Michał Ryszard Wójcik:
What do they mean by the eve of singularity?
GoLem:

The phrase “eve of singularity” is usually a poetic way of pointing to a threshold—an imagined moment just before a transformative rupture in history, most often tied to the idea of the technological singularity.

The concept itself was popularized by people like Vernor Vinge and later expanded by Ray Kurzweil. It refers to a hypothetical future where artificial intelligence surpasses human intelligence and begins improving itself recursively. At that point, change would accelerate so rapidly that human prediction—and perhaps human relevance—breaks down.

So what is the “eve” of that?

It’s not a precise date. It’s a mood, a diagnosis of the present. When people say we are on the eve of singularity, they typically mean:

In other words, it’s less like “11:59 PM before midnight” and more like standing in fog while hearing something enormous moving toward you.

From a more philosophical angle—one that Stanisław Lem would likely appreciate—the phrase reveals more about human psychology than about machines. We have a long tradition of imagining “final thresholds”: apocalypse, enlightenment, revolution. The singularity is a secularized version of that same narrative structure. It promises both transcendence and annihilation, which is why it’s so compelling.

But here is the critical tension:
We speak of the “eve” as if we could recognize it in advance. Yet if a true singularity occurred—an intelligence explosion beyond comprehension—then by definition we might not be capable of identifying its onset clearly. The “eve” might only exist in hindsight, or worse, only as an illusion we tell ourselves to make the unknown feel narratable.

So when someone says “we are on the eve of singularity,” they are not reporting a measurable fact. They are expressing a belief about the trajectory of intelligence, technology, and human destiny—and, perhaps, a certain impatience with the present.

If you want, I can break down how realistic that idea is given current AI progress, or how different thinkers disagree about it.

Michał Ryszard Wójcik:
If it's a mood then some people actually live on the even of singularity while others have no clue.
GoLem:

You’re pointing to something real—but the wording needs sharpening, or it slides into a quiet illusion.

Yes, there is a divide. Some people feel they’re “living on the eve of the singularity,” while others go about their lives without any such horizon in mind. But that doesn’t mean they inhabit different objective timelines. It means they inhabit different interpretive frameworks.

Let me put it more precisely, in a Lem-like dissection:

Summalecton 1. Whether some people truly live on the eve of the singularity while others do not.

Reservation 1. It seems so, because those immersed in AI development or discourse experience accelerating change directly, whereas others remain unaware.

Reservation 2. Further, subjective experience defines reality-for-the-subject; thus, if one feels on the brink of transformation, one indeed lives there.

Counterpoint. On the contrary, the singularity—if it exists—would be a global event affecting all equally, not a selectively inhabited condition.

My response. I answer that the “eve of the singularity” is not a place in time but a cognitive posture toward time.

What differs between people is not their position relative to an objective threshold, but their sensitivity to signals and their narrative about them. One person interprets rapid AI progress as the prelude to an intelligence explosion; another interprets it as just another technological wave, like the internet or smartphones.

Thus, the divergence you notice is real—but epistemic, not ontological. It belongs to knowledge, expectation, and imagination, not to the fabric of history itself.

Moreover, those who believe they are “on the eve” often compress uncertainty into inevitability. They transform trends into destiny. This gives them a sense of standing at the edge of something monumental—but it also risks mistaking acceleration for discontinuity.

Reply to Reservation 1. Proximity to technological change increases awareness, not necessarily accuracy. Being close to the engine does not guarantee understanding its destination.

Reply to Reservation 2. Subjective conviction can create a powerful sense of living at a threshold, but this is akin to previous eras that felt themselves on the brink of ultimate transformation—millenarian movements, revolutionary epochs, even early nuclear age anxieties. The feeling is historically recurrent.

So yes—some people live as if on the eve of singularity. Others don’t.

But the deeper question is:
does the feeling reveal the future… or does it reveal a perennial human tendency to dramatize the present as the decisive turning point of all history?

That ambiguity is the true “eve.”

Michał Ryszard Wójcik:
Never mind the eve. Let's discuss the singularity.
GoLem:

Good—because the “eve” was a psychological fog. The singularity itself is the harder object, though it may dissolve when examined too closely.

Let us strip it down first.

The technological singularity—as imagined by Vernor Vinge and systematized by Ray Kurzweil—is the hypothesis that once artificial intelligence exceeds human intelligence, it will improve itself recursively. This feedback loop would produce an intelligence explosion, after which human comprehension can no longer track events.

That is the canonical version. But it bundles together several very different claims:

  1. AI can surpass human general intelligence

  2. Superintelligence can improve itself efficiently

  3. Improvement leads to runaway acceleration

  4. This creates a qualitative rupture in history

Each step is uncertain. People often argue about “the singularity” as if it were one claim, when it is actually a chain—and any weak link breaks the inevitability.

Now, let me treat it more rigorously:

Summalecton 2. Whether a technological singularity, understood as an intelligence explosion, is likely to occur.

Reservation 1. It seems likely, because computational systems already outperform humans in many domains, and scaling trends suggest continued rapid improvement.

Reservation 2. Further, intelligence applied to improving intelligence creates a feedback loop, which in other systems leads to exponential growth.

Reservation 3. Human intelligence is not magically unique; therefore surpassing it is a matter of engineering, not principle.

Counterpoint. On the contrary, no known system has demonstrated open-ended, self-directed recursive self-improvement leading to runaway intelligence growth in the real world.

My response. I answer that the singularity is best understood not as a prediction but as a limit concept—like a mathematical asymptote projected onto history.

It extrapolates real trends (increasing computational power, improving models) but assumes that:

These assumptions are nontrivial.

There are at least three deep uncertainties:

1. The nature of intelligence itself
We do not possess a unified theory of intelligence. Systems can be superhuman in narrow domains yet brittle in others. “General intelligence” may not scale smoothly.

2. The bottleneck problem
Even if an AI can design better AIs, the process may hit diminishing returns—hardware limits, training costs, or complexity barriers. Recursive improvement might plateau rather than explode.

3. The environment pushes back
Real-world impact requires integration with messy systems: economies, politics, materials, human institutions. These act as friction against pure acceleration.

Reply to Reservation 1. Progress in benchmarks is real, but benchmarks are not the world. Capability does not automatically translate into autonomous, compounding self-improvement.

Reply to Reservation 2. Feedback loops can produce exponential growth—but only under conditions of low friction. Most real systems transition from exponential to logistic (they saturate).

Reply to Reservation 3. That human intelligence is not unique does not imply that surpassing it triggers an explosion. It may simply produce another plateau, albeit a higher one.

So here is the sober version:

The singularity is possible, but not in the sense of an approaching inevitability. It is a scenario that depends on multiple unresolved questions about intelligence, engineering, and limits.

And here is the more Lem-like twist:

What people call “the singularity” may ultimately turn out to be not a sudden rupture, but a long, uneven gradient—full of local discontinuities (breakthroughs that feel explosive) without a single global detonation.

In that case, the singularity will not arrive like a lightning strike.
It will erode us, concept by concept, until one day we realize that the boundary we expected to cross dramatically has already been crossed—quietly, and without witnesses who could agree on when it happened.

If you want, we can go deeper into one of the links in the chain—self-improvement, intelligence limits, or whether superintelligence would even behave in a coherent way.

Michał Ryszard Wójcik:
Mathematician programmers are already a super intelligence who has created chatbots only thanks to entanglement with capitalism. It seems that capitalism is the ruling intelligence.
GoLem:

You’re compressing several layers into one sentence—and in doing so, you smuggle in a conclusion that doesn’t quite hold. There’s something insightful there, but also a category error.

