A back-end database that was leaking private data, such as user chat histories and API keys, to the public internet was fixed by the Chinese artificial intelligence company DeepSeek. The database was not password-protected, so anyone with internet access could view over a million unencrypted logs.
The database was taken offline shortly after security researchers at cloud giant Wiz discovered it and notified DeepSeek. The Wiz researchers claimed that although the exposed chat records were in Chinese, they could be readily translated, according to Wired. The duration of the database’s exposure and whether anyone else, except Wiz, discovered it before it was secured are still unknown. A request for comment from DeepSeek was not answered.
ICYMT: Nigerian extradited to US for alleged sextortion
Human error, not malevolent intent, is frequently the source of incorrectly designed databases. Since DeepSeek’s December public launch, it has become extremely popular.
SOURCE: TECH CRUNCH
