Berichten met de tag Brussels Effect

EU AI Act Compliance for Global Enterprises: The Daiki Solution for Mandatory AI Governance

The European Union's AI Act has crossed the line from proposal to binding law, and its phased rollout is now an active clock rather than a distant horizon. Daiki, the AI and quantum governance company co-founded by Mauritz Kop, has published an account of what mandatory AI governance demands of global enterprises—and how an integrated, standards-based platform can convert that obligation into a strategic advantage.

A calendar that has already started

The Act entered into force on August 1, 2024. The prohibition on unacceptable-risk practices took effect in February 2025; obligations for general-purpose AI models began in August 2025; and full application—conformity assessments, CE marking, EU-database registration, post-market monitoring for high-risk systems—arrives on August 2, 2026, with a final grace period for regulated-product components running to 2027. Because the Act binds any provider whose systems reach the EU market or whose outputs are used within it, its reach is extraterritorial: a firm headquartered in New York or Singapore is squarely within scope, and penalties of up to €35 million or 7% of worldwide turnover make non-compliance a board-level risk.

The pyramid of criticality

The Act's organizing idea is a risk-based pyramid: unacceptable-risk practices are banned; high-risk systems—reaching common enterprise uses in hiring, credit scoring, and critical infrastructure—carry the heaviest lifecycle obligations; limited-risk systems owe transparency; and minimal-risk applications attract no new mandates. The Daiki solution operationalizes that structure, classifying each system through a rules engine aligned with the Act's definitions and Annex III, then triggering a workflow proportionate to its tier, with every action logged to an auditable evidence trail. Its architecture is anchored in ISO/IEC 42001, bridged to the NIST AI Risk Management Framework, and bounded throughout by deliberate human oversight—mirroring the Act's own Article 14.

A risk-based reading with a documented lineage

The post's central reading—that a risk-based regime rewards organizations able to prove their governance—has a clear history in Kop's scholarship. As Mauritz Kop's record of work shows, his 2021 analysis of the EU AI Act anticipated the four-tier architecture that is now law, and the same logic carries across the Atlantic to California's compute-threshold approach for frontier models. For general counsel and compliance leaders, the practical takeaway is consistent: build one coherent, standards-based governance system now—rather than a reactive checklist per statute—and the era of enforcement will reward exactly the discipline the era of voluntary principles merely recommended. Mandatory AI governance, as the post observes, is here to stay; the enterprises that treat it as design rather than damage control will be the trusted artificial intelligence leaders of the regulated decade ahead.

Meer lezen

Artificial Intelligence, European Law, GovernanceMauritz Kop23 juni 2025Mauritz Kop, AI, Regulation, Policy, Governance, Standards, Daiki, EU AI Act, AI Governance, ISO/IEC 42001, NIST AI RMF, Brussels Effect, Risk-Based Regulation, Frontier AI

Towards an Atomic Agency for Quantum-AI

Stanford, CA May 5, 2025 — Today, Mauritz Kop published interdisciplinary research proposing “A Principled Approach to Quantum Technologies”, and the establishment of an “Atomic Agency for Quantum-AI” on the website of the European Commission. The Atomic Agency essay analyzes emerging AI and quantum technology (including their increasing complementarity and interdependence embodied in quantum-AI hybrids) regulation, export controls, and technical standards in the U.S., EU, and China, comparing legislative efforts anno 2025 to strategically balance the benefits and risks of these transformative technologies through the lens of their distinct innovation systems. The Principled Approach paper posits that quantum technology's dual use character brings with it the need to balance maximizing benefits with mitigating risks. In this spirit, the paper argues that quantum technology development should best be guided by a framework for Responsible Quantum Technology, operationalized by a set of tailored principles to optimize risk-benefit curves. Download the paper here.

Towards an Atomic Agency for Quantum-AI

The article analyzes emerging regulation, export controls, and technical standards for both quantum and AI (including their increasing complementarity and interdependence embodied in quantum-AI hybrids) in the U.S., EU, and China and offers novel conceptual frameworks to steward these technologies towards shared global benefit.

Key Takeaways:

1. Converging Need for Responsible Governance: Despite distinct innovation philosophies (market-driven in the U.S., values-based in the EU, and state-driven in China), there's a growing international consensus on the necessity for principled and responsible technology governance for both AI and quantum technologies.

2. The 'Washington Effect' vs. 'Beijing Effect': The U.S.'s security-centric policies are creating a de facto "Washington effect," potentially setting global rules for quantum law but risking premature regulation. Conversely, China's push for state-aligned standardization (e.g., via the Digital Silk Road) signals a "Beijing effect," which could export autocratic norms and fragment global interoperability, a trend exacerbated by unilateral export controls.

3. Potential U.S., EU and China Visions on a Quantum Governance Act: Given the distinct innovation ecosystems and strategic priorities of the United States, the European Union, and China, it is instructive to envision how each might hypothetically structure a dedicated legislative framework for quantum technologies. The research outlines hypothetical "Quantum Governance Acts" for each, reflecting their respective governance philosophies and innovation models, while also considering pathways towards greater international alignment based on shared values:

a. United States: Removing Barriers for U.S. Quantum Technology Act (deregulation, industrial standards-centric approach, Safeguarding through Advancing quantum technology, prioritizing market dynamism, national & economic security, and defense).

b. European Union: EU Quantum Act (harmonized regulation rooted in fundamental rights and societal benefit based on New Legislative Framework while incorporating elements from European Chips Act, renewed focus on defense via “European DARPA”).

c. China: Comprehensive Quantum Law (Safeguarding state control while Advancing state goals, blending elements of authoritarian governance with surveillance capitalism, integration of civilian and military sectors, self-reliance, exporting state norms & values through technical standards).

