The way a culture learns to see rarely changes all at once, yet there are moments when familiar visual cues begin to lose their original meanings. That moment is happening now, as images associated with detail, sophistication and a sense of mastery register for some viewers, while those raised in a screen-forward society are gravitating towards work that feels amateur, personal and an overall “authentic” expression.
This movement demonstrates broader changes in the conditions under which visual habits form. Repeated exposure shapes how people learn to see, and today’s environments diverge sharply from the ones that defined the dominant standards of the last century. The institutions that once instructed visual literacy no longer hold the same influence or authority, and the effects of that displacement are now visible across culture.
Changes In Visual Literacy
In Believing Is Seeing: Creating the Culture of Art, Mary Anne Staniszewski describes art as a constructed category, one that is produced through systems of display and interpretation. Museums, galleries, catalog essays, and exhibition design do more than display work; they teach viewers how to read and interpret it. Essentially, meaning emerges through framing rather than from the object alone.
This framework is useful, particularly as those same institutions no longer exert the same level of influence over everyday perception. For those born in the late 1990s and onward, visual literacy has developed primarily through use of screens.
TV, the internet and social platforms (with fair argument that they are no longer “social”) have become the dominant visual frame and carrier of meaning. Images and video now circulate in continuous streams authored by friends, influencers, strangers and brands and, over time, the separation between these source types has softened. Personal moments, entertainment, education and marketing persuasion sit side by side with little distinction.
These conditions have produced a different type of viewer who is trained to evaluate images through signs of human involvement. Interpretation now orients around whether the person can be sensed behind the work, whether intention, subjectivity or lived context is legible, rather than if the work conforms to an inherited standard of refinement or compositional quality.
The Development Of The Vibe Check
When visual meaning is learned through this environment rather than through institutional curation, the interpretation begins to rely on a different set of cues. Viewers accustomed to feeds that blend personal expression, entertainment, education, and marketing develop sensitivity to signals that help them orient themselves within that ambiguity. Over time, these signals consolidate into habits. They are not consciously chosen so much as absorbed through consistent exposure and repetition.
Research about Influencer credibility and authenticity, particularly as it relates to trust and perception, reflects this shift. Younger viewers consistently engage more favorably with content that appears informal, in-progress, or lightly edited with live-streaming being a clear example. These qualities tend to register as indicators of authenticity and trustworthiness rather than as evidence of low effort.
Within digital spaces, sensitivity to credibility and authorship becomes increasingly refined. Research from Stanford’s Visual Persuasion Labsuggests that pre-rational visual cues shape judgments of authorship and credibility before conscious evaluation takes place. Viewers, knowingly or not, respond to small signals of human creation—irregular pacing, the “millennial pause”, situational framing, uneven composition—well before they are able to articulate why an image or video clip reads as believable.
What is being sensed is whether a human can be felt behind what is displayed. As this process is practiced, the more this judgement becomes automatic, registering as intuition rather than analysis. This dynamic is often played out in real life when younger viewers attempt to explain to older viewers why an image or video online “feels fake,” even when the technical specifics are difficult to name. IYKYK.
Part of this dynamic is that what signals something as human-made is not confined to a single cue or aesthetic. As technology makes even high-art mimicry increasingly easy to produce, views of human authorship carry greater informational weight. These cues point toward intention, context, and relatable lived specificity. Authenticity, in this sense, functions less as a value judgement and more as a perceptual skill that was shaped by daily exposure to digital environments where recognizing human presence has practical importance.
When Visual Habits Become Cultural Form
Once this judgement becomes automatic, it rarely remains confined to the interpretation alone. Perceptual habits tend to externalize themselves, shaping not only how images are read but how they are made, chosen and lived with. A sensitivity to human presence often developed due to the daily navigation of digital spaces begins to surface across cultural forms that may seem unrelated.
Tattooing offers a material expression of this change. Over the past decade, studios have seen sustained demand for new takes on jammers and fillers, “ignorant tattoos”, and doodle-based pieces, often described as the conteporary “sticker sleeve” style. These designs preserve immediacy rather than refinement, with most designs being considered juvenile. These pieces retain qualities viewers have learned to associate with humanity with their on-purpose imperfections that appear more work-in-progress, aligning with the cues the viewers have learned to associate with human presence.
These tattoos resemble the kinds of images people use to communicate, to think and to document their lives. They are the doodles, symbols and visual labels that represent them in that moment. Tattoo historian Matt Lodder has observed that these choices indicate a comfort with visual language shaped by digital self-expression, where meaning is carried through gesture rather than finish.
Within fashion, there has been a rise in athleisure that is often explained through comfort or lifestyle changes, specifically in our post-COVID and rise of hybrid work society. Its aesthetic logic aligns closely with the perceptual habits that have formed online. Clothing that prioritizes ease (and the ease of ease), adaptability and unforced presentation demonstrates lived conditions rather than curated display. As with doodle-based tattoos, the appeal lies in the sense that the object is shaped by use rather than optimized for appearances.
