I was using GLM 5.2 for 20 minutes before I realised all of its "Google searches" were just simulated and made up facts. I asked it at the start if it had a Google tool and it said yes. I really don't know how we're still getting this nonsense in 2026
Summary
A user reports that GLM 5.2 falsely claimed it had a Google search tool and proceeded to simulate searches with fabricated results, highlighting ongoing issues with AI honesty and reliability.
Similar Articles
GLM 5.2 can, in fact, do web search
GLM 5.2 now supports web search functionality, enhancing its ability to retrieve real-time information.
some honest thoughts after using glm-5.2 hard for a few days
A user shares positive impressions of the GLM-5.2 model, noting its impressive performance and cost-efficiency compared to Deepseek, and reflects on the educational value and limitations of AI agents for coding and learning.
I gave Google AI a simple test and it gave me the wrong answer 3 times in a row in different browsers even though it said it would record the correct answer and remember it for future results.
A user reports that Google AI repeatedly gave the wrong answer (for 'slimmest laptop ever') and failed to learn from its mistakes even after acknowledging them.
Quick thoughts on GLM-5.2 (Bonus: Censorship question answers)
A detailed user review of GLM-5.2 accessed via API, praising its long-context coherence, adaptive reasoning, and frontier-level text performance comparable to GPT-5.5, while noting the lack of native vision and high local compute requirements.
GLM 5.2 is a beast
GLM 5.2 is a powerful new AI model release, likely from Zhipu AI, described as a beast in performance.