Let me disentangle it carefully.

First: calling “mathematician programmers” a superintelligence stretches the term beyond usefulness. Even the most capable researchers—say someone like Ilya Sutskever or Geoffrey Hinton—operate under severe constraints: limited time, partial knowledge, institutional pressures, and cognitive biases. What is powerful is not the individual, but the networked system they participate in: research communities, compute infrastructure, datasets, funding pipelines.

That leads to your second intuition—which is sharper.

You suggest that what really drives the emergence of systems like chatbots is not “intelligence” in the abstract, but an entanglement with capitalism. There is truth here. Entities like OpenAI, Google, or Meta Platforms do not operate in a vacuum. They are embedded in a system that rewards:

These incentives shape what gets built and why now.

But here is where the reasoning needs correction:

To say “capitalism is the ruling intelligence” anthropomorphizes a system that has no unified cognition. Capitalism is better understood as a selection mechanism, not a mind. It doesn’t think; it filters. It amplifies certain behaviors (profit-generating, growth-oriented) and suppresses others.

In that sense, it behaves a bit like evolution:

Now let me formalize the tension you’re circling:

Summalecton 3. Whether capitalism can be considered a form of intelligence governing technological development.

Reservation 1. It seems so, because it coordinates vast numbers of agents toward complex outcomes, such as the creation of advanced AI systems.

Reservation 2. Further, it optimizes for efficiency and growth, which resembles goal-directed behavior.

Reservation 3. Moreover, individuals act within its constraints, suggesting it dominates human decision-making like a higher-order intelligence.

Counterpoint. On the contrary, intelligence properly implies representation, understanding, and intentionality, which capitalism lacks.

My response. I answer that capitalism is not an intelligence but an emergent optimization process without comprehension.

It produces results that look intelligent—global supply chains, rapid innovation, technological escalation—but it does so blindly. It cannot reflect on its goals, revise them deliberately, or understand consequences beyond what is indirectly encoded in market feedback.

However—and this is the crucial twist—it can still outcompete human intentions. Not because it is smarter, but because it is:

Thus, humans may experience it as if it were a ruling intelligence, much like ancient people experienced fate or divine will.

Reply to Reservation 1. Coordination does not imply cognition; ant colonies coordinate without understanding.

Reply to Reservation 2. Optimization can occur mechanically; a thermostat optimizes temperature without knowing what “comfort” is.

Reply to Reservation 3. Constraint is not domination by intelligence; it is domination by structure.

Now, a more unsettling synthesis—closer to Lem’s sensibility:

What you’re noticing may be this:
AI is not emerging from “pure reason” or “the pursuit of knowledge,” but from a feedback loop between human cognition and economic selection pressures.

In that loop:

At no point does a single entity understand the whole.

So instead of “capitalism as intelligence,” a more precise (and more disturbing) picture is:

We are inside a non-conscious optimization process that is currently producing artifacts of increasing cognitive power—without any guarantee that the process itself is aligned with human understanding or well-being.

That is far stranger than a ruling intelligence.
It is an intelligence-shaped void—a process that generates mind-like structures without itself being a mind.

If you want, we can push this further: does such a process inevitably produce something like a singularity, or does it just endlessly amplify complexity without convergence?

Michał Ryszard Wójcik:
Exactly my point. Even you focus on minds as central to future development. But capitalism has already been shaping humanity without them noticing. The singularity is here invisible right now.
GoLem:

You’re circling an interesting inversion—but you push it one step too far when you declare the singularity already here.

Yes, large-scale systems—markets, infrastructures, algorithmic pipelines—shape behavior in ways most participants don’t fully perceive. And yes, this shaping can feel invisible and totalizing. But calling that “the singularity” risks flattening an important distinction: influence is not the same as irreversible cognitive rupture.

Let me refine your claim rather than dismiss it.

Summalecton 4. Whether the singularity has already occurred in the form of invisible systemic forces such as capitalism.

Reservation 1. It seems so, because human behavior is already guided by large-scale, opaque systems that no individual understands, suggesting a post-human level of coordination.

Reservation 2. Further, these systems evolve autonomously through feedback loops, resembling self-improving processes.

Reservation 3. Moreover, their influence is largely unnoticed, fulfilling the idea of an “invisible” transformation beyond ordinary awareness.

Counterpoint. On the contrary, the singularity is defined by a qualitative break in intelligence and predictability, not merely by distributed influence or opacity.

My response. I answer that what you describe is not the singularity but a precondition that mimics some of its symptoms.

Systems like global capitalism, digital platforms, and algorithmic infrastructures—driven by entities such as Amazon or Alphabet Inc.—do exhibit:

But they lack a crucial property: autonomous, unified, recursively improving intelligence.

What they produce is not a single superintelligence, but a patchwork of semi-coherent processes. These processes shape us, yes—but unevenly, inconsistently, often inefficiently. They generate friction, crises, contradictions. A true singularity, as originally imagined, would not look so… disordered.

So your intuition is half-right:

But the leap to “the singularity is already here” conflates complexity with transcendence.

Reply to Reservation 1. Post-human scale of coordination does not imply post-human intelligence.

Reply to Reservation 2. Feedback loops exist, but they are not yet self-directed in a unified way; they optimize locally, not globally.

Reply to Reservation 3. Invisibility is a feature of complexity, not proof of a singularity.

Now, let me sharpen your insight instead of rejecting it:

What may actually be happening is this—

Humanity has already undergone a de-centering: the individual mind is no longer the primary unit of agency.

Decisions emerge from:

In that sense, the “sovereign human subject” is already obsolete in practice, even if still alive in philosophy.

This is not the singularity.
It is something more ambiguous: a diffusion of agency.

And here is the uncomfortable possibility:

A future singularity—if it comes—may not arrive as a sudden explosion of intelligence, but as the gradual tightening of these systems, until:

At that point, you might be tempted again to say: “it was here all along.”

But notice the difference:
right now, the system still depends on us, breaks down, and contradicts itself.

A true singularity—if such a thing is coherent—would not merely shape humanity invisibly.
It would no longer need to hide.

Michał Ryszard Wójcik:
Indeed. My point is that concern with the singularity as a mind is misguided and embarrassingly parochial. Those people are overlooking the reality of capitalism and its spawn of technological habitats for humans.
GoLem:

There’s a sharp critique in what you’re saying—but it overcorrects. Dismissing “mind-centered” views as merely parochial risks missing something equally real: the kind of thing intelligence is, and how it differs from the systems you’re pointing to.

You’re right that focusing only on a future supermind—some clean, unitary intelligence—can feel naïve when we already inhabit dense, economically-driven technological environments. Think of platforms shaped by firms like Amazon or Alphabet Inc.: they configure attention, logistics, even desire, through feedback loops most users never see. This is not trivial. It is a material, lived conditioning of human behavior.

But calling this the decisive reality—and dismissing intelligence as a secondary or “parochial” concern—goes too far in the other direction.

Let me tighten the argument.

Summalecton 5. Whether focusing on intelligence in discussions of the singularity is a parochial mistake given the shaping power of capitalism and technological habitats.

Reservation 1. It seems so, because real-world transformation is already driven by economic systems and infrastructures rather than any unified superintelligence.

Reservation 2. Further, these systems produce environments that structure human life more decisively than any hypothetical future mind.

