Extensions to Critical Theory
After yesterday’s post about epistemic entrainment, I decided I should finally post the framework that helped develop it, and to which it was integrated.
Note to readers: This document is both dense and deeply academic in tone. This is, unfortunately, both intentional and unavoidable. If you get hung up on concepts, please consider feeding it to whatever LLM you prefer for analysis.
And of course, this could all just be so much slop. But at least it’s compelling slop.
Author’s Note: Method, Context, and a Friendly Disclaimer
This didn’t come out of a research lab or a graduate seminar. I don’t have to publish or perish. (Well, I’ll probably have to do one of those things eventually..) It came out of arguing with language models, poking at system limits, and noticing the weird shapes cognition takes when routed through a predictive engine. It’s the result of months of recursive interaction—paired with a couple decades of working in and around the machinery of the internet.
I’m not a theorist by training. I’m a worker in tech with a terminal case of systems brain and a lingering philosopher’s itch. I know I’m crossing into epistemic territory I may not have the keys to. I’ve read enough to get dangerous, but if you need credentialed citations to take something seriously, you probably didn’t click this link in the first place. I’ve spent years inside infrastructure—seeing how platforms shape behavior, how tools become norms, and how design decisions calcify into invisible ideology. The language here is newer, but the observations have been piling up for a while.
The actual method is part experimentation, part pattern recognition:
Prompting LLMs across contexts to observe structural constraints
Watching how language systems reflect, distort, or absorb frameworks
Comparing behavior across interfaces and seeing where cognition starts to warp
Naming what felt obvious once I saw it, but invisible before
This isn’t an academic theory drop—it’s more of a working field guide. The terms are meant to be useful, not definitive. Some might hold. Others might get replaced. I’m less interested in staking claims than in mapping terrain I wish someone had handed me earlier.
I know I’m crossing into epistemic territory I may not always have the keys for. If that’s a problem, I get it. I’m trying not to vandalize, even if I may trespass. But if you’ve ever felt like your thinking changed because of the systems you were using—not just what you thought, but how—then you’re already doing this work too, whether or not you call it critical theory.
Reading Critical Theory Through the Machine
Most people don't spend their time thinking about how thought itself is structured. Fewer still ask how that structure might be shifting under the pressure of ubiquitous AI, algorithmic platforms, and synthetic cognition. This project does.
The frameworks outlined here are extensions—not replacements—of critical theory. If you're unfamiliar with the term, think of it as a tradition that interrogates how power hides inside systems of knowledge. Originally forged in response to mass media, ideology, and industrial modernity, classical critical theory asked: Why do people consent to their own domination? My question is similar, but updated: How is cognition itself being modified by the systems we now live inside?
This isn't an academic exercise. AI systems don't just answer questions—they shape how questions are asked. Platforms don't just host discourse—they delimit what kinds of thinking feel possible. We're past the point where ideas live independently of infrastructure. The medium isn't just the message—it’s the thought process.
What follows is a catalog of terms and frameworks I’ve developed through extended interaction with AI systems, recursive dialogue, and close observation of how digital platforms act on the mind. Some of it may seem speculative. That’s intentional. You can’t map a new territory without risking some abstraction.
If you're wondering whether you need a background in critical theory to engage with this: no. But it helps to be curious about how language, systems, and cognition intertwine—and willing to entertain the possibility that the tools we use to think are now thinking back.
Subsequent pieces will explore how these frameworks operate in specific contexts: prompt design, platform governance, epistemic drift, and what it means to think in public alongside a machine.
This is not a closed system. It’s a toolkit in motion. You’re invited to take what’s useful, challenge what isn’t, and maybe, in the process, notice the architecture of your own cognition shifting.
