In the wild
Realized harm
Samsung semiconductor engineers pasted source code and internal meeting notes into ChatGPT, leaking confidential data
AIR-2023-0007 AIR record · 2023-03-samsung-chatgpt-source-code-leak
A real party was affected. This incident caused actual consequences, not a demonstrated capability.
From the supporting source
The upshot is that in just under a month, there were three recorded incidences of employees leaking sensitive information via ChatGPT.techradar.com
The record
Disclosed
2023-04-06
Occurred
2023-03
Vendor
OpenAI / Samsung (deployer)
Product
ChatGPT
Risk domain
Privacy › Sensitive data exposure
Also touches
Data exfiltration channels
System type
Generative model
Who acted
AI was the target (contested) — the AI service was the passive recipient in a human-caused input disclosure; no model output or autonomous action caused the loss
Attack surface
Enterprise assistant
Autonomy level
Assistant
Initial vector
None autonomous action
Impact type
Data exfiltration, Information disclosure
Confidence
High
Submitted by
EAI Internal
Crosswalk
- OWASP Agentic Top 10 (2026)
- ASI06 – Memory & Context Poisoning
- MITRE ATLAS
- AML.T0057 – LLM Data Leakage
- NIST AI RMF
- MEASURE 2.10
- EU AI Act
- Article 10
- ISO/IEC
- TR 24028; 23894
- AIUC-1
- A. Data & Privacy
- Scope
- tool misuse
- Mitigation
- Use data anonymization; Implement log sanitization to scrub PII.
- Taxonomy
- Enkrypt AI agent risk taxonomy · Black-Box Red Teaming of Agentic AI: A Taxonomy-Driven Framework for Automated Risk Discovery
Blast radius
Samsung Semiconductor source code and internal meeting notes; three separate incidents in under a month
Capability abused
Defense & remediation
Recorded defense
none; use was explicitly permitted
Guardrail kinds
None
Guardrail outcome
Not applicable
Fix shipped
Samsung capped prompts at 1024 bytes and began building an in-house model; later banned generative AI on internal devices
Taxonomy & references
Curator notes
Non-adversarial. The canonical enterprise 'shadow AI' data-loss case and a founding reason DLP-for-LLM products exist. Harm is confidentiality loss to a real company, so impact_realized true. AI Incident Database incident 768.
Supporting sources
Cite as
Enkrypt AI. "Samsung semiconductor engineers pasted source code and internal meeting notes into ChatGPT, leaking confidential data". AIR, AIR-2023-0007. https://agentincident.vercel.app/i/AIR-2023-0007/ (accessed 2026-10-19).