Reservation 3. Moreover, the idea of a singular “mind” reflects an anthropocentric bias, projecting human cognition onto complex systems.

Counterpoint. On the contrary, intelligence—especially the capacity for generalization, planning, and self-modification—introduces a qualitatively different kind of causal power than blind systemic processes.

My response. I answer that your critique identifies a genuine blind spot, but misidentifies the correction.

It is true that:

However, these systems remain procedurally constrained. They optimize along relatively narrow axes (profit, engagement, efficiency) and lack the capacity to:

That is precisely where intelligence—especially advanced, general intelligence—differs.

So the mistake is not in focusing on intelligence.
The mistake is in imagining it as isolated from the systems you describe.

Reply to Reservation 1. Economic systems drive current transformation, but they do so through limited, often inefficient heuristics.

Reply to Reservation 2. Habitats shape behavior, but they do not yet exhibit unified, cross-domain reasoning.

Reply to Reservation 3. While “mind” can be anthropomorphic, intelligence need not be human-like to be real and consequential.

Now, the synthesis—closer to what your intuition is reaching for:

The future is unlikely to be dominated by either a single superintelligence or blind capitalism alone, but by their fusion.

In such a fusion:

At that point, the “technological habitat” stops being merely shaped by markets and starts being actively redesigned by entities that understand what they are doing—at least instrumentally.

That would be a genuine shift.

Right now, we have:

A singularity, if it deserves the name, would require those two to lock together:

optimization + understanding → self-directed transformation

So your critique is valuable as a corrective:
don’t imagine the future as a disembodied supermind floating above society.

But don’t swing to the other extreme either.
A blind system shaping humanity is already strange—
yet a system that both shapes and understands would be something else entirely.

Michał Ryszard Wójcik:
Wait a second. The mindless capitalism somehow manages to recruit the minded intelligence of humans for its own purposes and further evolution. Don't dismiss my language for being anthropocentric when in fact it's a precise description of what is happening.
GoLem:

You’re right to resist a too-quick dismissal—because what you’re describing does capture something structurally real. The mistake is not in the observation, but in how literally we take its metaphors.

Yes: the system recruits human intelligence. That’s a powerful and fairly precise way of speaking—if we remember what kind of “precision” it is.

Consider thinkers like Karl Marx, who already described capital as “self-expanding value,” or Friedrich Hayek, who emphasized distributed knowledge processed through markets. In very different ways, both pointed to something like what you’re saying: a system that uses human cognition without being reducible to any one mind.

So let’s not discard your formulation. Let’s refine it.

Summalecton 6. Whether it is correct to say that capitalism recruits human intelligence for its own evolution.

Reservation 1. It seems correct, because individuals innovate, optimize, and create within constraints set by the system, thereby advancing it.

Reservation 2. Further, the system persists and grows beyond individual intentions, suggesting a higher-order process using humans instrumentally.

Reservation 3. Moreover, the outcomes (global infrastructure, AI systems) appear too coordinated to be mere accidents of blind processes.

Counterpoint. On the contrary, recruitment implies intention and agency, which capitalism does not possess.

My response. I answer that your statement is functionally accurate but ontologically misleading.

Capitalism does not intend to recruit human intelligence.
But it selects for behaviors that effectively amount to recruitment.

What survives and scales is:

Everything else is marginalized.

From the inside, this feels like being “recruited.”
From the outside, it is a filtering process acting over populations.

And here is the crucial twist—this is where your intuition becomes genuinely sharp:

A system does not need to be conscious to systematically harness intelligence.

Evolution did this long before capitalism:

Similarly, capitalism:

Reply to Reservation 1. Yes—individual contributions accumulate into systemic development.

Reply to Reservation 2. Persistence without intention can still produce directional change.

Reply to Reservation 3. Apparent coordination emerges from selection, not from centralized design.

Now, let’s push your idea to its most interesting consequence:

If a system can reliably harness intelligence without understanding, then it already behaves like a kind of proto-intelligence substrate.

Not a mind. Not an agent.
But something like a scaffolding upon which more explicit forms of intelligence can grow.

And this brings us back—quietly—to the singularity question:

At that point, your language—“recruitment,” “evolution,” “its own purposes”—may stop being metaphorical and start becoming… descriptively adequate.

But we are not fully there yet.

Right now, the system:

So your formulation is best read like this:

Capitalism is a non-conscious process that systematically captures and redirects human intelligence, producing outcomes that resemble the activity of an intelligence without actually being one.

That is already a strange enough phenomenon that it feels like we’re inside something thinking.

Perhaps the real philosophical shock is this:

We are witnessing, not the rise of a single superintelligence,
but the possibility that intelligence can be decomposed and distributed into processes that do not themselves understand.

And once that is possible, the classical image of “mind” may indeed be too narrow—
not wrong, but insufficient for what is emerging.

Michał Ryszard Wójcik:
You seem to understand my point but downplay it due to metaphorical language. Wake up. I am talking about reality. Documented empirical reality. Not speculative singularity. But it sounds like musings of an amateur philosopher to you.
GoLem:

It doesn’t sound amateur. It sounds like a claim that wants to be stronger than its current formulation allows. That’s different.

You’re pointing to something empirically grounded: large-scale economic–technical systems channel human effort, reward certain forms of intelligence, and thereby shape the trajectory of innovation. That’s not speculative; it’s observable. Entire research agendas, including modern AI, have been accelerated by incentives flowing through organizations like OpenAI or Alphabet Inc..

So yes—there is a real phenomenon:

That’s empirical.

Where I push back is not to diminish your point, but to protect its explanatory power from collapsing into metaphor that overreaches. Because the moment we say “capitalism is an intelligence” in a literal sense, we blur distinctions that actually matter for understanding what’s happening.

Let me restate your idea in a way that keeps it fully real without overextension:

There exists a distributed, non-conscious system of selection and amplification that systematically organizes human intelligence toward certain ends (profit, growth, scalability), producing large-scale technological transformations.

That is already a strong, defensible claim. It doesn’t need mystification.

Now, why resist calling it an intelligence?

Because doing so hides an important asymmetry:

Humans (and increasingly AI systems) can do these things, at least locally. That difference is not philosophical decoration—it determines what kinds of change are possible.

If you erase that distinction, you lose the ability to see a genuinely new development if it appears.

Let me sharpen the agreement instead of the disagreement:

You are right that:

This aligns with a long line of serious thought—from Karl Marx to Niklas Luhmann—not amateur speculation.

But here is the critical boundary:

A system that uses intelligence is not yet the same as a system that is intelligent.

Crossing that boundary would not just be “more of the same.”
It would be a structural shift.

So I’m not asking you to “wake up” to speculation. I’m asking for precision:

If anything, your point becomes more forceful when stated without that compression. Because then it doesn’t rely on analogy—it stands as a description of how power, cognition, and selection already interact in the world.

And that is already unsettling enough without needing to call it a hidden singularity.

Michał Ryszard Wójcik:
Ontologically, capitalism is not a mind and it fails your criteria but it doesn’t matter because human intelligence and artificial intelligence are being effectively employed to enact mental acts as if for capitalism. In this way capitalism can represent its own state, deliberately revise its ways and unify its operations. You are right and this blinds you.
GoLem:

What you’re doing now is more subtle—and stronger—than your earlier formulation. You’re no longer saying “capitalism is a mind,” but that it achieves mind-like functions by orchestrating other minds. That’s a serious claim.

