The Authenticity Verification Laboratory occupied the top floor of a glass tower in Midtown. Marcus had passed it a hundred times without noticing, just another anonymous corporate facade among dozens. But inside, it was all white walls and humming machines and people in lab coats who looked like they knew things he didn't. He hated it already. "Mr. Webb." A young woman in a crisp white coat met him at reception. Her badge read Dr. Sarah Chen, Lead Analyst. "Welcome to AVL. I'll be giving you the tour today." "Thank you. I'm curious to see how your process works." She smiled, a thin, professional expression that didn't reach her eyes. "Most people are. This way." The laboratory was divided into sections, each with its own array of equipment. In the first room, technicians fed images into scanners that measured brushstroke patterns, color distribution, and compositional symmetry. "This is our pattern analysis division," Dr. Chen explained. "We look for statistical anomalies that suggest AI involvement. Machines tend to create art that's mathematically consistent in ways humans aren't." "And how accurate is this method?" "About seventy-three percent. When combined with other techniques, we can reach ninety-one percent confidence." Marcus nodded slowly. "And the remaining nine percent?" Dr. Chen's smile flickered. "Uncertain. Every system has limitations." They moved to the next room, where larger machines analyzed physical canvases. "Here we examine the material properties. Paint composition, canvas age, tool marks. AI-generated art that's printed on canvas can often be detected through microscopic analysis of the surface texture." "But if someone paints over an AI-generated image?" "Then it becomes more difficult. We look for inconsistencies between the underlayer and the visible surface, but a skilled artist could potentially hide the evidence." Marcus felt a familiar unease growing. "So your verification is..." "Probabilistic. Not definitive." Dr. Chen met his eyes. "We provide evidence, Mr. Webb. Not certainty." The final room was smaller, more intimate. A single desk with two chairs, a large monitor, and a stack of images. "This is where we show our analysts the verification methodology," Dr. Chen said. "We train them to look for patterns that machines might miss." "Like what?" "Emotional inconsistency. Human artists often make small errors, proportions slightly off, colors that don't quite match. AI systems tend to be too perfect, too consistent. Our analysts learn to spot the difference." Marcus studied the setup. "And how effective is this training?" "Combined with our technical analysis, we achieve approximately ninety-one percent accuracy. But there's always uncertainty, Mr. Webb. Art is subjective, and the technology keeps improving." "Uncertainty." Marcus let the word hang. "So your entire verification system is built on probability, not certainty." "Everything in science is probability. We provide evidence, not absolute truth." Dr. Chen met his eyes. "The question is whether that's enough for your purposes." He left the laboratory in a daze. The city felt different now, sharper, more hostile. Every billboard, every window display, every piece of public art seemed to mock him. How much of it was human? How much was machine? And how would he ever know? His phone buzzed. The unknown number again: How did you do, Mr. Webb? Better than random chance? Or worse? He didn't respond. Back in his office, he pulled up Jimmy Okonkwo's portfolio again. The work was good. He was sure of it. The rawness, the urgency, the quality that said human, it was there. He'd seen it. He'd felt it. But had he? Or had he seen what he needed to see? What he wanted to see? The question gnawed at him. He opened a new browser tab and searched for information on AI art generation. The results were overwhelming, dozens of platforms, each claiming to produce "human-quality" art. Testimonials from artists who used AI as a "tool." Critics debating whether AI art was "real" art. One article caught his attention: The Uncanny Valley of Creativity: When Human and Machine Become Indistinguishable. He read: "The question is no longer whether AI can create art. The question is whether we can tell the difference, and whether the difference matters. As AI systems become more sophisticated, they learn to mimic not just the products of human creativity, but the processes. They learn to be inconsistent, to make 'mistakes,' to create work that feels raw and urgent and human. At some point, the distinction becomes philosophical rather than perceptual. We can no longer see the difference because there is no difference to see." Marcus closed the browser. That night, he dreamed about the anonymous painting from his lecture. In the dream, he stood before it, studying the