We need fewer prompting frameworks and more "AI intuition"

We're approaching AI backwards. The industry obsesses over prompting frameworks while missing something fundamental: LLMs aren't traditional software tools.

They're alien intelligences that have absorbed the entirety of human culture. Not in some science fiction sense, but in the very real way they interpret language and meaning through processes no one fully understands—not even their creators.

While Claude and ChatGPT feel like friendly conversation partners, each response emerges from statistical predictions flowing through thousands of mathematical dimensions, all of which shift with context.

They're inscrutable and unlike any software we've encountered before.

Beyond Technical Frameworks

System prompts, fine-tuning, and LLM wrappers work well for specific tasks, but they don't help you sense the model's deeper patterns. This might sound abstract, but developing "AI intuition" becomes increasingly valuable as models grow more powerful and opaque.

Consider ChatGPT 4.5: measurably worse on benchmarks yet far more creative than its predecessors. When pressed to explain this, even the labs building these systems retreat into circular explanations about "pre-training" and "reasoning capabilities."

Meanwhile, users cut straight to what matters: "the vibe is different." That's not imprecision—it reveals something crucial about this technology and the skills we need going forward.

After hundreds of hours across ChatGPT, Claude, Deepseek, and others, spanning countless use cases, one pattern emerges clearly: effective collaboration with these alien intelligences demands more than technical competence. It requires intuitive wisdom grounded in philosophy, culture, and art.

These models encode human meaning at unprecedented scale. The new essential skill becomes sensitivity to how meaning shifts across contexts—a form of dimensional literacy that lets you navigate between statistical gravity and creative possibility.

Introducing Thoughtform

This is why I'm developing Thoughtform: a creative philosophy focused on cultivating AI intuition among designers, writers, philosophers, and artists.

Not to replace technical skills, but to complement them with the navigational awareness these tools actually require. Whether AGI arrives or not, AI's economic and societal impact demands immediate attention.

What matters most right now is teaching creative professionals to bridge human meaning and machine intelligence. Because beneath every model's surface lies a multidimensional creative wilderness.

Technical frameworks can build the tools, but only intuition can guide you through the territory—showing you not just what the machine can generate, but where it should wander and why certain directions lead to wonder rather than statistical mediocrity.