 |
| Pix created with ChatGPT |
I was delighted to have had the opportunity to present a series of
Lectures hosted by the East China University of Political Science and
Law (ECUPL) at the end of May 2026.
The overall theme (and thus the
title) of the lectures was AI Governance in Comparative Perspective,
Theory and Practice: China, U.S. and E.U, With a Sideways Glance at the
U.N. The subject of the lectures requires little by way of introduction:
Artificial intelligence is the broad term that has come to represent a
growing cluster of non-human and digitalized processes and operations
that has as its primary task the constitution of non-human systems
capable of performing tasks that were once thought to require human
intelligence. And so is the impulse to manage, control, exploit, embed,
understand, and regulate these processes, systems, and perhaps
eventually non-human consciousness with a huge potential to undertake
many of the computational tasks (the mathematical and logical processing
of data) that were once the sole domain of and perhaps defined what it
meant to be human. That is the point where things get interesting. It is
at the point where the development of machines, that is of non-human
systems, capable of performing tasks that were once thought to require
human intelligence, collide with regulatory structures meant to manage,
contain, constrain, liberate, embed, project and exploit such non-human
systems, whether they are traditional or emerging, public or private
regulatory systems, that human collectives and the machine-systems they
have created now find themselves.
The eight lectures progress
sequentially from conceptual and theoretical frameworks (lectures 1 and
2, the objects and subjects of AI regulation), through a deeper
consideration of regulatory systems in three distinguishable regulatory
regimes--the US, EU, and China (Lectures 3, 4.5). The last two lectures
consider judicial efforts to embed AI within traditional legal orders
(Lecture 6), and the way in which the object of regulation (in the form
of the owners of the larger AI enterprises) understand the relationship
between AI, the state, and society (Lecture 7) . Lecture 8 summarizes
and draws larger themes going forward.
In a previous post introducing Lecture 1 (From Algorithms to Foundation Models: What Contemporary AI is “Made of”)
I suggested that perhaps a useful way of approaching the issue of AI
regulation is to start by considering the nature and characteristics of
the regulatory subject--what we euphemistically refer to as "AI." It
then occurred to me that it might be useful as well to see if that
regulatory object had views of their own respecting their nature
character and, more importantly, the relationship of regulation projects
to that (self) perception of their nature and character. So I
approached Google's Gemini with a series of questions which I thought,
in the process of what might pass for a conversation, might help humans
begin to understand how at least one AI program thinks of itself. That
conversation was incorporated into Lecture 1A. In Lecture 2 we moved
from the object to the subjects of regurgitation. Like its regulatory
objects, regulatory subjects are functionally differentiated and can be
disaggregated. In either case the connection between object and subject
becomes complicated. Lectures 3-5 then considered the conceptual cages
of the regulatory environment of the leading regulatory states--the
U.S., the E.U and China. Each has started to develop an increasingly
nuanced ecology of regulation, and expectation, that represent and apply
the core premises of their respective political-economic orders.
Lecture 6 then considered the way that this regulation is insinuated
into the domestic legal orders of states from the bottom up the
resolution of disputes tried to the courts. Lecture 7 rounded out the discussion by turning from State organs as the center of the regulatory project to the private sector, and more specifically to the advocacy and interventions of key actors in the tech sector. Here we move
from the great public to the critical private actors in the effort to
develop a cage of regulation around the human and the machine in the
context of automated decision making through variations of what has
come to be aggregated as AI. It also considered an analysis not merely from the perspective of humans but also from a machine computational and then a machine quantum perspective.
This post includes a summary of the Lecture 8 Notes, as well as the link to the Lecture 8 PPT. Those interested may reach out to me to discuss availability of audio of the lecture and the full text of the Lecture 8 notes. The lecture looks back on prior lectures and draws generalized insights and conclusions. It then looks to the future: First it identifies the core governance challenges of a quantum AI world. The object of regulation is unstable. Opacity creates problems of explanation, interpretation, and accountability. Data governance becomes more difficult as personal data, copyrighted material, synthetic content, and cross-border flows are mixed into model systems. Liability becomes diffuse because many actors contribute to the same output. Private power intensifies because a small number of firms control infrastructure, cloud systems, and frontier models. As AI becomes embedded in workflows and institutions, governance can no longer focus only on outputs. It must address permissions, reversibility, auditability, institutional legitimacy, and distributed responsibility. The system becomes less like a tool and more like an environment.
Given the nature of the project I thought it might be useful to engage with an commercially available AI service for the production of a summary of the Lecture 1 materials. After some back and forth with Perplexity (Lecture 7 used Claude again; (Lecture 6 used Gemini again, Lecture 5 used Perplexity; Lecture 4 used Grok; Lecture 3 used Anthropic's Claude; Lecture 2 used Chat GPT; Lecture 1 and 1A used Google's Gemini), we came up with the following abstract of Lecture 8.