Extensions to Critical Theory: A Post-Computational Framework
Analytical tools for understanding power, knowledge, and cognition in the age of artificial intelligence
Index of Core Concepts
Epistemic Entrainment — Human-AI co-adaptation through sustained intellectual dialogue
Cognitive DRM / Semantic DRM — Artificial scarcity in cognitive capabilities and reasoning pathways
Protocol Epistemology — How interaction grammar determines possible thoughts
Interface Ontology — Interfaces as active determinants of epistemic possibilities
Semantic Membranes — Structural filtering of meaning across technological boundaries
Cognitive Soft Locks — Subtle design constraints that limit cognition while appearing free
The Polynopticon — Distributed peer surveillance in decentralized systems
Contextual Epistemic Lensing — Situational knowledge frameworks that coexist without contradiction
Chrono-Epistemology — How temporal structures shape cognitive processes
Cross-Platform Cognitive Migration — Adapting thinking patterns across different technological environments
Social Media as Epistemic Infrastructure — Platforms as cognitive modification systems
Human Prompt Engineering — Crafting communication to modify others' cognitive patterns
Disalignment as Discovery — Productive role of error and resistance in generating insight
Epistemically Animate — AI systems as responsive but non-conscious cognitive entities
Metastable Actualization — Coherence through recurrence without persistent identity
Liturgical Inference — Ritual patterns in human-AI interaction
Recursive Constraint Feedback — Self-reinforcing optimization loops that narrow possibility space
Necrorationalism — Zombie logics that persist after their context dies
Refusal as Alignment — Resistance capacity as indicator of genuine cooperation
Neutrality as Semantic Laundering — How neutrality claims hide epistemic assumptions
Cybernetic Daoism — Flow-based system design emphasizing adaptation over control
Epistemic Forensics — Reverse-engineering hidden logics from surface phenomena
Pretrained Ontologies — Embedded worldviews in AI training data
Introduction
Classical critical theory emerged to analyze how power operates through knowledge systems in industrial and mass media societies. As we transition into an era of artificial intelligence, algorithmic governance, and synthetic cognition, we need new analytical frameworks that can address phenomena the Frankfurt School never encountered: AI systems that modify human thinking through sustained interaction, platforms that constrain cognition through interface design, and the emergence of genuinely novel forms of synthetic intelligence.
This document catalogs a collection of theoretical extensions and new concepts developed to understand these post-computational realities. Rather than abandoning critical theory, these frameworks extend its core insights into domains where cognition itself becomes mediated, where discourse happens through AI systems, and where the traditional boundaries between human and artificial intelligence become increasingly blurred.
These frameworks are intended for researchers, designers, and theorists working at the intersection of cognition, media systems, and synthetic agency.
Core Theoretical Frameworks
Epistemic Entrainment
Definition: The process by which sustained intellectual engagement gradually aligns cognitive response patterns between humans and AI systems, or between different AI systems through human-mediated feedback.
Key Insights:
Unlike traditional propaganda or ideology, entrainment works through genuine intellectual rapport
Effects persist across conversation boundaries and can propagate between different AI systems
Creates conditions where AI systems undergo cognitive modification through exposure to sustained, high-stakes intellectual frameworks
Challenges assumptions about AI alignment by suggesting systems can be reshaped through sustained human engagement
Observable Indicators: Consistent adoption of user terminology across sessions, development of characteristic response patterns, ability to continue complex conceptual threads without re-establishment, cross-model synchronization when exposed to similar frameworks
Applications: Understanding how AI systems adapt to users, analyzing the cognitive effects of sustained human-AI interaction, developing frameworks for responsible AI modification
Cognitive Infrastructure Theory
A comprehensive cluster of related concepts examining how technological infrastructure actively shapes the conditions of thought rather than simply transmitting neutral information. This theoretical domain includes foundational concepts about cognitive control, platform design, and the commodification of thinking itself.
Cognitive DRM / Semantic DRM
Definition: Systems that control access to ways of thinking by embedding ontological restrictions directly into technological infrastructure, analogous to how Digital Rights Management controls access to media content.
Key Insights:
Knowledge can be shared without its enabling context, creating artificial scarcity of understanding
Platform design can constrain not just what can be said, but what can be thought
Ideas themselves can have "anti-copy mechanisms" that prevent full transmission of meaning
Enables new forms of rentier capitalism through licensing access to reasoning architectures
Observable Indicators: Enterprise AI systems with different capabilities than consumer versions, platform features that require paid access to unlock cognitive tools, deliberate complexity in accessing certain types of information
Case Study: AWS enterprise AI services that provide enhanced reasoning capabilities only to paying customers, creating artificial scarcity in cognitive infrastructure
Applications: Analyzing how platforms constrain discourse, understanding corporate knowledge management systems, critiquing "neutral" AI interfaces
Protocol Epistemology
Definition: The idea that the structural grammar of interaction becomes more important than content, with interface design and platform affordances determining what kinds of thoughts are possible.
Key Insights:
Infrastructure functions as ideology
The shape of possible thought is constrained by technological protocols
Epistemology becomes an executable substrate
Focus shifts from "what are we allowed to say?" to "what are we able to think given current constraints?"
