Modern AI models possess hidden capabilities that emerge suddenly and consistently during training, but these abilities remain concealed until prompted in specific ways, according to new research from Harvard and the University of Michigan.

The study, which analyzed how AI systems learn concepts like color and size, revealed that models often master these skills far earlier than standard tests suggest—a finding with major implications for AI safety and development.

"Our results demonstrate that measuring an AI system's capabilities is more complex than previously thought," the research paper says. "A model might appear incompetent when given standard prompts while actually possessing sophisticated abilities that only emerge under specific conditions."

This advancement joins a growing body of research aimed at demystifying how AI models develop capabilities.

Anthropic researchers unveiled "dictionary learning," a technique that mapped millions of neural connections within their Claude language model to specific concepts the AI understands, Decrypt reported earlier this year.

While approaches differ, these studies share a common goal: bringing transparency to what has primarily been considered AI's "black box" of learning.

"We found millions of features ...