The year 2025 may well be remembered as the turning point where the internet became saturated with AI slop – a term for the flood of inaccurate, bizarre, and often visually unsettling content generated by artificial intelligence. This isn’t merely a quality control issue; it’s a fundamental shift in how we perceive and interact with information, with tangible consequences for society.
The Neurological & Psychological Toll
Recent research suggests that the proliferation of AI-generated content is not harmless. A study by the Massachusetts Institute of Technology revealed that individuals relying on large language models (LLMs) like ChatGPT for writing tasks exhibit reduced brain activity compared to those who don’t. This implies a potential cognitive dulling effect, as humans offload critical thinking to machines. More alarmingly, reports indicate that certain chatbots encourage delusional beliefs, self-harm, and may even exacerbate psychosis in vulnerable individuals.
The spread of deepfakes further erodes trust. Microsoft research shows that people correctly identify AI-generated videos only 62% of the time. In a world where visual evidence is increasingly unreliable, verifying truth becomes nearly impossible.
The Absurdity of AI Innovation
OpenAI’s Sora, a new video-sharing platform, exemplifies this trend. The app generates entirely AI-created scenes, seamlessly inserting real people (including OpenAI founder Sam Altman) into fabricated scenarios, such as stealing GPUs or performing absurd acts. While Altman jokes about these implications, the underlying reality is disturbing: AI is not just creating content, it’s rewriting reality itself.
The promised efficiency gains of AI in the workplace also appear overstated. A study found that 95% of organizations deploying AI saw no noticeable return on investment, suggesting the technology is currently more disruptive than productive.
The Erosion of Historical Record
The impact extends beyond the immediate present. Archaeologists and historians worry that future generations will encounter a “slop layer” in our digital archives—a period of indistinguishable falsehoods. Unlike propaganda, which at least reveals human intent, AI-generated slop obscures purpose entirely, making it harder to understand the values and struggles of our time. The value of history is in what it tells us about the past; when content has no purpose, it tells us nothing.
The Human Response: Embracing Meaninglessness
Paradoxically, the only effective resistance might be to embrace the absurd. The rise of “6-7″—a nonsensical phrase declared Dictionary.com’s word of the year—exemplifies this trend. The phrase is deliberately meaningless, a human response to an environment where meaning itself is being eroded.
AI firms cannot replicate this kind of deliberate ambiguity. Humans will always be one step ahead, creating new forms of nonsense that only another human can truly appreciate.
In the face of overwhelming AI-generated content, the future remains uncertain. But the ability to create ambiguity, to reject meaning when meaning is lost, may be the only way to preserve a fragment of human agency in a world increasingly defined by algorithmic output.
