Beyond Binary Systems—Queering the Machine

Artificial intelligence, once confined to the pages of speculative fiction and the chalkboards of theoretical mathematics, has evolved into one of the most influential forces shaping 21st-century society. From chatbots to facial recognition, from content curation to predictive policing, AI systems mediate the way we engage with reality. But as these systems become more autonomous and embedded in our daily lives, a provocative question has emerged in the worlds of queer theory and tech ethics alike: Can artificial intelligence be queer?

To some, the question may seem absurd—after all, queerness is rooted in lived experience, human subjectivity, and the negotiation of power, identity, and desire. Machines, by contrast, are tools designed to perform functions. But therein lies the deeper challenge. AI is not born neutral. It is built by humans, trained on human data, and embedded with human values—often uncritically and unconsciously. These values are not just economic or political; they are also cultural, gendered, and normative.

To queer AI, then, is not to suggest that Siri might come out as nonbinary, or that ChatGPT might demand hormone therapy. Rather, it is to interrogate the structures, assumptions, and logics that shape the design, deployment, and discourse around artificial intelligence. It is to ask how systems built on binary classifications, hierarchies of normalcy, and optimization of efficiency might resist or reinforce the kinds of exclusions that queer and trans people have long fought against. It is to recognize that identity, creativity, embodiment, and meaning-making are not simply “human” capacities to be replicated, but contested terrains shaped by culture, power, and resistance.

This post explores how queerness can illuminate the hidden assumptions of AI design, the dangers of digital bias, and the radical possibilities of reimagining intelligence itself. We draw on the voices of queer and trans technologists, artists, and scholars who are challenging the dominant paradigms of technology, and we examine how queerness—understood not just as identity but as method—might offer a blueprint for more ethical, inclusive, and imaginative forms of digital life.

Binary Code, Binary Thinking: The Queer Limits of Machine Logic

At the heart of nearly all modern computing lies binary code: ones and zeroes, yes and no, true and false. This digital binary mirrors and reinforces the social binaries that queer theory has long critiqued—male/female, straight/gay, normal/deviant. The foundational logic of computation, while efficient and elegant in design, often becomes a metaphorical and structural limitation when applied to human behavior and identity.

AI systems trained on massive datasets operate by finding patterns and reinforcing statistical regularities. While this might seem objective, it often encodes normative assumptions. Gender, for instance, is frequently treated as a fixed binary in data schemas and model architectures. Many facial recognition systems require users to be categorized as either “male” or “female,” excluding or misclassifying transgender, nonbinary, and gender-nonconforming individuals.

In a now-infamous example, a Google Photos algorithm labeled Black individuals as “gorillas”—a deeply offensive error rooted in biased training data and a lack of diverse representation. Similarly, facial recognition systems have been shown to misgender and misidentify trans people, particularly those whose appearances do not conform to Western gender norms. The problem is not just in the data—it is in the very assumptions of what AI is supposed to do: categorize, optimize, reduce complexity.

As Sasha Costanza-Chock writes in Design Justice, “Design is never neutral. It either reinforces or challenges existing systems of power.” When AI systems are built without attention to queer and marginalized experiences, they do not simply omit these identities—they render them illegible, invisible, or wrong.

To queer AI, then, is to resist the binary epistemologies that dominate machine learning. It is to insist on plurality, fluidity, contradiction, and ambiguity—not as bugs in the system, but as features of reality worth honoring.

Queer Technologists Rewriting the Script

Fortunately, a growing number of queer and trans technologists are pushing back against the dominant paradigms of AI development. Artists, programmers, ethicists, and hackers are exploring what it means to build technologies that reflect queer ways of knowing, being, and imagining.

Among them is Os Keyes, a nonbinary researcher whose work critiques how gender is operationalized in algorithmic systems. In their paper “The Misgendering Machines,” Keyes argues that many AI tools fundamentally misunderstand gender as a static, observable trait, rather than a lived and constructed identity. “What these systems are doing,” they write, “is reinforcing the idea that gender is a thing that can be seen, rather than a role that is played or an identity that is felt.”

Another important voice is Laine Nooney, a historian of computing who has explored how gender and sexuality shaped early computer culture, not just in who was allowed to build systems, but in how those systems were imagined. By recovering forgotten stories and re-centering queer contributions, scholars like Nooney challenge the myth of the lone (cis-male, straight, white) tech genius and open the door to more inclusive narratives.

Meanwhile, artist-technologists like Zach Blas have created speculative works that critique and subvert surveillance technologies. Blas’s “Facial Weaponization Suite” creates masks that confuse facial recognition software, offering a poetic and political intervention into the logic of biometric control. In doing so, Blas asks not just whether AI can be ethical, but whether resistance to AI itself can be an act of queer futurism.

These interventions show that queerness in tech is not just about representation—it is about reimagining the very foundations of what technology is and does.

