We All Have a "Serious Work" AI and a "Just Vibing" AI. When Did That Happen?

A recent article on Dev.to explores the unconscious hierarchy users develop when interacting with various AI models. The author notes that they have naturally categorized different AI tools based on their specific strengths and limitations rather than general intelligence. For instance, Claude Opus is reserved for complex coding tasks, while Gemini is preferred for calculation-heavy problems and real-time information retrieval. ChatGPT is utilized for quick, low-stakes queries. The author realized this system existed when an overly complex AI agent failed at a simple task, highlighting that effective AI usage is about matching the tool's capabilities to the specific requirements of the job. The piece concludes by inviting readers to reflect on their own subconscious AI routing systems and the moments that defined their personal workflows, emphasizing that these hierarchies are often built through trial and error to minimize frustration.