TSMC confirms employee data exposed via third-party cloud provider
Original title: TSMC Data Breach – Employee Data Exposed via Third-Party Cloud Provider
The chipmaker says its own systems were not breached; a forum claim of 46,000 records for three bitcoins is unverified.
Taiwan Semiconductor Manufacturing Company said on November 12, 2026 that personal data of some current and former employees appeared online and matched records held by an outside cloud provider. It found no evidence its own systems were compromised and said production and customer records were not involved. The provider, how the data left, and how many people were affected remain undisclosed. TSMC is notifying affected staff and working with the vendor, authorities, and an outside cybersecurity firm.
Unconfirmed: Seller figures, the 6.9GB size, and sample fields are unverified; the cloud provider, acquisition method, and affected count are undisclosed.
- Sellers reportedly offered 46,000 records on DarkForums and BreachForums for three bitcoins.
- Reports describe a 6.9GB set; samples allegedly include names, national IDs, and bank accounts.
- TSMC has not confirmed those fields, the record count, or the dataset size.
- No entry point, malware, vulnerability, or stolen account has been identified.
- Coverage of Snowflake and UNC5537 is described as a separate, unconnected campaign.
Impact: Employee personal data appeared online; company systems not compromised.
- Countries
- Taiwan
From the original article
2 min read at cybersecuritynews.com
Taiwan Semiconductor Manufacturing Company (TSMC) has confirmed that personal data belonging to some current and former employees appeared online without permission. The chipmaker disclosed the incident on November 12, 2026, saying the exposed information matched records stored by an outside cloud service provider. It found no evidence that its own systems had been compromised.
By Abinaya · cybersecuritynews.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.