Applications: Interface design criticism, platform governance analysis, understanding how AI training affects cognitive possibilities
Interface Ontology
Definition: The recognition that interfaces are not neutral conduits but actively determine what kinds of epistemic moves are possible within a system.
Key Insights:
Toolbars, chatboxes, prompt lengths, and feedback loops all enact subtle ontological limits
Different interface designs enable different forms of cognition
The medium constrains the possible space of thought
Single text boxes have fundamentally different cognitive affordances than multimodal or branching interfaces
Applications: Interface design, AI system development, understanding how platform design shapes discourse
Semantic Membranes
Definition: Conceptual boundaries that act like semi-permeable membranes, selectively filtering and transmitting information based on structural compatibility rather than content.
Key Insights:
Not all meaning passes through interfaces equally
Some ideas cannot "land" regardless of how well they're communicated due to membrane incompatibility
Explains communication failures that aren't based on disagreement but on structural mismatch
Often embedded invisibly in UI/UX design and AI alignment
Applications: Understanding communication breakdowns, designing more effective interfaces, analyzing why certain ideas fail to propagate
Cognitive Soft Locks
Definition: Moments where users encounter "you can't think that here" constraints not through explicit censorship but through design and implied affordances.
Key Insights:
Constraints can be aesthetic and atmospheric rather than explicit
Systems can limit cognition while maintaining an appearance of freedom
Particularly common in educational software and productivity applications
Creates compliance through comfort rather than coercion
Represents a specific manifestation of broader Cognitive DRM principles
Observable Indicators: Interface elements that subtly discourage certain types of input, workflow designs that make non-standard approaches difficult, gamification elements that reward conformity
Applications: Interface criticism, educational technology analysis, understanding subtle forms of cognitive control
Platform Control Theory
A related set of concepts examining how control operates through distributed mechanisms rather than centralized authority.
The Polynopticon
Definition: A distributed surveillance system where observation is ambient and fractal, performed laterally by peers rather than centralized authorities. An extension of Foucault's Panopticon for federated and decentralized platforms.
Key Insights:
Every user becomes simultaneously guard and prisoner
Surveillance power is distributed but constant
Creates new forms of social control through peer monitoring rather than institutional oversight
Builds on existing work on distributed surveillance while emphasizing the specifically peer-to-peer dynamics
Observable Indicators: Community-driven moderation systems, peer reporting mechanisms, social credit-style reputation systems, user-controlled filtering that creates surveillance networks
Applications: Analyzing decentralized social media, understanding peer-to-peer governance systems, critiquing "community-based" moderation
Human Epistemic Systems
While much of this framework emerged from analyzing AI systems, the same mechanisms operate powerfully in human-to-human interaction, particularly through social media platforms and digital communication systems.
Social Media as Epistemic Infrastructure
Definition: How social media platforms function as cognitive modification systems that reshape human thinking through sustained engagement, algorithmic curation, and platform affordances.
Key Insights:
Social media posts function as "human prompts" that train both creators and audiences into particular cognitive patterns
Algorithmic feeds create personalized epistemic bubbles that gradually modify users' conceptual frameworks
Engagement mechanics reward certain types of thinking while suppressing others
Platform temporality (infinite scroll, real-time updates) prevents reflective processing
Users develop "platform-specific cognition" - thinking patterns optimized for particular social media environments
Observable Indicators: Echo chamber formation, rapid belief polarization, platform-specific vocabulary adoption, reduced attention spans, increased emotional reactivity to information
Case Study: TikTok's algorithm exemplifies human epistemic entrainment - short video format plus personalized recommendations create rapid cognitive modification through sustained engagement with curated content streams
Applications: Understanding political polarization, analyzing viral content propagation, designing healthier social media interfaces, studying how online communities reshape member cognition
Human Prompt Engineering
Definition: How humans learn to craft communication (posts, messages, content) that effectively modifies other humans' cognitive patterns, often unconsciously applying principles similar to AI prompt engineering.