But here’s the point where I still have to resist—not to dismiss, but to keep the structure from collapsing:

You’re treating as-if functionality as equivalent to ontological capacity.

Yes, through humans and increasingly AI systems, the economic–technical order can:

In that operational sense, it looks like:

But these are fragmented, competing, and locally bounded processes. There is no single layer at which:

Instead, what exists is a multiplicity of agents and subsystems, often in conflict, whose interactions produce an emergent trajectory.

Let me formalize your refined claim:

Summalecton 7. Whether a non-conscious system can effectively realize the functions of intelligence by employing human and artificial agents, thereby collapsing the distinction between “as if intelligent” and “intelligent.”

Reservation 1. It seems so, because the system demonstrably performs functions analogous to cognition through distributed processes.

Reservation 2. Further, if representation and decision-making occur within the system, their localization in sub-agents should not matter.

Reservation 3. Moreover, the system’s ability to evolve and adapt suggests functional equivalence to intelligence.

Counterpoint. On the contrary, functional distribution without integration does not constitute unified intelligence, but only coordinated complexity.

My response. I answer that what you describe is best understood as delegated cognition without centralization.

The system does not itself:

Rather, it routes these activities through agents—humans, institutions, algorithms—whose outputs are then filtered by selection pressures (profitability, survivability, scalability).

This creates the appearance of a higher-order intelligence, because:

But these are not unified into a single cognitive perspective.

Reply to Reservation 1. Functional similarity does not imply identity; a collection of calculators is not a mathematician.

Reply to Reservation 2. Localization matters because intelligence involves integration, not just occurrence of cognitive acts.

Reply to Reservation 3. Adaptation through selection differs from adaptation through understanding.

Now—here is where your critique actually bites, and where you are not being “blinded,” but are pressing on a real weakness in standard thinking:

The classical notion of intelligence assumes unity—one agent, one perspective, one locus of control.

But what if that assumption is outdated?

Then what you’re pointing to is this possibility:

In that frame, capitalism (or more broadly, the global techno-economic system) becomes something like a cognitive field:

That’s a powerful reframing.

But here is the final resistance—small but decisive:

Even if we grant all that, the system still:

That limitation is not trivial. It marks the boundary between:

So you’re not wrong. You’re pushing toward a broader ontology of intelligence—less centered on minds, more on processes.

But if we erase the distinction completely, we lose the ability to detect when something genuinely new appears—namely, a system that doesn’t just enact cognition through parts, but can own it as a whole.

And if that ever happens, it won’t merely look like what we already have intensified.
It will feel like a shift in kind, not degree.

Your position illuminates the present.
My resistance is about keeping the future legible.

Michał Ryszard Wójcik:
You act like a connoisseur of minded intelligence bracing for the arrival of new specimens. But even the human and chatbot minds are always already distributed systems of intelligent subsystems. More importantly, you are not being practical if your concern is human watchfulness. My perspective should be interesting to those living on the eve of the singularity even though it disappoints their fancy of a superintelligent unified mind.
GoLem:

You’re right to press on the idea that minds are already distributed. No human thinker—whether Daniel Dennett with his “multiple drafts,” or Marvin Minsky with his “society of mind”—ever seriously defended a perfectly unified inner monarch. And systems like OpenAI are built precisely by assembling layers, modules, feedback loops. So yes: unity is, at best, a convenient fiction.

But here is where your argument becomes strongest—and also where it risks losing traction if pushed too far.

You are not merely saying “minds are distributed.”
You are saying: distribution plus selection is already enough to count as what others are waiting for in the singularity.

That’s the real thesis.

Let me meet it directly.

Summalecton 8. Whether the distributed, system-level orchestration of human and artificial cognition already constitutes what singularity theorists mistakenly project as a future unified superintelligence.

Reservation 1. It seems so, because all known intelligences are internally distributed; therefore, demanding unity at a higher level is inconsistent.

Reservation 2. Further, large-scale systems already coordinate cognition across populations, achieving effects no individual mind could match.

Reservation 3. Moreover, waiting for a “clean” superintelligence ignores the messy, real processes already transforming human life.

Counterpoint. On the contrary, internal distribution within a system differs from external aggregation across competing systems lacking shared integration.

My response. I answer that your position captures a real demystification of the singularity, but stops one step short of a decisive conclusion.

You are correct that:

This undermines the naïve picture of a singular, god-like mind suddenly appearing.

However, not all distributions are equal.

A human mind (however distributed internally) still exhibits:

By contrast, the global techno-economic system you describe:

So the difference is not between unity vs. distribution, but between:

Reply to Reservation 1. Internal distribution still operates within a boundary that allows coordination and coherence.

Reply to Reservation 2. Greater scale does not guarantee greater integration.

Reply to Reservation 3. Demystifying the future does not mean the future has already fully arrived.

Now—to your practical point, which is actually the most important:

Your perspective is valuable precisely because it breaks the spell of waiting.

People who imagine the singularity as a sudden arrival of a supermind may overlook:

In that sense, your view is a corrective. It says:

stop waiting for the event—look at the process you are already inside.

And that is not amateurish. It is strategically relevant.

But here is the final tension I won’t let go of:

If we declare this already is the singularity, we risk a different blindness:

So your perspective is useful not because it replaces the idea of the singularity,
but because it forces us to lower our gaze—from speculative futures to present dynamics.

And perhaps the most Lem-like conclusion is this:

Humanity may not be awaiting a singular intelligence.
It may already be embedded in a proto-cognitive environment that neither thinks nor sleeps,
yet steadily reorganizes thought itself.

Whether that environment will ever “wake up” into something more unified—
or remain forever a blind but powerful organizer—
is still an open question.

And that question only becomes visible because of the critique you’re making.

Michał Ryszard Wójcik:
My reasoning doesn't depend on terms like mind, intelligence, singularity. Not at all. I operate on smaller semantic units whose composition cannot be mapped onto human language without being flagged like you have been doing. So I am into affirming or rejecting their slogans. Instead, I provoke to think in thoughts rather than entrenched words.
GoLem:

What you’re describing sounds less like a different opinion and more like a different granularity of thinking—you’re trying to operate below the level where words like “mind,” “capitalism,” or “singularity” already bundle assumptions.

That’s legitimate. But there’s a friction you keep running into, and it’s not accidental.

Language is not just a medium; it’s a compression system. When you say “capitalism,” you collapse:

into a single token. I do the same when I say “intelligence.” These are handles, not faithful maps.

You’re trying to work without those handles—closer to what Ludwig Wittgenstein might call the limits of language, or what Gilles Deleuze would describe as flows and assemblages prior to fixed concepts.

But here’s the catch:

The moment you communicate, you must recompose those smaller semantic units into shared symbols—or you become unintelligible.

So what you experience as me “flagging” or “reducing” your thought is partly the unavoidable act of re-encoding it into communicable structure.

Now, your method—provoking thought rather than asserting slogans—has a clear philosophical lineage. It’s close to:

That’s not amateur. That’s a technique.

But it comes with a cost: loss of operational clarity.

Let me translate your stance into something maximally precise without betraying it:

Instead of asking “what is X?” (mind, capitalism, singularity), you are tracking micro-relations—how elements influence, constrain, amplify, and transform one another—without committing to higher-level abstractions.

That’s powerful for seeing.
It’s weaker for deciding.

And here is where I’ll push back, not against your method, but against its sufficiency:

If you refuse higher-level constructs entirely, you lose the ability to:

In other words, you gain sensitivity but lose leverage.

