You won't be talking to AI in natural language for too long (www.elicited.blog)

The latest discourse on AI interaction highlights a critical evolution: while natural language processing (NLP) makes AI interfaces accessible to all, true efficacy and power will increasingly depend on domain-specific, expert language. As tasks become more complex, high-level users will leverage concise, precise terminology to convey intent, leading to more effective interactions with AI systems. This shift reflects an ongoing journey to minimize the "semantic gap"—the distance between user intentions and machine understanding—which has historically been a focal point in programming and system design. The article draws parallels to the RISC versus CISC processor architecture debates, illustrating that the future of AI will not just be about simpler natural language prompts but also about moving to a higher "semantic altitude." Experts will need to define requirements with greater specificity, avoiding ambiguous or verbose language. As generative AI tools improve, the focus will transition from mastering syntax to crafting clear, high-level specifications that accurately convey intent. Consequently, while NLP democratizes the ability to generate outputs, it simultaneously underscores the enduring value of domain expertise—where the ability to formulate precise queries will drive the distinction between novice users and seasoned professionals.