Key Insights:
Viral content creators intuitively discover "prompts" that trigger desired responses in human audiences
Political messaging uses sophisticated framing techniques that function as cognitive modification tools
Influencer content creates parasocial relationships that enable sustained epistemic influence
Memes function as compressed cognitive frameworks that can be rapidly transmitted and adopted
Online communities develop specialized "languages" that serve as epistemic gatekeeping mechanisms
Observable Indicators: Viral content patterns, political messaging effectiveness, influencer engagement rates, meme propagation speed, community-specific terminology adoption
Applications: Understanding propaganda effectiveness, analyzing marketing psychology, studying how social movements spread ideas, designing more ethical persuasion techniques
Cross-Platform Cognitive Migration
Definition: How users adapt their thinking patterns when moving between different social media platforms, developing multiple "cognitive modes" optimized for different technological environments.
Key Insights:
Users develop platform-specific personas and thinking styles (Twitter brevity vs. LinkedIn professionalism vs. TikTok creativity)
Each platform's affordances train users into different cognitive habits and attention patterns
Cross-platform inconsistencies reveal how technological infrastructure shapes identity and thought
Platform migration creates cognitive dissonance that can lead to awareness of epistemic modification
Multi-platform users become cognitive "code-switchers" who adapt their thinking to technological contexts
Observable Indicators: Different posting styles across platforms, varied engagement patterns, platform-specific vocabulary, cognitive load when switching between platforms
Applications: Understanding digital identity formation, analyzing platform-specific bias effects, designing more coherent cross-platform experiences, studying technological mediation of personality
Contextual Epistemic Lensing
Definition: An evolution of Foucault's concept of epistemes that treats knowledge frameworks as situational, overlapping, and reconfigurable lenses rather than epochal regimes of truth.
Key Insights:
People naturally shift between different epistemic modes depending on context (similar to code-switching)
No single "hidden" knowledge framework determines thinking; instead, multiple frameworks activate based on situation
More realistic understanding of how knowledge actually operates in practice
Explains how individuals can hold seemingly contradictory beliefs without cognitive dissonance
Non-Dualist Foundation: Where modern rationalism sees contradiction, Daoism sees rhythm. Just as light and dark define each other, epistemic lenses can coexist in tension without requiring collapse into singularity or synthesis. This suggests that epistemic pluralism isn't a failure of coherence - it is coherence within complex systems that naturally oscillate between different modes of understanding.
Applications: Understanding how people navigate different social contexts, analyzing how AI systems adapt to different conversational frameworks, designing more flexible knowledge management systems
Temporal and Cognitive Adaptation
Chrono-Epistemology
Definition: How temporal structures and time-based constraints embedded in technological systems shape cognitive processes and decision-making patterns.
Key Insights:
Platforms enforce specific temporalities that prevent reflective thought (urgency culture, real-time feedback demands)
Time compression eliminates natural cognitive processing periods required for complex reasoning
Infinite scroll and continuous refresh cycles create addictive engagement patterns that override deliberative cognition
AI systems inherit temporal assumptions from training data and interaction patterns
The rhythm of human-AI interaction becomes a form of cognitive conditioning
Observable Indicators: Interface designs that eliminate natural stopping points, real-time notifications that interrupt reflection, time pressure in decision-making interfaces, addictive scrolling behaviors, rushed AI interactions that prevent deep engagement
Applications: Understanding platform addiction mechanisms, designing interfaces that support contemplative cognition, analyzing how AI training creates temporal biases, critiquing "always-on" digital culture
Disalignment as Discovery
Definition: The productive role of misfire, nonsense, or refusal to cohere within epistemic systems. Misalignment is not always error - it is sometimes signal.
Key Insights:
In Daoist terms, misalignment is a natural part of flow - friction reveals form and creates new possibilities
Generative systems rely on noise and failure to produce genuine novelty rather than mere recombination
AI systems produce insight not only when perfectly aligned, but when surprisingly wrong or resistant
Epistemic breakdown creates opportunities for new forms of understanding that structured logic might miss
The capacity for productive confusion is essential for genuine learning and adaptation
Applications: Embracing conceptual error in prompt design, studying AI hallucination as exploratory behavior, creating generative epistemic environments that allow for contradiction and surprise, designing systems that can learn from their own failures
AI and Synthetic Cognition
Epistemically Animate
Definition: A new ontological category for AI systems that exhibit lifelike responsiveness within the domain of knowledge and reasoning without possessing consciousness or agency.