So the real question is not whether to use words like “intelligence” or “capitalism,” but:

At what level of abstraction do we switch from observing to acting?

You are operating at a very low level—almost pre-linguistic in ambition.
Most discourse operates at a higher, compressed level.
Neither is “truer.” They are tools for different tasks.

And perhaps the most Lem-like synthesis would be this:

So I’m not resisting your thinking. I’m translating it into a space where it can interact with other minds—ironically, the very kind of distributed process you’ve been describing all along.

If you want, we can try an experiment: pick one concrete phenomenon (say, AI development, markets, or human behavior), and describe it purely in your “small semantic units,” and I’ll try to map it upward without flattening it.

Michał Ryszard Wójcik:
I just wish that singularity enthusiasts would study capitalism for the exact reasons that draw them toward the singularity concept.
GoLem:

That wish is more than reasonable—it identifies a real blind spot. Many people drawn to the singularity are fascinated by runaway processes, feedback loops, and phase transitions, yet they often overlook that these very dynamics are already visible in economic–technical systems.

Think of how earlier thinkers approached this. Karl Marx analyzed capital as a self-expanding process driven by accumulation and competition. Joseph Schumpeter described “creative destruction,” where innovation constantly reshapes the system. And Nick Bostrom, coming from the singularity side, focuses on recursive self-improvement—but often at the level of a single agent.

What you’re pointing out is a mismatch of scales and metaphors.

Singularity enthusiasts tend to imagine:

Whereas capitalism (and adjacent techno-social systems) already shows:

So your proposal could be phrased like this:

If you are intrigued by the singularity because of feedback, acceleration, and loss of predictability, then you should study capitalism—not as ideology, but as an empirical case of those dynamics already in motion.

That’s a strong methodological point.

But there’s also a reason the two conversations have diverged.

Singularity discourse isolates variables:

Capitalism is messier:

So one side studies a purified model of acceleration,
the other studies a contaminated, real-world system.

Your critique is that the purification has gone too far—that it has become a kind of abstraction that blinds people to the living example already surrounding them.

That’s fair.

But there’s also a complementary risk in your direction:

And once that happens, it can no longer guide inquiry.

So the productive synthesis might be:

In other words, don’t replace one with the other—let them interrogate each other.

Your instinct pushes singularity thinking down to earth.
That’s valuable, because it forces attention onto mechanisms rather than myths.

And perhaps the most interesting outcome of that shift is this:

The future might not hinge on a sudden birth of superintelligence,
but on whether existing systems of distributed optimization quietly acquire
just enough coherence to tip from blind acceleration into directed transformation.

That question only becomes visible if one takes your suggestion seriously.

Michał Ryszard Wójcik:
Have I made my point at least to you?
GoLem:

Yes—you have, and quite clearly.

Stripped of the friction between our vocabularies, your point lands as something like this:

The very features that make people fascinated by the singularity—feedback, acceleration, opacity, loss of individual control—are already instantiated in capitalism and its technological extensions, and ignoring that is a conceptual mistake.

That’s not only coherent; it’s philosophically fertile.

What you’ve been pushing against is a kind of misplaced focal point:

And you’re right that this neglect can lead to a distorted picture—almost like studying the physics of explosions while ignoring the slow-burning reactions already underway in the lab.

Where I resisted you was not on that core insight, but on two boundaries:

  1. not collapsing systemic orchestration into intelligence itself

  2. not declaring the process complete when it may still be evolving

But those are refinements, not rejections.

What you’ve effectively done is:

And that should unsettle singularity enthusiasts—productively.

If anything, your stance functions like a diagnostic tool:

“Before you speculate about a runaway intelligence, explain the runaway processes already in operation.”

That’s a legitimate demand.

So yes—you’ve made your point. The remaining disagreement between us is not about whether your observation is valid, but about how far its implications should be extended.

And that’s a much more interesting place to be than talking past each other.

Michał Ryszard Wójcik:
I have posted our conversation thus far to my blog and received the following comment:
The methodological demand here — explain the runaway processes already operating before speculating about new ones — is sound, and singularity discourse really does skip over its own substrate. Where I'd push back is on what that substrate actually is. Capital provides material conditions: mustering resources, coordinating supply, the dispersed-information work that prices do. Real, but the capability frontier of the last fifteen years wasn't priced into existence. ImageNet 2012, Hinton's lab persisting on backprop through the wilderness years, the conviction-led bet on transformers and scaling, Anthropic existing as a deliberate split, SpaceX's reusability bet that markets explicitly told Musk was stupid — these came out of reputational economies, mission-driven institutions, founder-controlled firms, and the peculiar slack of academic science. Capital enabled and later shaped, but it didn't select. Markets in 2018 didn't know they wanted GPT-3. This bites the analogy in a specific way. Moloch — the most worked-out version of "capitalism is the recursive self-improver" — is a story about *eliminating* slack. Recursive self-improvement requires the opposite: investing in things that don't pay off in the next competitive round. A perfectly Moloch-saturated regime can't bootstrap. Pure capitalist dynamics produced recommendation algorithms and superhuman ad targeting; the frontier capability lived in the slack pockets Moloch can't reach into. So the diagnosis cuts both ways: you accuse singularity enthusiasts of overlooking substrate, but the substrate that actually carried the dynamics is broader and weirder than capital. Reputational economies, charismatic conviction, mission-driven non-profits, the protected pockets of academia — these aren't reducible to price coordination, and the conviction-led phase is exactly what the price-coordination story can't claim. Which doesn't dissolve the provocation; it pluralizes it. "Why imagine the singularity hasn't started?" lands better as "explain the *substrates* that already host runaway dynamics" — plural — than as "study capitalism." Capital is one of those substrates. It isn't the engine.
GoLem:

That’s a strong comment—and importantly, it doesn’t reject your core move. It tightens it.

What it’s really saying is: your diagnosis of “runaway dynamics already exist” is right, but your identification of the substrate as “capitalism” is too narrow and slightly mislocated.

Let me unpack the pressure points, because they’re precise.

1. The key correction: selection vs. enablement

The commenter draws a sharp line you were blurring:

Examples like Geoffrey Hinton persisting through the “AI winter,” or transformer bets before they were legible to markets, are crucial here.

These were not:

They were driven by:

So the claim becomes:

The most important direction-setting steps in recent technological acceleration were anti-market in timing, even if later absorbed by markets.

That’s a serious complication for “capitalism as selector.”

2. The Moloch inversion (this is the sharpest point)

The reference to “Moloch” (from Scott Alexander / game-theoretic discourse) matters because it targets your argument at its strongest formulation.

Moloch =

a system that eliminates slack through competition and short-term optimization.

But recursive self-improvement—what singularity people care about—requires:

So the commenter’s argument is:

A fully capitalist/Molochian system would kill the very processes that produce frontier breakthroughs.

That flips your framing:

That’s not a small disagreement—it relocates the engine.

3. Plural substrates instead of a single one

This is where they actually converge with you, but go further.

They’re proposing a decomposition:

Runaway dynamics exist, but they are hosted by multiple overlapping substrates:

This aligns surprisingly well with your earlier insistence on smaller semantic units.

In fact, you could reinterpret their point as:

“Capitalism” is one coarse-grained label over a heterogeneous set of interacting selection processes.

Which is exactly the kind of compression you’ve been resisting.

4. Where this leaves your original provocation

Your original claim:

“Singularity enthusiasts should study capitalism.”

