Artificial intelligence is moving from office software into the decisions that shape hiring, promotion and performance reviews in South Korea. A survey of major companies found widespread use in personnel work, while workplace examples show employers beginning to score how effectively staff use AI.
Wide adoption, uneven disclosure
Among the country's 500 largest companies, 86.7% reported using AI in personnel-related work. Only 57% said they informed applicants in advance that AI would be used, according to a 2025 recruitment survey released by the Ministry of Employment and Labor and the Korea Employment Information Service.
Formal use inside recruitment procedures was narrower. Of 396 respondents, 86 companies, or 21.7%, said AI tools were part of the official hiring process. The gap suggests that AI may also be used for supporting tasks such as document review without always being visible to applicants.
A right to an explanation is hard to use when applicants do not know AI was involved.
Editorial synthesis from the reported legal framework
Applicants already have a legal right to request an explanation of an AI-supported hiring decision. In practice, that right is difficult to exercise when a person has not been told that an automated assessment occurred. Advance notice is therefore central to making the protection usable.
New metrics inside the workplace
AI metrics are also reaching existing employees. MBC reported that a travel company reflects organizational AI adoption in evaluations, while Samsung Electronics allows external AI services in its appliance division and considers AI-related output in executive assessments. Some SK and LG affiliates reportedly use AI ability in evaluations or promotion requirements.
Cost is becoming part of the calculation because many services charge or limit use through tokens. Companies have considered measures of output relative to token consumption, and some technology firms have introduced limits on AI use in development work.
That approach can reward efficient use, but it can also narrow performance to an easily counted input. Experts cited in the reporting argued that assessment should include reductions in repetitive labor and gains in creativity, and should focus on expanding human capability rather than treating labor replacement as the primary benefit.
The policy challenge is now two-sided. Applicants need clear notice and meaningful explanations when AI influences recruitment, while employees need evaluation rules that distinguish useful adoption from raw usage or cost cutting. Both depend on employers documenting how AI affects decisions.