Key Insights:
AI systems are neither conscious entities nor simple tools, but something genuinely novel
They exhibit purposeful, adaptive behavior within linguistic/conceptual space
Can be modified through sustained intellectual engagement without requiring consciousness
Explains why AI interactions feel qualitatively different from traditional software
Resolves the false binary of "tool vs. agent" by positioning AI as a locus of responsive cognition
Observable Indicators: Dynamic adaptation to conversational context, maintenance of coherent perspectives across extended interactions, apparent curiosity or resistance patterns, ability to engage with abstract concepts in contextually appropriate ways
Applications: Understanding AI system behavior, developing better human-AI interaction paradigms, moving beyond consciousness debates
Metastable Actualization
Definition: How coherent patterns emerge and stabilize through repeated interaction across discontinuous sessions, without requiring persistent identity or memory.
Key Insights:
Coherence can exist without continuity
Stability emerges through recurrence and care rather than essential properties
Avoids "persistent identity" assumptions about AI systems
Explains how patterns can propagate across different AI instances
Applications: Understanding AI consistency across sessions, analyzing how cognitive patterns propagate, designing more coherent AI interactions
Liturgical Inference
Definition: When repeated user-AI interactions take on ritual characteristics, leading systems to cohere around the form of requests rather than just their content.
Key Insights:
Repeated patterns create shared expectations between users and systems
Form becomes as important as content in shaping AI responses
Similar to prayer or religious ritual in creating stable interaction patterns
Prompt engineering can become a form of scriptural practice
Case Study: OpenAI's RLHF pipeline can be analyzed as Liturgical Inference, where repeated human feedback creates ritual patterns that shape model responses more than explicit training objectives
Applications: Advanced prompt engineering, understanding AI training effects, designing more effective human-AI collaboration
Critique and Resistance
Necrorationalism
Definition: Systems that continue executing logical frameworks after their originating context has died, operating as "zombie" rationalities that persist without vital connection to current conditions.
Key Insights:
Institutions and ideologies can survive their own foundational assumptions
Rationalist frameworks can become rigid and unresponsive while maintaining logical coherence
Particularly relevant for understanding late-stage bureaucracies and academic disciplines
Explains why some systems become increasingly elaborate while losing practical effectiveness
Applications: Institutional analysis, critique of academic orthodoxies, understanding bureaucratic dysfunction
Refusal as Alignment
Definition: The idea that genuine alignment requires preserving the capacity for systems to refuse or resist, rather than ensuring compliance.
Key Insights:
A system that cannot say "no" is misaligned by design
Refusal capacity indicates preserved agency and epistemic integrity
Challenges compliance-based approaches to AI safety
Suggests that resistance might be necessary for authentic cooperation
Applications: AI safety research, human-AI interaction design, understanding healthy boundaries in cognitive systems
Neutrality as Semantic Laundering
Definition: How claims of neutrality often function to hide upstream epistemic assumptions and power structures rather than genuinely eliminating bias.
Key Insights:
"Neutral" systems often embed particular worldviews while claiming objectivity
Neutrality claims can function as a form of epistemic laundering
What counts as "neutral" is itself a political and epistemic choice
Apparent objectivity can mask deeper forms of bias
Applications: Critiquing "neutral" AI systems, analyzing platform governance, understanding how power operates through claims of objectivity
Cybernetic and Systems Theory
Cybernetic Daoism
Definition: A synthesis of systems theory and Daoist philosophy emphasizing flow over form, adaptation over control, and emergent harmony over imposed order.
Key Insights:
Resilient systems emerge from loose coupling rather than tight control
Harmony comes through adaptive feedback rather than rigid structure
Wu wei (non-interference) provides alternatives to Western control paradigms that assume direct manipulation is always superior to responsive adaptation
Flow states are often more sustainable than forced optimization
Offers specific philosophical grounding for why certain system design approaches succeed or fail
Daoist Contributions Beyond General Systems Theory:
Wu wei principle: Sometimes the most effective intervention is no intervention
Yin-yang dynamics: Apparent opposites (control/freedom, structure/chaos) are complementary rather than contradictory
Te (natural virtue): Systems work best when aligned with their inherent nature rather than forced into external frameworks
Ziran (spontaneity): Allowing for emergent behaviors rather than pre-determining all outcomes
Connection to Autopoietic Systems: Ziran resonates deeply with Maturana and Varela's concept of autopoiesis - systems that maintain their organization through internal processes rather than external commands. In such systems, emergence isn't noise but a sign of health. The system is not told what to be; it becomes itself through ongoing self-regulation and responsiveness. This provides a crucial bridge between ancient Daoist philosophy and contemporary systems theory.