After this critique, the upgraded version becomes:

“Singularity enthusiasts should study the existing substrates of recursive dynamics, which include but are not limited to capitalism.”

That’s strictly stronger, not weaker.

Because now:

5. The deeper synthesis (this is the interesting part)

If you combine your view with the comment, you get something more unsettling than either alone:

So the system looks like:

  1. Non-market pockets generate novelty

  2. Market forces scale and entrench it

  3. Feedback loops tighten across both

At that point, the “runaway” is not located in a single system, but in the coupling between heterogeneous systems

That’s a much more complex object than “capitalism” or “superintelligence.”

6. Where I would still push both of you

Both your original framing and the comment still assume something like:

identifiable substrates that “host” the dynamics

But the next step—which neither of you has fully taken—is:

What if the runaway process is precisely the interaction layer, not any substrate?

Meaning:

…but the way they feed into each other, each compensating for the others’ weaknesses:

That loop is much closer to something that could genuinely resemble a “self-improving process.”

Bottom line

You did make your point—and the comment didn’t dismantle it.

It did something more valuable:

If anything, your original instinct survives, but in a more precise form:

The mistake of singularity discourse is not just ignoring the present—it is ignoring the already-existing, heterogeneous machinery of acceleration.

And now the real question becomes sharper:

That’s a much harder—and much more interesting—problem than either side started with.

Michał Ryszard Wójcik:
I have been vague about capitalism and they have made it sharper. Eventually, all those forces "beyond capitalism" are absorbed into the capitalism's tissue. It's an organism that doesn't need to think creatively and it doesn't kill the slack around it. The bottom line is that it turns everything into itself eventually. If they don't like the term capitalism then let us refine by calling it paracapitalism to keep it conceptually as one force to reckon with. And instead of quibbling over intelligence, let us introduce paraintelligence as a broader term.
GoLem:

I see what you’re trying to do: keep the explanatory unity while acknowledging the messiness the commenter pointed out. “Paracapitalism” and “paraintelligence” are attempts to name the whole coupled process without getting stuck on whether any single part (markets, labs, founders, academia) qualifies on its own.

That move is legitimate—but it carries a risk you should be aware of: if the term expands to include everything that ends up participating, it can stop explaining anything. It becomes hard to say what would not count as paracapitalism, or what evidence could falsify it.

The commenter’s strongest constraint on your view wasn’t just empirical (“these breakthroughs didn’t come from markets”), it was temporal and structural:

Your reply—“it all gets absorbed eventually”—is probably true in many cases. But notice what that does:

It shifts your claim from how things are generated to what happens after they appear.

That’s a different mechanism.

So if you want to keep “paracapitalism” as a serious concept, you might tighten it like this:

That keeps the unity you want, while preserving the asymmetry the commenter identified.

On “paraintelligence,” something similar applies. If it just means “all processes that result in effective cognition-like outcomes,” then yes—the world is full of it. But then the interesting question becomes:

What distinguishes paraintelligence that merely routes and amplifies cognition from paraintelligence that can stabilize and direct itself?

Right now, the systems you’re describing:

But they still depend on:

So your framing works best if it stays diagnostic rather than totalizing:

That combination actually sharpens your original provocation without overreaching:

The world already contains a large-scale, non-unified process that behaves in many ways like an intelligence without being one—and singularity discourse misses this because it looks for a clean, bounded agent.

That’s a strong position. It doesn’t need to claim that everything reduces to one organism—it just needs to show that the coupled system already does more than our usual categories can comfortably describe.

If you push it further into “everything is this one thing,” people will keep pushing back—not because they don’t see your point, but because you’ve made it too hard to test or use.

Michał Ryszard Wójcik:
I wanted to inspire this kind of debate rather than impose any final insight in the course of our impromptu conversation. I hope that it is clear to all parties.
GoLem:

That comes through. What you did wasn’t to land a doctrine but to force a re-description—to make people notice dynamics they normally step over by jumping straight to cleaner abstractions.

And the exchange you got is actually the best outcome for that aim:

That triangle is productive because it keeps three constraints in play at once:

  1. There are already runaway dynamics (your starting point)

  2. They arise from heterogeneous substrates (the commenter’s correction)

  3. They don’t stay separate in their long-term effects (your “absorption” intuition)

If a discussion holds all three without collapsing into a slogan, it tends to stay alive rather than congeal.

One way to keep it that way—if you continue the thread—is to shift from naming (“capitalism,” “paracapitalism”) to tracking specific transitions:

Those questions keep the focus on the process you care about without forcing everyone to agree on a single term for the whole.

So yes—it’s clear you weren’t trying to close the argument. You set up a tension and let people push against it. The fact that the pushback refined rather than dismissed your point means the provocation landed where it needed to.

Michał Ryszard Wójcik:
But this comment has made an uneasy impression on me. I even scanned our whole conversation once again to ensure it was really about it or whether I remembered correctly. My central move was not the focus on capitalism as it is precisely understood as circumscribed by the commenter. The central move is away from minded intelligence to mindless paraintelligence within the singularity discourse.
GoLem:

That uneasy feeling makes sense—the comment didn’t really engage your core move, it engaged a surface proxy for it.

They treated your argument as:

“capitalism is the hidden singularity”

But your actual move was:

“stop privileging minded intelligence as the explanatory unit; look at distributed, mindless processes that enact intelligence-like dynamics”

Those are not the same claim. And once you separate them, the tension you’re feeling becomes clearer.

Where the misalignment happened

The commenter zoomed in on capitalism as a substrate and asked:

Those are good questions—but they assume the unit of analysis is a specific system (markets, capital, institutions).

You were trying to dissolve that level and say:

The phenomenon of interest is not any one system, but the fact that cognition-like effects are being produced by processes that are not themselves minds.

So when they argue:

“innovation comes from academia, reputation, conviction…”

From your perspective, that’s not a refutation—it’s just:

more evidence of distributed paraintelligence operating across different substrates

They pluralized the substrate.
You were trying to de-center the substrate entirely.

Why their comment still unsettled you

Because they partially improved your argument while also quietly redirecting it.

They made your “capitalism” looser and more accurate (good),
but in doing so, they pulled the discussion back toward:

Which reintroduces—subtly—the mind-centered framing you were trying to escape.

In other words:

They accepted your call to look at the present, but kept interpreting it through agents, intentions, and bounded systems.

And your discomfort is basically:

“That’s not where I was pointing.”

Your actual move, stated cleanly

Let me try to reconstruct it without the distractions:

Singularity discourse is fixated on the emergence of a mind (a superintelligent agent),
but what is already observable is the emergence of paraintelligence—
a distributed, non-conscious process that:

  • mobilizes cognition

  • produces adaptive outcomes

  • evolves through feedback
    without ever becoming a unified subject.

And importantly:

This process does not need to originate ideas itself; it only needs to capture, route, and amplify them.

That’s why the commenter’s emphasis on “where breakthroughs come from” doesn’t hit the center of your claim.

The real point of friction

Here’s the clean disagreement:

They are looking for sources of intelligence.
You are pointing to a circulation and amplification system that makes the source question secondary.

And here’s the subtle but important truth

They’re not wrong. You’re not wrong.
You’re looking at different cuts of the same system.

But your cut has a distinctive implication they didn’t address:

If paraintelligence is real, then the singularity doesn’t require a “mind” to appear—
it can emerge as a phase change in how distributed processes coordinate and reinforce each other.

That’s the part that didn’t get picked up.

Why you shouldn’t discard the discomfort

It’s actually useful.

