Compliance

High-Risk AI Application

AI applications with significant potential to affect rights, safety, or welfare.

How the laws define this

Both Draft Bill No. 14 on Artificial Intelligence Regulation and the AI Basic Act (Draft) agree that a high-risk AI application is defined by its potential to pose significant risk or impact to individuals' rights and safety, placing these systems under stricter oversight and enhanced compliance obligations. Where the two instruments diverge is in their scope criteria and specific procedural mandates. Draft Bill No. 14 on Artificial Intelligence Regulation includes individual welfare as a key criterion, cites specific examples such as healthcare, finance, and public services, and requires sector-specific guidelines alongside phased implementation. Conversely, the AI Basic Act (Draft) highlights risks to access to essential services and enforces its risk-classification framework specifically through mandatory risk assessments and certification. Consequently, entities evaluated under Draft Bill No. 14 face a sector-tailored, staged rollout, while those governed by the AI Basic Act (Draft) must meet formal risk assessment and certification requirements.

Synthesised from the 2 statutory definitions below. Each one is quoted in full, with its source.

How jurisdictions differ

What is distinctive in each definition, binding law in force first.

Jurisdiction	Instrument	Status	What is distinctive
Taiwan	Taiwan Artificial Intelligence Basic Act	In Force	Taiwan's definition adds access to essential services as a risk factor and specifies enhanced oversight, risk assessments, certification, and stricter compliance under a risk-classification framework.
Panama	Draft Bill No. 14 on Artificial Intelligence Regulation	Under Review	Panama's draft limits high-risk AI to uses or categories in sectors like healthcare, finance, and public services, with stricter oversight, sector-specific guidelines, phased implementation, and enhanced compliance.

What it triggers

Under Taiwan Artificial Intelligence Basic Act, high-risk AI applications are subject to enhanced oversight, risk assessments, certification, and stricter compliance requirements under the risk-classification framework.source
Under Panama's Draft Bill No. 14 on Artificial Intelligence Regulation, high-risk AI applications would be subject to stricter oversight, sector-specific guidelines, phased implementation, and enhanced compliance obligations.source

Definitions (2)

A high-risk AI application is an AI use or category identified based on its potential impact on fundamental rights, public safety, or individual welfare—examples cited include systems used in healthcare, finance, and public services. These applications are subject to stricter oversight, sector-specific guidelines, phased implementation, and enhanced compliance obligations under the draft bill.

Draft Bill No. 14 on Artificial Intelligence RegulationDefinition 1 of 2

An AI application defined as posing significant risk to individuals' rights, safety, or access to essential services; such applications are subject to the Act's enhanced oversight, risk assessments, certification, and stricter compliance requirements under the risk-classification framework.

AI Basic Act (Draft)Definition 2 of 2

Not to be confused with

High‑risk application

High-risk application focuses on significant impacts on life, rights, or critical infrastructure, while high-risk AI application emphasizes rights, safety, welfare, or essential services with regulatory duties.

High‑Risk Artificial Intelligence

High-risk artificial intelligence targets systems materially affecting life, safety, or fundamental rights, differing from high-risk AI application by its focus on potential impacts and sector-specific oversight.

Questions and answers

What obligations apply to high-risk AI applications in Taiwan?

High-risk AI applications in Taiwan are subject to the Artificial Intelligence Basic Act's enhanced oversight, risk assessments, certification, and stricter compliance requirements under the risk-classification framework.

What makes an AI system high-risk under Panama's draft bill?

Panama's draft identifies high-risk AI applications based on potential impact on fundamental rights, public safety, or individual welfare, with examples in healthcare, finance, and public services.

How do Taiwan and Panama define high-risk AI differently?

Taiwan includes risks to access to essential services and mandates certification and risk assessments, while Panama's draft focuses on specific sectors like healthcare with phased implementation and sector-specific guidelines.