Home
Official Confirms AI-Generated Image of Pedestrian with Brooms Was Pulled From Circulation

An unusual public service advertisement on a notice board in Linfen, Shanxi, has drawn public attention due to its bizarre imagery. Pedestrians appear oddly positioned, and a broom seems to emerge from a sanitation worker’s garbage bag. Distorted details suggest AI generation, prompting local authorities to dispatch personnel on August 26 for immediate action. The management committee’s construction department confirmed an emergency inspection on the evening of the 25th, followed by the prompt removal of the problematic advertisement.
Text-focused review overlooked visual errors
On August 27, local officials confirmed the ad was created by a partner advertising firm using AI tools and has since been taken down. Staff acknowledged that the review process prioritized textual slogans, failing to verify visual accuracy. The area is now conducting a comprehensive audit of all outdoor advertisements, with plans to replace any similar flawed content immediately.
The absurd visuals highlight AI’s inability to grasp real-world logic. Large language models learn pixel patterns, styles, and elements from vast datasets but lack understanding of physical common sense, human anatomy, or logical object relationships. They combine elements based on frequency without assessing whether postures or placements make sense. Simple prompts often lead to forced stacking of elements, resulting in distorted limbs and other “AI hallucinations.” When these illogical errors bypass human oversight, they manifest as physical advertisements.
Related article
Musk Says 99% of SpaceX\'s Value Will Come From AI Rockets, Space Exploration Becomes Side Business
SpaceX has shared a video of Elon Musk’s recent address to staff. As he traced the company’s progress, he laid out for the first time the three core pillars of Starship, Starlink, and AI, with artificial intelligence taking center stage. Musk emphasi
Universal Robots Unveils Gen 7 for Physical AI Deployment
Will Healy, Director of Product & Industry Marketing at Universal Robots, credited for the imageWill Healy, Director of Product & Industry Marketing at Universal Robots, emphasizes that following the Gen 7 release, AI-driven robots must prioritize ad
GPT 5.5 Leads AI Security Race, DeepSeek Tops Cost-Efficiency
Kasra Rahjerdi, a security researcher, recently published a significant report detailing practical tests on the security reasoning of major large language models. He achieved this by constructing a deliberately vulnerable book review application. In
Related Special Topic Recommendations
Comments (0)
0/500

An unusual public service advertisement on a notice board in Linfen, Shanxi, has drawn public attention due to its bizarre imagery. Pedestrians appear oddly positioned, and a broom seems to emerge from a sanitation worker’s garbage bag. Distorted details suggest AI generation, prompting local authorities to dispatch personnel on August 26 for immediate action. The management committee’s construction department confirmed an emergency inspection on the evening of the 25th, followed by the prompt removal of the problematic advertisement.
Text-focused review overlooked visual errors
On August 27, local officials confirmed the ad was created by a partner advertising firm using AI tools and has since been taken down. Staff acknowledged that the review process prioritized textual slogans, failing to verify visual accuracy. The area is now conducting a comprehensive audit of all outdoor advertisements, with plans to replace any similar flawed content immediately.
The absurd visuals highlight AI’s inability to grasp real-world logic. Large language models learn pixel patterns, styles, and elements from vast datasets but lack understanding of physical common sense, human anatomy, or logical object relationships. They combine elements based on frequency without assessing whether postures or placements make sense. Simple prompts often lead to forced stacking of elements, resulting in distorted limbs and other “AI hallucinations.” When these illogical errors bypass human oversight, they manifest as physical advertisements.
Musk Says 99% of SpaceX\'s Value Will Come From AI Rockets, Space Exploration Becomes Side Business
SpaceX has shared a video of Elon Musk’s recent address to staff. As he traced the company’s progress, he laid out for the first time the three core pillars of Starship, Starlink, and AI, with artificial intelligence taking center stage. Musk emphasi
Universal Robots Unveils Gen 7 for Physical AI Deployment
Will Healy, Director of Product & Industry Marketing at Universal Robots, credited for the imageWill Healy, Director of Product & Industry Marketing at Universal Robots, emphasizes that following the Gen 7 release, AI-driven robots must prioritize ad
GPT 5.5 Leads AI Security Race, DeepSeek Tops Cost-Efficiency
Kasra Rahjerdi, a security researcher, recently published a significant report detailing practical tests on the security reasoning of major large language models. He achieved this by constructing a deliberately vulnerable book review application. In











