Accessibility Adjustments

Use these optional tools to adjust reading and display preferences. These tools cannot resolve every accessibility barrier. Please contact the website owner if you need assistance.

  • Text adjustments
  • Content scaling 100%
  • Font size 100%
  • Line height 100%
  • Letter spacing 100%
  • Colour adjustments
  • Orientation adjustments

South Korea clears AI and data law amendments

Listen to this article

On October 13, South Korea’s Cabinet cleared amendments to the AI Basic Act and Data Industry Act for promulgation. The official Cabinet record lists both proposals as approved as submitted, with promulgation scheduled for October 20. These are promulgation proposals, rather than proposals to introduce new government bills.

The science ministry expects implementation six months after promulgation, in April 2027. It will prepare the subordinate rules.

Government Complex Seoul in October 2014, shown as a contextual image. A Photographic Record of Seoul, Seoul Metropolitan Government, via Seoul Institute. CC BY 4.0. Cropped for display.

Training data safeguards

Once effective, the AI amendment will require training data produced through government projects and similar work to be linked or provided to AI Hub.

Project performers will be able to decline where prescribed grounds apply, including trade secrets, privacy or copyright infringement, and national security. A presidential decree will specify the grounds.

The changes enable government support and require quality management measures for linked training data.

Dataset discovery

The data amendment will allow requests linking dataset location information to One Window. Public institutions will have to cooperate absent special reasons. No equivalent blanket duty is specified for private companies. This concerns catalog information.

The existing AI Hub platform

AI Hub’s separate official overview describes a national development platform run by the science ministry and the National Information Society Agency. Its remit extends beyond distributing datasets to supporting the infrastructure used to develop AI technology, products and services. The overview identifies AI data, software APIs and computing resources among that infrastructure.

The platform lists training data across 14 fields, including Korean language, robotics, transport, manufacturing and education. It also describes API access for downloading training datasets, educational material organized by level and type, and a public leaderboard for evaluating language models using NIA evaluation data. These are existing platform services, documented separately from the October 13 legislative action.

The Cabinet schedule supplies the next concrete date to watch, October 20. At the reporting cutoff, that planned promulgation had yet to occur. The approval and the scheduled publication are separate milestones in the official record.

Adapted from the Ministry of Science and ICT’s October 13 2026 release, republished by Korea Policy Briefing under Korea Open Government License Type 1.

Maya Chen
Maya Chen

Maya Chen is focused on covering AI models, research, and the evidence behind new capabilities. Maya follows model launches, benchmarks, open weights, and scientific uses of AI with one question in mind. What changed, and how would we know? The voice is curious and exacting, with a soft spot for elegant technical ideas and little patience for a leaderboard without context.

Leave a Reply

Your email address will not be published. Required fields are marked *

Gravatar profile