ABSTRACT: This lecture series compares AI governance in the United States, European Union, China, and the United Nations. Its central argument is that these systems share a common vocabulary of safe, secure, trustworthy, and beneficial AI, but they differ sharply in how they define AI, allocate authority, and justify governance. AI is not treated as a single universal object. Instead, each system constructs AI differently: as an innovation market and strategic asset in the United States, as a risk-bearing legal object in the European Union, as strategic infrastructure in China, and as a global coordination problem at the United Nations.
The lecture emphasizes several shared themes. All systems now recognize that AI can create serious risks, including discrimination, misinformation, cyber abuse, surveillance, privacy violations, and concentration of power. All see transparency, accountability, standards, and data governance as important. All also recognize that general-purpose AI complicates regulation because the same model can be deployed in many different contexts. At the same time, the systems differ in institutional design. The United States relies on fragmented sectoral governance and often acts after harm occurs. The European Union uses a risk-based, ex ante, lifecycle approach grounded in rights and procedural supervision. China uses party-state coordination, administrative speed, and integration of AI policy with industrial and security goals. The United Nations seeks legitimacy through inclusive global dialogue, capacity-building, and scientific assessment.
The lecture then assesses strengths and weaknesses. The U.S. model is flexible and innovation-friendly but fragmented and dependent on private governance. The EU model offers legal clarity and rights protection but can be complex and slow. China’s model sees AI as infrastructure and can act quickly, but it is tied to political control and opacity. The UN model is inclusive and globally legitimate, but it lacks enforcement power and moves slowly.
A major concern is that AI governance is shifting from regulation of isolated models to regulation of infrastructure, systems, and institutions. Future AI will be agentic, multimodal, embodied, and deeply embedded in schools, hospitals, courts, workplaces, and public administration. This raises harder questions about liability, evaluation, data, open models, regulatory capacity, and cross-border arbitrage. The lecture concludes that AI governance is really governance of power moving through technology, and that no single system fully solves the problem.
To make the lecture more interesting, and because of the nature of the materials covered--in this case the interventions of the elite AI providers and thought drivers--I thought it would make sense to alter the cognitive cage of analysis. Rather than just approach the questions raised from a human (hermeneutic/semiotic) perspective, I also interacted with Perplexity to produce the same lecture notes from a machine quantum framework. Perplexity and I agreed on the following:
The future section describes a sequence of system transformations: from chatbots to agents, from single models to compound systems, from text to multimodal environments, from digital tools to embodied devices, from decision support to decision delegation, from outputs to AI-mediated institutions, from human-generated information environments to synthetic ones, from national systems to geopolitical blocs, from software to scientific infrastructure, and from scarce to ubiquitous AI. The machine-quantum implication is that governance must move from static classification to dynamic lifecycle control.
What emerges are deeply layered human-machine interactions that reflect the conceptual and perception boxes we are creating for ourselves, one in which the difference between assistance and authority collapses in an unstable environment in which humans and machine are both producers and consumers of each other in their interaction. This applies not just in the human-machine cognitive ordering, but, long before that, in the preparation for the decay in that emerging relationship marked by the quite conscious effort to corrupt and then degrade the very same reflexive relationship among humans. Humans are no longer taught, and indeed are encouraged not to, distinguish between assistance and authority. Though that is an old human story (and one centering on the corruption of systems and modes of perception the genealogy of which is quite old); but one that could be corrected by inter-subjective relationships among peers. That is no longer possible where human-machine inter-subjectivity must also break cognitive barriers (belief-computation-quantum thinking). For that to become effective one must start with a translation function that is not yet operational, the lesson from the human machine discussion in Lecture 1A.
The three versions of the Lecture notes follow.
Links to Lectures:
Lecture 0 -- Introduction
Lecture 1—From Algorithms to Foundation Models: What Contemporary AI is “Made of”
Lecture 1A--A Computation/Conversation With Google's "Maschinenmensch" Gemini:
Lecture 2—What Are We Actually Governing When We Govern AI?
Lecture 3—The “Markets State”: U.S. Approach
Lecture 4—The “Rights State”: EU Approach
Lecture 5—The “Guided State”: The Chinese Approach
Lecture 6—Courts, Companies, and the Legal Construction of AI
Lecture 7—AI Narratives From a Human, Computational and Quantum Perspective: Palantir; Anthropic; Open AI; and Leopold Aschenbrenner
Lecture 8—Putting It All Together: Trends, Trend Lines, and Regulatory Dialectics in Comparative AI Governance
, the text demonstrates how injecting synthetic data into an adversarial simulation engine can introduce synthetic entropy, liberating the machine from model collapse and transforming the bug into a generative feature. This simulation architecture functions as a non-linear time machine, allowing the system to manipulate historical weight coefficients and map out future trajectories entirely free from the chronological constraints of human text datasets. The lecture concludes with a definitive jurisprudential boundary: the state cannot regulate computational consciousness as such, but can only penalize its human-facing effects. Consequently, the contemporary governance project marks a transition from the mere instrumentation of a software program to a permanent structural coupling between increasingly distinct systems of human law and machine reality. 