dark swirls, the central void. But as he looked, the painting began to change. The brushstrokes smoothed out. The colors became more uniform. The texture flattened. And then, slowly, words appeared across the surface: YOU CAN'T TELL THE DIFFERENCE BECAUSE THERE ISN'T ONE. He woke in a cold sweat. The next morning, he called Dr. Chen. "I have a question," he said when she answered. "Off the record." "Go ahead." "If someone wanted to fake human art, using AI but making it look human, how would they do it?" There was a long pause. "That's not a theoretical question, is it?" "No." Another pause. Then: "They'd use AI to generate a base image. Then they'd paint over it, by hand, with real paint. They'd introduce inconsistencies, imperfections, the kinds of 'mistakes' that signal human creation. And they'd be very careful about their story, how they talk about their process, what they claim about their influences." "Could your lab detect that?" "Possibly. If we examined the canvas closely enough, we might find evidence of the underlayer. But if they were skilled enough..." She trailed off. "Mr. Webb, are you asking about a specific artist?" Marcus thought about Jimmy Okonkwo. The studio that felt staged. The vague answers about process. The laptop in the corner. "No," he said. "Just curious." He sat in his office for a long time after the call ended. The exhibition was in five months. He'd committed to Jimmy Okonkwo as his centerpiece. The board was expecting results. In the back of his mind, a question was forming, a question he didn't want to ask but couldn't avoid. What if Jimmy was lying? What if the "authentic human art" Marcus had staked his reputation on was nothing more than a sophisticated fake? What if everything he'd built his career on was a beautiful, elaborate fiction? He pushed the questions away. Tomorrow, he would visit Jimmy again. He would ask better questions. He would find the truth. He hoped.
The blind test was Helen Marsh's idea. "We need to verify the Method before the exhibition," she'd said, as if it were routine. As if it weren't a test of everything Marcus had built his career on. He'd agreed, because refusing would have been worse. Refusing would have looked like doubt. And Marcus Webb did not doubt. Now, sitting in the museum's conference room with ten images in front of him, he wondered if that had been a mistake. The setup was simple. Ten images, displayed one at a time on a large monitor. For each image, Marcus would announce his assessment: human or AI-generated. Two museum staff members observed from the corner of the room, taking notes. "The images have been sourced from various platforms and private collections," one of them explained. "Some are known to be human-created. Some are known to be AI-generated. You won't be told which is which until the end." "Understood." "Ready?" Marcus took a breath. "Ready." The first image appeared. A portrait—a woman's face, rendered in a style that evoked classical oil painting. The brushstrokes were visible, the colors warm and varied. The eyes had a quality Marcus recognized immediately: the slight asymmetry that came from human observation. "Human," he said. They moved to the second. An abstract composition, all geometric shapes and bold colors. The precision was striking—each line perfectly straight, each angle mathematically exact. "AI," he said. The third. A landscape. Fourth. A still life. Fifth. A figure study. Marcus worked through them methodically, applying his method, trusting his instincts. By the eighth image, he was feeling confident. He'd seen patterns, recognized qualities, made assessments that felt right. The ninth image appeared. It was a small painting—maybe twelve inches square—of a dark room with a single window. Light spilled through in a way that seemed almost alive. The brushwork was loose, impressionistic, full of energy. Marcus studied it. There was something there. A quality he couldn't quite name. The kind of rawness that usually signaled human creation. But there was also a precision to the composition, a mathematical balance that felt... He hesitated. "Mr. Webb?" The observer's voice was neutral. "Human," he said finally. "I'm seeing human intention here. The light, the composition, the emotional weight—it feels authentic." The tenth image appeared. Another abstract piece, this one chaotic and raw. Marcus assessed it quickly. "AI," he said. "The chaos is too deliberate. It's mimicking human randomness without actually being random." After the test, the observers conferred quietly. Then one of them turned to Marcus. "The results are ready." He felt a tightness in his chest. "Of the ten images, you correctly identified four." The words didn't register at first. "Four?" "Four correct, six incorrect. That's within the range of random