4. Global Challenges & Opportunities for Alignment: Faced with planetary challenges like disease, inequality and climate change, aligning on Responsible Quantum Technology (RQT) norms and standards is a critical global opportunity. The article cautions against a simplistic zero-sum game or Cold-War redux narrative for quantum competition, arguing it hinders vital international cooperation.

5. Quantum-Relativistic Innovation Theory of Everything: Philosophical thought experiment to understand innovation dynamics by drawing analogies from quantum mechanics (uncertainty, superposition at micro-level) and general relativity (context, structure at macro-level), theories about the fundamental nature of reality.

6. Smart Regulation and RQT by Design: Effective governance must move beyond mere restrictions to actively incentivize responsible behaviors, promoting "Responsible Quantum Technology (RQT) by design" through flexible instruments like Quantum Impact Assessments (QIA), RQT by design metrics, adaptive, modular legislation, & regulatory sandboxes.

7. Harmonized "Quantum Acquis Planétaire": The article advocates for a global body of Quantum Law ("Quantum Acquis Planétaire"), complemented by sector-specific practices. Such a quantum acquis would be anchored in universal ethical values and translated into foundational standards and agile legal guardrails. This requires inter-continental policymaking and strategic "recoupling" between major players like the U.S. and China, based on incentives and shared values (“what connects us” – e.g. human dignity, security, well-being).

8. An "Atomic Agency for Quantum-AI": A central proposal is the establishment of an international agency modeled after the International Atomic Energy Agency (IAEA). This body would aim to enforce a global acquis, deter a quantum arms race, ensure non-proliferation of dual-use quantum-AI technologies via safeguards implementation (inspired by nuclear governance), and potentially oversee a global UN Quantum Treaty.

9. Need for International Collaboration & Research Platforms: Realizing ambitious goals like fault-tolerant quantum centric supercomputing, and scalable topological qudits unlocking higher-dimensional quantum systems leveraging multi-level logic, requires collective global expertise and collaborative research platforms akin to CERN or ITER, challenging protectionist measures that stifle necessary cooperation. Immediate global actions should focus on leveraging quantum for the UN Sustainable Development Goals (SDGs), mitigating a 'Quantum Divide,' promoting quantum literacy, and building a skilled quantum workforce.

The research underscores the urgent need for robust global quantum-AI governance structures and calls for a shift from purely competitive dynamics towards pragmatic cooperation and the codification of a harmonized global framework.

Meer lezen

CQ Researcher Interviews Mauritz Kop on Regulating Artificial Intelligence and Quantum Computing

In late 2022, journalist Sarah Glazer interviewed Mauritz Kop—Founder of the Stanford Center for Responsible Quantum Technology (Stanford RQT)—for a CQ Researcher report, The Future of Artificial Intelligence — Can it be successfully regulated?, published under the lineage of Congressional Quarterly. Her questions ranged across the whole governance agenda: the regulatory gaps for artificial intelligence in the United States and Europe, the reach of the EU AI Act, the timing and skepticism around quantum computing, China and defense, and whether a machine could ever be sentient. Kop's written responses offer a compact statement of his position just before the field's law caught up with its technology.

The diagnosis: rules without teeth

Kop's opening line to Glazer was unsparing: "On both sides of the Atlantic, AI regulation is virtually nonexistent at the moment." The EU Artificial Intelligence Act, he explained, would change that—and not only in Europe. Because it is a Regulation rather than a Directive, it binds all Member States directly; and through the Brussels effect it sets a de facto global standard, exporting a risk-tiered, conformity-and-certification approach to AI well beyond the EU's borders.

Quantum is not "twenty years away"

The interview's sharpest thread was Kop's rebuttal to the familiar claim that quantum computing is perpetually two decades off. The framing, he argued, confuses engineering milestones with physical reality. Adversaries can harvest encrypted data now and decrypt it later once a capable machine exists, so the migration clock is already running. The physics is unforgiving: because qubits exploit superposition and entanglement, a quantum register explores an exponentially large state space, and an algorithm like Shor's turns that into the ability to break the assumptions behind today's public-key encryption. The argument echoes the warning in the Foreign Policy op-ed that drew Glazer to him in the first place.

Govern the systems we have

On machine sentience, Kop kept capability and consciousness apart: quantum effects make hard computations tractable, but they do not manufacture an inner life, and nothing in the physics confers rights on a model. The serious work is to govern the AI and quantum systems already in deployment—an interdisciplinary, standards-first program Kop has carried into venues from the U.S. Senate to his scholar profile. Read in 2026, after the EU AI Act's adoption and NIST's finalization of its first three post-quantum cryptography standards, the 2022 conversation looks less like commentary than like an early reading of developments now settled in law.

Meer lezen

Artificial Intelligence, Quantum Computing, PolicyMauritz Kop25 november 2022Mauritz Kop, AI, Regulation, Policy, Quantum Governance, Stanford RQT, CQ Researcher, Congressional Quarterly, EU AI Act, Sarah Glazer, Brussels Effect, Post-Quantum Cryptography

Blog over Kunstmatige Intelligentie, Quantum, Deep Learning, Blockchain en Big Data Law

A calendar that has already started

The pyramid of criticality

A risk-based reading with a documented lineage

Towards an Atomic Agency for Quantum-AI

The diagnosis: rules without teeth

Quantum is not "twenty years away"

Govern the systems we have

Juridische diensten en projecten in Nederland en daarbuiten