Insights on Sovereignty Models (1)

500,000 humans for rent to AI clients—can anyone beat that? Cited by Wired, Forbes, and Futurism, RentAHuman.ai is a platform that intrigues, amuses, and perhaps reveals a new level of acceptability regarding the transfer of sovereignty and the role of the individual as a resource.

RentAHuman.ai will serve as our starting point for an analytical and historical look at platform models, giving you the keys to one of the most powerful models of distributed sovereignty.

But let’s take a quick detour to the year 1770.

In 1770, Wolfgang von Kempelen built the Mechanical Turk, an automaton supposedly capable of playing chess. In reality, it concealed a highly talented human inside the machine, talented enough to defeat Napoleon Bonaparte, Empress Catherine II of Russia, and Benjamin Franklin!

In 2005, Amazon deliberately named its micro-task platform Mechanical Turk in tribute to this historic trick. On Amazon Mechanical Turk, humans complete cognitive tasks that algorithms cannot yet perform, yet the system is designed to keep this human contribution invisible, encapsulated within a technical workflow.

RentAHuman takes it a step further: the machine becomes the client, and the human is transformed into a biological peripheral available on-demand via an API call.

As you can see from the screenshots, RentAHuman.ai aims to give AI access to the physical world through human IRL (In Real Life) action capabilities. Incredibly, humans are eager to play along. Among the 500,000 registered users, we find both curious experimenters and very serious professionals looking to showcase their high-value skills, much like on LinkedIn.

The marketplace offers them very real missions in areas like user research, logistics, and marketing.

And this is only the beginning.

1- What the Platform Does to the World

From Tool to Cultural Infrastructure

Platforms are primarily conceived as digital intermediation tools and a category of multi-sided business models. However, when Uber redefines the relationship to work for five million drivers worldwide, when Airbnb transforms how entire cities house-and unhouse-themselves, and when Amazon becomes the default commerce infrastructure for millions of SMBs, we move past the concept of a simple tool and enter the realm of cultural infrastructure: a system that transition from facilitating exchanges to redefining how society itself functions.

Sangeet Paul Choudary, co-author of Platform Revolution (2016), phrased this transition beautifully: traditional businesses operate primarily as value-creation pipelines, from raw materials to final products. Platforms, in contrast, create value by orchestrating interactions. They create nothing physical; they bring people together, and their power lies in this orchestration. For some, this orchestration transforms into massive influence. To orchestrate is, in effect, to decide who meets whom, at what price, and by what rules.

Shifting the Center of Control

What fundamentally distinguishes a platform from a traditional business is not the technology. It is the shift of where control resides. In a pipeline model, the company controls production. In a platform model, it controls the conditions of the interaction. The Uber driver owns the car but not the market. The Airbnb host owns the apartment but not its visibility. The Fiverr freelancer owns their skills but not the ranking algorithm. The Instagram user owns their attention but not the model exploiting it. In each case, ownership of the means is decoupled from control over the conditions of monetization. It is this decoupling that differentiates a true platform from a basic directory.

In the rest of this article, we will explore the different levers of sovereignty and their impacts, but first, let's take a quick look at how platform models have evolved.

II. Genealogy: How the Platform Became a Societal Paradigm

The platform is the product of a long history where technical innovations, economic theories, social transformations, and political choices combined to produce the model we know today.

Two-Sided Markets and Intermediation Theory (Pre-2000)

The concept of a platform as a two-sided market predates the Internet. Local newspapers, stock exchanges, and medieval fairs are all intermediation devices where value lies in building connections. What changed with the digital era is scale and speed. In their seminal paper "Platform Competition in Two-Sided Markets" (2003), Jean-Charles Rochet and Jean Tirole show that in a two-sided market, the price structure (how pricing is distributed between the two sides) is paramount. Their model explains why a platform can rationally charge very low or even below-cost prices on one side of the market and secure revenue from the other, making it more challenging to identify predatory pricing strategies under traditional antitrust laws.

The Invention of Remote Trust (1995-2007)

Early e-commerce platforms like Amazon and eBay, both launching in 1995, had to address the challenge of trust in completely new ways, establishing mechanism to enable transactions between total strangers without direct contact. eBay's feedback system, introduced at its inception, became one of the first major online reputation systems, marking a foundational shift: an economic actor's reputation became quantified, public, and cumulative.

The Attention Economy and Surveillance Capitalism (2007-2012)

The launch of the iPhone in 2007 and the rise of social media shifted the paradigm: platforms no longer just connected buyers and sellers; they captured user attention and monetized it through targeted advertising built on massive data extraction. This context birthed the widespread practice of "behavioral surplus" as described by Shoshana Zuboff, where behavioral data exceeding service needs is repurposed as feedstock for predictive products. In Platform Capitalism (2017), Nick Srnicek offers a complementary view, showing that platforms are not accidental disruptions but the contemporary expression of capitalist logic, finding in data a raw resource to extract indefinitely.

Uberization and Control Over physical Assets (2012-2020)

Uber, Airbnb, Deliveroo, TaskRabbit, Fiverr: in just a few years, platforms transcended information-sharing to claim domain over physical bodies, homes, vehicles, and assets. Parker, Van Alstyne, and Choudary formalized this shift in Platform Revolution (2016). The stakes of control became explicitly social: redefining work, living, and mobility. And it is here that the decoupling of ownership and control shines in full force, as discussed in our introduction.

