New research shows that computers can become smarter if they mimic the way humans think.

Microsoft researchers have proposed a new AI training technique called “Algorithm of Thoughts” (AoT) that aims to improve the efficiency and human-like reasoning capabilities of large language models (LLMs) like ChatGPT. It’s one of the ways researchers are trying to harness human wits to boost artificial intelligence (AI).

“Machine learning algorithms are very good at identifying correlation, but not necessarily causation,” Alex Raymond, head of AI at Doppl.ai, a generative AI company, told Lifewire in an email interview. “AI can’t explain its reasoning the way a human would. Humans have a more grounded and symbolic understanding of the world, derived from both axiomatic wisdom and empirical learning.”

Microsoft researchers claim in their new paper that the new algorithmic method could be revolutionary because it “steers the language model toward a more efficient problem-solving path,” according to a published research paper. The technique uses “in-context learning,” which allows the model to explore different solutions in a structured way.