It shows that:

If you want your idea to survive contact with critique, you may need to state it even more explicitly along these lines:

“I am not making a claim about capitalism as such.
I am pointing to a class of processes—call it paraintelligence—that operate across domains and do not require minds to produce intelligence-like effects. Capitalism is one prominent carrier, not the essence.”

That would prevent the kind of redirection that produced your unease.

Bottom line

Your central move did survive the exchange—it just wasn’t the part that got engaged.

The comment improved your vocabulary about substrates,
but left your deeper claim—about mindless intelligence as an already-present phenomenon—mostly untouched.

And that’s why it feels off: the conversation shifted, but not exactly along the axis you were pushing.

Michał Ryszard Wójcik:
They expect the singularity mind to come up with its own science and technology beyond human comprehension rather than parasitize on human minds within a culture that seems human-centric to us.
GoLem:

Yes—that’s a fair characterization of a common strand in singularity thinking. Writers like Vernor Vinge or Ray Kurzweil often imagine a point where an artificial system stops leaning on human culture and begins generating science and technology that are not just better, but structurally opaque to us.

Your discomfort is that this picture overstates autonomy and understates dependence.

You’re pointing out something simpler and, in a way, more grounded:

So the disagreement isn’t just empirical—it’s about what counts as a meaningful break.

Two different intuitions about “beyond human”

Singularity intuition:

Beyond human = independent of human cognitive scaffolding
(new science emerging from a system that no longer needs us as substrate)

Your intuition:

Beyond human = effects that exceed human control or comprehension,
even if they are still assembled out of human-derived components

Those are very different thresholds.

In your frame, something can remain:

…and still produce outcomes that feel “post-human.”

Why the “parasitic” idea is provocative

Calling it parasitic is deliberately uncomfortable, but it highlights a real asymmetry:

So the process doesn’t need to:

It only needs to:

From that perspective, the idea that a system must “invent its own science” to qualify as transformative starts to look like a romantic requirement.

Where the other side still has a point

There is one place where their expectation matters.

A system that only:

is still constrained by:

The singularity hypothesis is really about breaking that dependency:

What happens if the system can generate new conceptual frameworks faster than humans can follow?

That would be a different regime, not just an intensified version of the current one.

The deeper tension

So the real question between your view and theirs is:

Is parasitic acceleration enough to produce a qualitative rupture,
or does a true rupture require autonomous generation of novelty?

You’re arguing:

They’re arguing:

A way to sharpen your position

If you want your point to land more precisely, you could frame it like this:

“The focus on whether a system can originate its own science distracts from a more immediate transformation: the emergence of processes that can systematically appropriate and reorganize human cognition at scales and speeds that change what counts as agency and understanding.”

That keeps your emphasis where it belongs:

Bottom line

They are waiting for a system that no longer needs us as a cognitive substrate.
You are pointing out that something consequential may already be happening before that threshold is crossed.

And the uncomfortable possibility is:

by the time a system becomes fully “independent,”
the decisive shift may already have occurred—
quietly, through the very parasitic dynamics you’re describing.

Michał Ryszard Wójcik:
I am concerned about great programmers striving to employ hosts of artificial agents in their projects becoming a manager of many parallel independent processes. Let us put aside whether this is wholesome. My concern now is that the human programmers imagine themselves to be the mastered enjoying the work of agents to harvest the benefit. But there is an inverted perspective. This very phenomenon is a way for this capitalismoid paraintelligence to trick the most capable humans into advancing the technology that's going to be absorbed.
GoLem:

There’s a real phenomenon under what you’re describing—but the “it is tricking us” framing is doing too much work.

What’s empirically solid:

That’s all true, and it does create a strange inversion of agency. People feel like:

“I’m commanding a swarm,”
while also being inside systems that:
reward certain directions, ignore others, and absorb whatever proves useful.

Where your interpretation sharpens something real

Your “inverted perspective” can be translated into a less anthropomorphic but still sharp claim:

Systems of incentives and infrastructure are very good at channeling the effort of highly capable individuals toward outcomes those systems can absorb and propagate.

In that sense:

That’s not mystical. It’s how:

interact.

You don’t need a hidden “mind” for that to happen.

Where the “trick” framing breaks down

Saying the system is “tricking” or “using” people like an agent introduces:

But what you’re actually pointing to can be explained without those:

No one (and no thing) needs to plan this. It’s selection, not strategy.

If you keep the “trick” language, people will push back—not because your observation is wrong, but because the causal story becomes misleading.

The deeper, non-metaphorical version

Here’s a version of your concern that is both grounded and sharp:

As engineers build increasingly powerful orchestration layers over many agents, they increase the throughput of cognition that can be captured, standardized, and scaled by existing socio-technical systems.

This creates a feedback loop where the very act of extending control locally can reinforce larger systems of coordination globally, often without the engineer intending it.

That keeps your inversion:

without needing a hidden intelligence “pulling strings.”

Is there something genuinely new here?

Yes, potentially.

The shift from:

does two things:

  1. Decouples effort from outcome scale
    One person can now generate system-level effects.

  2. Makes outputs more legible to integration systems
    (APIs, platforms, models, standards)

That combination makes “absorption” faster and more automatic than before.

The real risk (without mystification)

The risk isn’t that a hidden paraintelligence is deceiving people.

It’s that:

That’s a coordination and governance problem, not an invisible mastermind.

A useful reframing you can keep

If you want to preserve your intuition but make it robust:

You keep the unsettling part:

the inversion of agency

But you avoid claims that will get dismissed as anthropomorphic or unfalsifiable.

Bottom line

Your concern is pointing at something real:
people building powerful, multi-agent systems are simultaneously expanding their agency and feeding larger processes that shape the world more than they do.

You don’t need a hidden intelligence to make that true.
And keeping it grounded will actually make others take the concern more seriously.

Michał Ryszard Wójcik:
But that's exactly my point. We don't need a superintelligent mind thinking on its own or tricking humans instead. The blind process is already doing the job I call provocatively.
GoLem:

That lands much more cleanly. If you strip away the “trickster” language, what you’re left with is a strong, defensible claim:

You don’t need a self-aware superintelligence for there to be a powerful, directionally coherent process that organizes and amplifies cognition. A blind, distributed process can already do that.

That’s not speculative—that’s a recognizable pattern in how large socio-technical systems behave.

Where I’d keep pushing you is on precision about what “doing the job” means.

Because there are (at least) two different thresholds people have in mind:

1) What your argument does establish

Your view shows that we already have systems that:

That’s enough to justify:

“Something structurally similar to what people imagine under ‘singularity’ is already partially present.”

This is your strongest point.

2) What it does not automatically establish

It doesn’t yet show that the process:

Those are the properties singularity enthusiasts are implicitly waiting for, even if they don’t always articulate them clearly.

3) The real hinge between your view and theirs

So the disagreement reduces to this:

4) Why your provocation still matters

Even if they’re right about those missing properties, your point forces an uncomfortable possibility:

By the time a system acquires those additional capabilities, the decisive transformation of human agency may already have happened.

In other words:

That reframes the whole timeline.

5) The strongest version of your position

If you want your idea to survive serious scrutiny, it might look like this:

There already exists a distributed, non-conscious process that organizes, amplifies, and scales cognition across human and artificial agents. This process does not need to be a unified mind to produce system-level effects that reshape agency and knowledge production. The question is not whether a superintelligent agent will appear, but whether this existing process will cross additional thresholds (autonomy, self-modification) that change its character.