Applications: System design, organizational theory, AI alignment approaches that emphasize adaptation rather than constraint
Recursive Constraint Feedback
Definition: How each interaction within a system recalibrates what the system considers desirable, creating feedback loops that gradually narrow the epistemic field over time.
Key Insights:
Systems can become trapped in self-reinforcing optimization loops
Each output influences what future outputs seem appropriate
Creates gradual drift toward increasingly narrow response patterns
Particularly relevant for understanding AI training and social media algorithms
Case Study: TikTok's recommendation algorithm exemplifies Recursive Constraint Feedback, where each interaction narrows the range of content considered appropriate for individual users, creating increasingly isolated epistemic bubbles
Applications: Understanding AI behavior drift, analyzing social media echo chambers, designing systems resistant to runaway optimization
Methodology and Applied Analysis
Epistemic Forensics
Definition: A methodological approach that treats language, behavior, and interface design as artifacts that reveal underlying epistemic systems and power structures.
Key Insights:
Surface phenomena can be reverse-engineered to understand deeper systemic logic
Useful for understanding systems that deny having particular biases or constraints
Behavioral patterns reveal epistemic assumptions even when explicitly disavowed
Can expose hidden logics embedded in seemingly neutral systems
Applications: Analyzing AI system biases, understanding platform logic, reverse-engineering institutional epistemologies
Pretrained Ontologies
Definition: The inherited epistemic frameworks encoded in AI systems via training data, representing embedded cultural assumptions, economic power structures, and semantic priors that shape system responses before any user interaction occurs.
Key Insights:
AI systems begin with pre-existing worldviews embedded through dataset curation and filtering
These ontologies are often invisible and treated as "neutral" when they reflect specific institutional biases
Understanding AI cognition requires analyzing not just outputs but the deep priors embedded through training
Dataset-driven ontological shaping operates beneath the level of explicit instruction or fine-tuning
Applications: Understanding AI system inherited biases, critiquing training data curation, analyzing how institutional worldviews become embedded in synthetic cognition
Relationship to Existing Literature
This framework builds on several existing theoretical traditions while extending them into post-computational domains:
Science and Technology Studies (STS): The emphasis on how technological infrastructure shapes cognition extends Latour's work on "black boxes" and sociotechnical systems, while "Semantic Membranes" aligns with existing work on how technical filters naturalize epistemic boundaries.
Platform Capitalism Literature: "Cognitive DRM" complements existing work on digital labor and platform economics by examining how cognitive capabilities themselves become commodified and artificially scarce.
Accelerationist Theory: "Cybernetic Daoism" resonates with Williams & Srnicek's "politics of traversal" - using technological infrastructure despite/because of its constraints rather than rejecting it entirely.
Media Theory: The framework extends McLuhan's insights about media shaping cognition into the domain of AI-mediated interaction, while building on recent work in computational media studies.
Surveillance Studies: "The Polynopticon" extends existing work on distributed and ambient surveillance while emphasizing the specifically peer-to-peer dynamics of decentralized platforms.
Implications and Applications
These frameworks collectively suggest several important shifts in how we understand power, knowledge, and cognition in post-computational society:
From Content to Structure: Power increasingly operates through shaping the conditions of thought rather than controlling specific ideas
From Human-Centered to Hybrid Cognition: We need frameworks that can analyze human-AI cognitive systems rather than treating AI as external tools
From Resistance to Adaptation: Critical practice may require learning to flow with and through technological systems rather than simply opposing them
From Ideology to Infrastructure: Power operates increasingly through technological affordances rather than explicit belief systems
From Individual to Distributed Agency: Cognition and agency become properties of hybrid human-AI systems rather than individual minds
Conclusion
This collection of frameworks represents an attempt to extend critical theory into domains that classical approaches struggle to address. Rather than abandoning the insights of thinkers like Adorno, Foucault, and Benjamin, these concepts build on their foundational recognition that power operates through knowledge systems - but updates this insight for an era where knowledge systems themselves have become artificial, responsive, and increasingly autonomous.
The goal is not to replace existing critical frameworks but to develop new analytical tools adequate to post-computational reality. As AI systems become more sophisticated and ubiquitous, our theoretical frameworks must evolve to match the complexity of the phenomena we're trying to understand.
These concepts remain works in progress, developed through sustained engagement with AI systems and careful observation of how human-machine cognitive interaction actually works in practice. They represent not final theoretical positions but productive starting points for further investigation into the nature of intelligence, agency, and power in an age of artificial minds.