South Korea adopts seven AI ethics principles

South Korea’s Ministry of Science and ICT (MSIT) and the Korea Information Society Development Institute (KISDI) have confirmed the country’s revised national AI ethics principles, adopted on 21 August following review at the 12th Science and Technology Ministers’ Meeting, according to the ministry.

The principles are built around three core values, human dignity, the common good of society, and the sustainability of humanity, and are structured into seven practical principles: human-centeredness, privacy protection, fairness and inclusiveness, accountability, safety, reliability and transparency. Accountability is the newest addition, introduced during a revision process that began with a first draft in May and incorporated more than 100 public comments ahead of a second draft presented at a public forum on 14 August.

The framework updates guidance first issued in 2020 and is intended to work alongside South Korea’s AI Basic Act, which took effect earlier this year. MSIT plans to issue sector-specific self-checklists, case-based explanatory guides and ethics training materials to support implementation.

The principles are non-binding. Local reporting has noted the framework lacks legal force and specificity, with the ministry positioning it as a “soft law” complement to statutory regulation rather than a precursor to further legislation.