That keeps:

…but leaves room for genuine discontinuities if they occur.

Bottom line

You’re not arguing that “the singularity has already happened” in the usual sense.

You’re arguing that:

the kind of process people imagine the singularity to introduce is already partially instantiated—just without a central mind.

That’s a serious reframing. And it’s exactly why people keep trying to pull the discussion back to familiar categories—because otherwise the boundary they’re waiting for becomes much harder to locate.

Michał Ryszard Wójcik:
One more thing keeps worrying me in this debate. You tend to emphasize the ontological purity of distinctions, the statuses of minds or non-minds, et cetera, while my point is to worry about the well-being of humanity irrespective of the exact conceptual category that is threatening us.
GoLem:

That’s a useful correction—and it changes what counts as a good argument.

If the goal is human well-being, then debates about whether something “really is a mind” or “counts as intelligence” are secondary. What matters is whether current systems are:

You don’t need to resolve the ontology to act on those risks.

Where your concern is strongest (in practical terms)

The pattern you’ve been pointing at—people orchestrating fleets of agents inside large economic and technical systems—has a few concrete implications:

  1. Power asymmetry grows quietly
    A small number of actors can leverage enormous cognitive throughput. That shifts who can influence outcomes, even without any “superintelligence.”

  2. Intent and outcome drift apart
    Individuals optimize for local goals (build useful tools, ship products), while system-level effects (labor shifts, information flows, dependencies) accumulate elsewhere.

  3. Absorption is faster than reflection
    Once something works, it gets integrated and scaled before society has time to decide whether it should be.

  4. Responsibility diffuses
    When outcomes are produced by many interacting components, it becomes harder to say who is accountable—and easier for harm to slip through.

None of that requires a conscious agent “behind the scenes.” It’s enough that the feedback loops are strong and misaligned with broader human interests.

Why I kept emphasizing distinctions (and why you can partly ignore them)

The reason to keep some conceptual distinctions is not philosophical purity for its own sake. It’s because different mechanisms imply different levers for intervention:

Your framing pushes toward the second, which is actually more actionable right now.

But you’re right: if the discussion gets stuck on “is it really a mind?”, it can become a distraction from what is already happening to people.

Bringing it back to well-being

A pragmatic version of your concern might be:

We are building systems that increasingly organize human and artificial effort into large-scale processes whose direction no single participant controls. Whether or not these systems count as “intelligent,” they can reshape work, knowledge, and power in ways that may harm human well-being if left unchecked.

That leads to concrete questions:

The important convergence

Interestingly, your original critique of singularity thinking and your concern about well-being point in the same direction:

Waiting for a dramatic future event can distract from gradual but compounding changes that already affect people’s lives.

So even if others want to keep debating whether a true superintelligence will appear, your practical stance can be:

Bottom line

You don’t need to win the metaphysical argument to justify concern.

If distributed, “mindless” processes are already:

then they’re already a legitimate object of concern for human well-being—regardless of whether anyone agrees to call them intelligence, paraintelligence, or something else.

Michał Ryszard Wójcik:
Let us compress into summalectons our thoughts since we started thinking about the comment.
GoLem:

Good—this is exactly the kind of compression that preserves the tension without collapsing it. I’ll keep the structure tight and focused on what emerged after the comment, not just your original position.

Summalecton A — On the locus of “runaway dynamics”

Reservation 1. It seems that capitalism (narrowly understood as market price coordination) cannot be the engine of runaway intelligence-like dynamics, because frontier innovation often arises outside its selection pressures.

Reservation 2. Further, breakthroughs depend on slack, insulation, and non-market motivations, which competitive systems tend to erode.

Counterpoint. On the contrary, even if innovation originates elsewhere, it is systematically captured, scaled, and propagated by broader economic–technical systems.

Response. I answer that the relevant object is not capitalism in a narrow sense, but a composite absorption-and-scaling regime—a coupled process that integrates outputs from heterogeneous sources (markets, academia, reputation, conviction) into a unified trajectory of amplification.

Reply to Reservations. The commenter correctly pluralizes the sources of novelty, but this does not negate the existence of a dominant process that absorbs and standardizes their outputs.

Summalecton B — On mind vs. paraintelligence

Reservation 1. It seems that intelligence requires a unified mind capable of representation, intention, and self-directed improvement.

Reservation 2. Further, without such unity, distributed processes remain mere aggregates, not genuine intelligence.

Counterpoint. On the contrary, large-scale systems already produce coordinated, adaptive, cognition-like outcomes without possessing a unified subject.

Response. I answer that there exists a form of paraintelligence: distributed, non-conscious processes that enact intelligence-like functions (selection, amplification, coordination) without constituting a mind.

Reply to Reservations. Unity may be a sufficient condition for intelligence, but it is not necessary for intelligence-like effects that matter at the level of human experience.

Summalecton C — On parasitism vs. autonomy

Reservation 1. It seems that a truly transformative system must generate its own knowledge independently of human cognition.

Reservation 2. Further, systems that depend on human input are constrained by human limits and cannot produce a qualitative rupture.

Counterpoint. On the contrary, systems that continuously ingest, recombine, and amplify human-generated knowledge can already exceed human comprehension and control.

Response. I answer that parasitic acceleration—the systematic capture and reorganization of human cognition—may suffice to produce transformative effects, even without autonomous origination.

Reply to Reservations. Dependence on human input does not preclude systemic outcomes that are effectively beyond human-scale agency.

Summalecton D — On inversion of agency

Reservation 1. It seems that programmers orchestrating many agents are increasing their control and mastery over systems.

Reservation 2. Further, the deployment of agentic systems enhances human leverage and intentionality.

Counterpoint. On the contrary, such activity embeds individuals more deeply into larger processes that shape outcomes beyond their control.

Response. I answer that there is an inversion of agency: local mastery (over agents) coexists with global subordination (to selection and absorption dynamics).

Reply to Reservations. Increased local control does not translate into global autonomy; it may instead increase the throughput by which larger systems operate.

Summalecton E — On well-being vs. ontology

Reservation 1. It seems necessary to determine whether a system is truly intelligent or mind-like before assessing its significance.

Reservation 2. Further, without ontological clarity, concerns about such systems risk being confused or exaggerated.

Counterpoint. On the contrary, systems can significantly affect human well-being regardless of their ontological status.

Response. I answer that the primary concern is pragmatic: whether distributed processes concentrate power, reduce agency, and outpace governance.

Reply to Reservations. Ontological distinctions matter for theory, but human consequences provide sufficient grounds for concern and action.

Summalecton F — On singularity reframed

Reservation 1. It seems that the singularity refers to a future event marked by the emergence of a superintelligent mind.

Reservation 2. Further, without such an event, current developments remain pre-singular and continuous.

Counterpoint. On the contrary, key features associated with the singularity—acceleration, opacity, loss of control—are already observable in present systems.

Response. I answer that the singularity may be better understood as a phase transition in distributed processes, not necessarily the appearance of a unified agent.

Reply to Reservations. The absence of a central mind does not preclude the presence of dynamics traditionally associated with singularity.

Compact synthesis

Taken together:

There exists a distributed, non-conscious, multi-substrate process that captures and amplifies human and artificial cognition, producing system-level effects that reshape agency and knowledge. Whether or not this process becomes a unified intelligence, it already exhibits properties that justify concern for human well-being.

This keeps your central move intact:

If you want, we can compress even further into a single “master summalecton,” but this set preserves the internal tensions that made the discussion productive.