains of unapproved AI tools and is working with the AI company on the cleanup, according to Montgomery.”
Risks
Read from the sources; not yet reviewed by a person.
Controls
HarnessNot effective (bypassed) Tool allow-list
Control failure chain
L1 · ModelProvider not namedModel provider
No source mentions a control at this layer.
Model version not linked to a model card.
L2 · HarnessHarness not namedAgent framework vendor
Other controls for this risk at this layer, not mentioned in the sources: Network egress control.
L3 · ApplicationDeployer not namedDeployer
No source mentions a control at this layer.
Other controls for this risk at this layer, not mentioned in the sources: Sensitive-data checks on outbound data, Output filtering and redaction, Human approval for high-impact actions, Data minimisation, Least-privilege access for the agent, Activity logs.
L4 · Infrastructure / targetOrganisation hit
No source mentions a control at this layer.
EffectiveNot effectiveNot effective (bypassed)Did not existNot stated in sourcesNot yet checkedAttacker's ownOrder: model · harness · application · target
Controls for this risk: result in this incident
| Control | Part of the system | Result |
|---|---|---|
| Tool allow-list | L2 Harness | Not effective (bypassed) |
| Network egress control | L2 Harness | Not stated in sources |
| Least-privilege access for the agent | L3 Application | Not stated in sources |
| Human approval for high-impact actions | L3 Application | Not stated in sources |
| Output filtering and redaction | L3 Application | Not stated in sources |
| Activity logs | L3 Application | Not stated in sources |
| Data minimisation | L3 Application | Not stated in sources |
| Sensitive-data checks on outbound data | L3 Application | Not stated in sources |
Evidence
- What was done File and surrounding prompts deleted; unapproved AI domains blocked
No source states: when it happened, how many records were exposed, the direct loss, a count of the other parties affected, any court case, the cause, what set it off, what was done.
Show the full account: 7 items from the primary sources
What happened, in order
- May 5 The bank found out about the incident.
- May 7 The company decided the incident was material.
- May 11 CB Financial filed a Form 8-K disclosing the incident.
Not dated in the sources:
- The bank already offered an approved AI tool to cleared employees before the incident.
- The file and surrounding prompts were deleted after upload was confirmed.
- After the incident the bank blocked unapproved AI tool domains and worked with the AI company on cleanup.
- After the incident the bank tightened employee access to customer data and expanded data-security policies.
Show every field, with its source and grade
What happened
Impact
Context
Notes on gaps
Reversibility is not stated and is left out. The data-loss-prevention policies are mentioned, but the sources do not say they failed or held for the browser upload, so no DLP control is linked. The 36-hour rule and Gramm-Leach-Bliley duties are general background from the article, not facts the bank stated about its own notices, and the article says it is unclear whether these were met.
LR-2026-0605 · version 1 · published 2026-11-01 after automatic checks; not yet checked field by field by a person · JSON