Experts in artificial intelligence have expressed skepticism about the ability of large language models (LLMs) to achieve artificial general intelligence (AGI). While LLMs have shown rapid progress, they are considered by researchers such as Gary Marcus and Richard Sutton to be reaching a point of diminishing returns and a temporary focus in AI development. Sutton emphasized that future breakthroughs are more likely to come from scaling computational power rather than mimicking human cognitive processes. Predictions on the timeline for AGI vary, with some suggesting it could take 5 to 8 years if optimistic, or 10 to 15 years if more conservative, but all agree that major breakthroughs are necessary before AGI can be realized. The conversation highlights a need for diversified approaches in AI research beyond LLMs, which remain far from achieving true general intelligence.