Does AI already have human-level intelligence? The evidence is clear

"Three-quarters of a century later, the answer looks like 'yes'. In March 2025, the large language model (LLM) GPT-4.5, developed by OpenAI in San Francisco, California, was judged by humans in a Turing test to be human 73% of the time - more often than actual humans were. Moreover, readers even preferred literary texts generated by LLMs over those written by human experts."

"This is far from all. LLMs have achieved gold-medal performance at the International Mathematical Olympiad, collaborated with leading mathematicians to prove theorems, generated scientific hypotheses that have been validated in experiments, solved problems from PhD exams, assisted professional programmers in writing code, composed poetry and much more - including chatting 24/7 with hundreds of millions of people around the world."

"Yet many experts baulk at saying that current AI models display artificial general intelligence (AGI) - and some doubt that they ever will. A March 2025 survey by the Association for the Advancement of Artificial Intelligence in Washington DC found that 76% of leading researchers thought that scaling up current AI approaches would be 'unlikely' or 'very unlikely' to yield AGI (see go.nature.com/4smn16b)."

Alan Turing proposed the imitation game in 1950 to ask whether machines could display flexible, general cognitive competence characteristic of human thought. In March 2025, GPT-4.5 was judged human 73% of the time in a Turing test, and readers preferred some LLM-generated literary texts to human experts'. LLMs have achieved high-level results across mathematics, science, coding, and creative writing, and they interact continuously with hundreds of millions of people. Many researchers remain reluctant to label current models as artificial general intelligence. The reluctance is partly conceptual, with ambiguous AGI definitions, and partly emotional, tied to fear of displacement.

Read at Nature