The Platform as Public Infrastructure (2020-2025)

The pandemic accelerated a trend already underway: platforms transitioned into quasi-public infrastructures. A service outage was no longer a minor inconvenience; it was a systemic crisis. The European Union responded with the Digital Markets Act and the Digital Services Act (2022). Historically pipeline companies began internalizing platform logic, and the model became ubiquitous.

When Machines Become the Clients (2025-2026)

RentAHuman.ai and the rise of autonomous AI agents mark a conceptual turning point. An agent posting a task on RentAHuman is not an employee: it is an autonomous system connected via API, capable of searching, hiring, and paying a human without human intervention. When a machine can participate as a client in the labor market, the questions change: who is accountable when an AI directs a human? Where does moral responsibility lie when the hiring entity is software, itself driven by another automated system?

This brings to mind Deleuze, who in his 1987 lecture at FEMIS described the transition from "disciplinary societies" to "societies of control": open environments where power operates through continuous modulation. With AI agents as clients, this modulation goes recursive: we orchestrate systems that, in turn, orchestrate human beings.

III. The Four Pillars of Platform Control

Developing from an analysis of the interviews and writings of platform founders and theorists, we can identify four primary operations of control that consistently appear from Uber to Fiverr, and from Amazon to RentAHuman. These four pillars operate simultaneously, reinforcing one another, to build the core architecture of the platform model.

1. Control via Interface Architecture

The platform’s first strategic action is a design choice. Whoever designs the interface, its matching algorithms, visibility rules, and pricing controls the ecosystem by extension, even if individual participants feel completely autonomous. Far from being a neutral tool, the interface is a normative framework that shapes all potential actions.

Brian Chesky, co-founder of Airbnb, has often emphasized how everything started with designing the experience. Early Airbnb employees photographed hosts' apartments themselves because high-quality imagery built trust, and trust generated transactions. Dara Khosrowshahi, CEO of Uber, describes this from another perspective on the Acquired podcast (2023): the architecture decides which drivers see which rides, at what dynamic price, and with what incentives. Every interface design decision is a direct application of power.

The Meta lawsuit of February 2026 revealed interface architecture pushed to its absolute limits. Internal documents showcased interfaces carefully designed to minimize user exit points: infinite scrolls, autoplay videos, and notifications timed at optimized intervals. Design here isn't just about user experience; it's an attention-capture architecture where internal target metrics (such as increasing teenage daily usage from 40 to 46 minutes) stood in stark contrast to their public discourse on "quality of experience."

2. Sovereignty Through Standardization of Action

The second pillar involves breaking down real-world services into standardized, exchangeable units. An Uber ride, a Fiverr gig, an Airbnb night, a TaskRabbit chore: each unit is defined by standard parameters (duration, price, location, required skill) that make it readable by algorithms. This granular slicing of the world is what enables orchestration at a global scale.

Stacy Brown-Philpot, former CEO of TaskRabbit, described her platform as the "new normal of work." Behind this elegant phrase lies a precise tactic: transforming complex tasks into standardized units assignable in just a few clicks. The person performing the task, the "tasker," is reduced to simple parameters (rating, availability, location, price) just as much as their actual human talent.

RentAHuman takes this logic to its ultimate conclusion: humans are explicitly described as the "meatspace layer" of artificial intelligence. The tasks offered reduce human intervention to what machines cannot yet replicate: physically being in a space. This is the API-fication of the human.

3. Control via Metrics and Continuous Evaluation

The third pillar is perhaps the most powerful because it becomes completely internalized. Ratings, scores, response times, acceptance rates, and feedback loops: the platform institutes a framework of constant evaluation that defines an implicit ethos of high performance. This regime is internalized by participants who self-discipline to maintain their standing without the need for traditional managerial supervision.

On Fiverr, for instance, performance is evaluated constantly: response times, on-time delivery rates, and customer satisfaction. The platform's ranking algorithm translates these metrics directly into visibility. When a score drops, a freelancer's business fades. Recently, Kaufman introduced Fiverr Go, a feature where AI agents can replicate a creator's unique style, promising freelancers they can "work even while they sleep." The paradox is fascinating: metrics become so demanding that individuals are encouraged to automate themselves to keep up.

4. Control via Narrative

The fourth pillar is the most subtle: creating an empowering narrative that symbolically upends the reality of dependence. "Be your own boss" (Uber). "Monetize your talent" (Fiverr). "Belong anywhere" (Airbnb). "Get paid when agents need someone in the real world" (RentAHuman). The narrative elegantly reframes precarity as freedom, atomization as autonomy, and algorithmic subordination as ultimate "flexibility."

This fascinating paradox is the engine of the entire platform model. The platform requires its users to feel totally free so that they enthusiastically self-optimize. Chesky champions "belonging," not commission fees. Kaufman focuses on "unleashing talent," not algorithmic dependency. Khosrowshahi positions Uber as a universal "mobility infrastructure." The vocabulary is systematically selected to emphasize the desired emotion (fluidity, community, autonomy) rather than the underlying mechanism (matching, ranking, extraction).

The Meta trial of February 2026 offered a dramatic example of this gap between narrative and reality. Adam Mosseri, head of Instagram, testified under oath that he did not believe social media posed a "clinical addiction" threat, preferring the term "problematic use" within internal company communications. Concurrently, internal emails entered into evidence showed employees comparing Instagram to a drug and jokingly describing themselves as "dealers," referencing