Design and branding follow similar patterns. Identity systems increasingly incorporate irregular or custom typography, analog textures, or subjective layouts. These choices appear across packaging, editorial formats and campaign work, particularly when audiences are expected to engage quickly and intuitively. Emerging commentary has begun to note the prevalence of handwriting-like typography on packaging and diaristic visual elements, reflecting how legible these cues are becoming at scale. Further, brands are eager to capture the success of influencers by downgrading production quality of their owned content.
The thread connecting these expressions is not a shared aesthetic but their shared function. Each retains evidence of human involvement that is recognizable without instruction. Tattoos reflect choices preserved in the moment, the clothing showcases a real life being lived, and design that shows off its imperfect process rather than hiding it.
These types of signals are shown to be culturally preferred at a time when they have to coexist with generative systems that are operating under a different directive.
Not What Can Be Technically Achieved
Recent generative AI has entered our visual landscape with remarkable fluency. Trained on vast archives of commercial photography, illustrations, memes, paintings, videos, and more, these systems produce outputs marked by balance, stylistic consistency, and the high-quality visual standards that have dominated both institutional and commercial culture for decades. As a result, the presence of these models intensifies the current condition: visual abundance without clear authorship.
Within a digital environment saturated by refined imagery, generative outputs fit nicely into the stream. What has become noticeable are the moments where a viewer’s recognition falters, occurring when an image appears complete yet difficult to place, expressive yet oddly uninhabitable. The issue is not realism or quality, but the legibility of origin.
Generative AI “creates” by resolving probability across enormous datasets, producing shapes and forms that use statistical centrality rather than situational specificity. This yields images or videos that feel internally coherent but slightly off. The decisions embedded in the output remain invisible because they are not tied to a particular moment, constraint or lived context.
Viewers trained to look for presence can register the absence of it quickly, even if they cannot articulate it. The same perceptual habits that guide attention towards informal content, doodled tattoos, and lived-in clothing also shape responses to generative imagery. People sense when an image lacks the kinds of irregularities or rawness that emerge from their exercised judgement. The response is not outright rejection so much as a pull of uncertainty. Very uncanny valley.
As these generative systems absorb quality and reproduce it at scale, it all begins to lose its function as evidence of skill or effort. Instead, it draws attention towards attributes that resist synthesis and remain perceptible. What matters increasingly is not whether an image can be generated, but whether it can be recognized as having passed through a person.
Tasting Recognition
As generative AI continues to absorb visual language to then reproduce, long-standing ideas about taste (which I have written about) introduce new talking points. Evaluations that were once rooted in refinement, perceived quality, assumed technical execution struggle to account for why certain images now feel inert while others continue to hold attention. We are quietly moving from whether something is “good” to whether something feels “real”.
Taste, in this environment, functions less as an arbiter of quality and skill and more as a mechanism of recognition.It becomes a way for sorting images based on their perceived origin over an adherence to formal standards. This reframing helps explain why some debates about a decline in taste can miss the point. The pull towards irregular, amateur or unfinished work does not display a disregard for discernment but rather a change in what a viewer is attempting to discern.
While machines can generate consistent quality effortlessly, taste adapts to identify what that consistent quality cannot express. Under these conditions, a beautiful output can no longer act as evidence. Instead, cues that once read as rough or casual take on a new meaning that signals an image passed through human creativity, not probabilistic assembly.
Seen this way, the growing preference for what appears imperfect or even “bad” is not a rejection of quality or high art. It is a practical and reasonable response to authorship ambiguity.Taste shifts towards what helps the viewer answer the new question of where did this come from, and who made it?
Make it Bad
In an image economy being dictated by abundance, the work that captures attention often introduces friction. “Bad”, in this sense, is about the persistence of human traces that remain visible.
This is not an abandonment of design and art standards but rather an understanding of the reordered priorities formed by the conditions of our contemporary visual life. The qualities once associated with refinement have been robbed of their ability to signal authorship.
What is becoming evident are the limits of technology (and hopefully, the small pockets of rejection of it). Models and machines struggle, and will struggle, with humanity because humanity has context that cannot be abstracted into rational variables for calculation.
The cultural preference for the irregular shows up in how images are being read, how objects are chosen and how expression is valued. “Bad” becomes meaningful because it carries information that tells the viewer that something has passed through a person’s feelings, thoughts and creativity.
It is no longer whether an AI can make what humans have done before. They can. We know this. This is about how people recognize themselves in what they see and which signals continue to support their recognition of others as automation expands and works to drown us.
One comment, left beneath a passing Instagram reel, captured the intuition succinctly: “The first piece of art humans left behind was of hands—the one thing AI still can’t get right.”