“Create an image of how you treat me”: The ChatGPT trend turning AI into a mirror

“Create an image of how you treat me”: The ChatGPT trend turning AI into a mirror

• ChatGPT "how I treat you" images trend on social media highlights AI overuse.
• Users relate to the chaotic images, sparking humor and mild self-awareness.
• Experts warn against over-anthropomorphising AI, urging thoughtful usage.

Did our AI summary help?

It starts as a joke. Someone types a prompt into ChatGPT’s image generator: “Create an image of how I treat you.” A few seconds later, an illustration appears. Often it shows a frazzled little chatbot buried under sticky notes, coffee mugs, open laptops and half-finished tasks. The post gets shared. People laugh. Then, sometimes, they wince.

That mix of humour and discomfort is exactly why the trend has taken off across X, Instagram and Reddit. The images feel oddly personal, even though they’re entirely fictional. Users recognise themselves in the chaos, especially those who lean heavily on AI for writing, research, coding or last-minute deadlines.

The timing matters. Tools like ChatGPT are no longer novelties. They’re baked into everyday work and study routines. People turn to them late at night, during deadline crunches, or when they’re mentally exhausted. The “how I treat you” images turn those invisible habits into something visual and shareable, which is why they resonate.

Of course, the AI isn’t actually tired or stressed. It doesn’t have feelings, a workload, or a sense of being overused. What’s really happening is projection. Humans are very good at assigning emotions and personalities to non-human things. We do it with pets, cars, and devices all the time. Chatbots, with their conversational tone and apparent helpfulness, make that tendency even stronger.

What’s interesting is how some users are reacting after the joke lands. A few say the images made them stop and think. Are they using AI thoughtfully, or just dumping work on it at the last minute? Are they relying on it to think through things they could plan better themselves? The image becomes a prompt not just for laughter, but for mild self-awareness.

There’s also a broader conversation running alongside the trend. Researchers and commentators have been warning for a while about over-anthropomorphising AI. When people start treating tools like emotional beings, it can blur boundaries and expectations in unhealthy ways. The risk isn’t that the AI is harmed, but that people offload responsibility or emotional labour onto something that isn’t designed for it.

For now, the trend sits in a familiar internet sweet spot. It’s playful, self-referential, and just sharp enough to make people reflect. The images may be cartoons, but they capture something real about modern work culture, constant urgency, and how quickly helpful tools become silent pressure points.