Our Stance on the EU AI Act
This page is for informational purposes. It does not constitute legal advice.
MIRAI AI ORCHESTRA operates strictly as an AI automation infrastructure provider. We prioritize the safe integration of AI into business workflows, aligning with the core principles of the European Union Artificial Intelligence Act.
1. System Classification
We do not develop or deploy General Purpose AI (GPAI) models exceeding the systemic-risk threshold of $10^{25}$ cumulative training floating-point operations (FLOP). Our tools are categorized as specific business automation solutions, designed to assist—not replace—human judgment.
2. Deployer (Customer) Responsibilities
Under Article 26 of the EU AI Act, deployers have specific obligations. We design our systems to ensure you maintain full structural control:
- Human Oversight (Art. 26(1)): Our “Stop-first behavior” ensures significant actions require explicit human approval.
- Log Keeping (Art. 26(5)): Tamper-resistant records are maintained for accountability.
3. Timeline Integration
Following the August 2024 enforcement, we continuously monitor the phased implementation timeline—prohibited AI (6 months), GPAI (12 months), and high-risk AI (24–36 months)—to update our framework functions proactively.