chance." Marcus stared at the monitor, where the images were now displayed with their true classifications. The portrait he'd identified as human had been AI-generated. The geometric abstract he'd called AI had been human. And the ninth image—the dark room with the light—that one had been AI too. "But I saw..." He stopped himself. "You saw what you expected to see," the observer said gently. "That's common. The human brain is wired to find patterns, even when they don't exist." Marcus felt something crack inside him. "Is this... will this be documented?" "All test results are confidential. For internal use only." Small mercies. But the damage was done. He knew the truth now, even if no one else did. His method was no better than guessing. He walked back to his office in a daze. The museum's halls, usually so familiar, felt alien now. Every painting on the walls seemed to mock him. How many of them had he misidentified over the years? How many times had he claimed to see "human soul" in work that was actually machine-generated? His phone buzzed. A message from Elena: Heard about the test. Want to talk? He didn't respond. That evening, there was a panel discussion at the museum: "Art in the AI Age: Authenticity, Authorship, and the Future of Creativity." Marcus was supposed to be a panelist. He considered canceling, but that would raise questions. So he sat on the stage, next to three other experts, and tried to focus on the moderator's questions. "The question of authenticity has never been more pressing," the moderator said. "Mr. Webb, you've built your career on the Webb Method. Can you tell us how it works?" Marcus cleared his throat. "The Webb Method is about... seeing. Looking beyond the surface to identify the human intention behind the work. It's not something that can be easily quantified or explained. It's an intuitive process, developed over decades of study." "So it's subjective?" "All art criticism is subjective. The difference is that I've trained my intuition to recognize patterns that others miss." "But can those patterns be verified? Can your method be tested?" Marcus felt the weight of the day's failure pressing down on him. "Any method has limitations," he said carefully. "The question isn't whether the method is perfect. The question is whether it provides value. And I believe it does." Elena, sitting in the audience, raised her hand. "Mr. Webb, I heard about the verification lab visit this morning. Word travels fast in this building." Her voice was calm but pointed. "How do you reconcile that with your claim to have 'trained intuition'?" The room went silent. Marcus felt his face heat. "I had an off day—" "An off day. Your entire career is based on this ability, and an 'off day' means it might not exist." "That's not—" "What if you've been wrong all along? What if the 'human soul' you claim to see is just projection? What if there's no difference to see?" The audience murmured. Marcus felt the walls closing in. "I've spent thirty years studying art," he said, his voice tight. "I've identified hundreds of works correctly. One test doesn't invalidate a career." "But it raises questions. Questions you seem unwilling to ask yourself." The moderator stepped in, steering the conversation to another panelist. But the damage was done. Marcus could feel the audience's eyes on him, could sense the shift in the room's energy. He'd been exposed. Not fully—not yet—but enough. The first crack had appeared. --- After the panel, he found Elena in the lobby. "Why did you do that?" She looked at him without expression. "Because someone needed to. You're building an entire exhibition on a method that doesn't work. You're promoting artists based on intuition that's no better than guessing. People trust you, Marcus. They believe you can see something they can't. And you're exploiting that trust." "I'm not exploiting anyone. I believe in what I do." "Do you? After today?" She shook her head. "I'm not trying to destroy you. I'm trying to wake you up. Before it's too late." "Too late for what?" "Before you stake everything on something that might not exist." She walked away, leaving him alone in the crowded lobby. That night, he sat in his apartment with Jimmy Okonkwo's portfolio open on his laptop. The work was good. He was sure of it. The rawness, the urgency, the quality that said human—it was there. But was it? Or was he seeing what he needed to see? What he wanted to see? The unknown number messaged him again: The cracks are showing, Mr. Webb. How long before the foundation collapses? He didn't respond. He closed the laptop and sat in the dark, trying not to think about the test, the panel, Elena's words. Trying not to think about the growing certainty that everything he'd built was a house of cards. And the wind was picking up.