Data is Never Just Data: Bias, Erasure, and the Violence of Inclusion

Mainstream narratives around AI bias often focus on the need for more inclusive datasets. The assumption is that if we simply add more diverse identities into the training data, the system will become fairer. But this approach can be dangerously simplistic.

As Ruha Benjamin explains in Race After Technology, inclusion without structural change often becomes a form of “cosmetic diversity”—a way to appear progressive without addressing deeper injustices. Simply adding more queer or trans people into surveillance databases does not make the technology less harmful; it simply makes its violence more evenly distributed.

Moreover, inclusion itself can be coercive. Many trans people do not want to be accurately identified by surveillance systems. For some, invisibility is a form of safety. As the trans scholar and artist Micha Cárdenas has argued, “opacity”—the refusal to be fully known or categorized—can be a radical and protective gesture.

The push for “ethical AI” must therefore go beyond fairness metrics and demographic parity. It must ask more fundamental questions: Who decides what data is collected? Who benefits from its analysis? Who is harmed by its deployment? And can we imagine forms of intelligence that are not premised on domination, extraction, and control?

Queer theory offers a lens to ask these questions—not just to improve technology, but to transform our relationship with it.

Rethinking Intelligence: From Optimization to Imagination

One of the most profound contributions that queer theory can make to AI discourse is a redefinition of intelligence itself. In most current frameworks, intelligence is defined in terms of problem-solving, prediction, efficiency, and control. These values reflect the priorities of industrial capitalism, not necessarily the full spectrum of human (or non-human) cognition.

What if intelligence were defined not by optimization, but by imagination? Not by conformity to norms, but by the ability to challenge them?

In this sense, queerness is not just a critique of AI—it is a mode of intelligence in its own right. Queer life requires constant adaptation, creative survival, the ability to navigate hostile systems with grace and subversion. It involves emotional acuity, relational dexterity, and a deep understanding of power and performance. These are not deficiencies to be corrected by algorithms—they are insights to be honored.

Indeed, some of the most exciting experiments in AI and machine learning are embracing non-normative logics. Generative art tools like GANs (Generative Adversarial Networks) produce strange, surreal outputs that defy conventional standards. Queer artists and coders are using these tools to explore themes of fluidity, transformation, and multiplicity.

Even the idea of a “non-binary” AI is gaining traction—not in the sense of gender identity alone, but as a broader challenge to systems that demand binary inputs and outputs. Projects like Genderless Voice seek to create AI voices that do not conform to traditional gender norms, raising questions about embodiment, recognition, and the limits of machine-mediated identity.

These experiments do not offer easy answers. But they open space for a different kind of conversation—one that centers creativity, complexity, and critique.

Queer AI in Practice: Building Otherwise

What would it mean to build AI systems that are explicitly queer—not just inclusive of LGBTQ+ identities, but rooted in queer epistemologies? Some guiding principles might include:

• Ambiguity over clarity: Embracing uncertainty, fluidity, and multiplicity rather than reducing everything to fixed categories.
• Opacity over transparency: Recognizing that not all identities want to be visible or knowable to systems of surveillance.
• Imagination over optimization: Prioritizing creative expression and ethical reflection over speed and efficiency.
• Relationality over autonomy: Designing systems that foster interdependence, care, and community rather than isolation and control.
• Critique over compliance: Building tools that challenge dominant paradigms, rather than reinforce them.

Examples of these principles can be found in speculative design projects, activist tech initiatives, and community-driven data practices. The Trans Tech Social Enterprises network supports trans-led innovation in technology. The Design Justice Network develops frameworks for participatory, equitable design. Even the act of refusing to build certain technologies—such as facial recognition tools—is a form of queer resistance.

These efforts suggest that queerness is not an add-on to AI ethics—it is a foundation for imagining technology differently.

Queering the Future—Ethics, Resistance, and Digital Liberation

The question “Can AI be queer?” is not just about machines—it is about us. It is about the values we encode into the systems we build, the stories we tell about intelligence and identity, and the futures we dare to imagine.

In a world increasingly shaped by algorithms, queerness offers a vital counterpoint. It reminds us that not all logic is linear, not all knowledge is quantifiable, and not all truths are singular. It teaches us to value ambiguity, to resist the violence of normativity, and to embrace the beautiful messiness of being.

To queer AI is to reject the idea that progress means conformity. It is to insist that liberation cannot be programmed, but it can be dreamed, designed, and demanded. It is to understand that ethics are not a checklist, but a commitment to justice, care, and transformation.

As queer technologist Danielle Brathwaite-Shirley puts it, “If the future is not designed by us, then we will not be in it.” The call, then, is clear. Queer the code. Disrupt the system. Reimagine the machine.

Because queer theory is not just for people. It belongs in our algorithms, too.