AI accelerates scientific research but risks narrowing academic scope

Research indicates that while artificial intelligence significantly accelerates scientific output, it may also be narrowing the scope of academic inquiry. Studies show that researchers using AI tools publish approximately three times more papers and receive nearly five times more citations than those who do not. However, AI-driven research tends to have a 4.6 percent narrower thematic range compared to non-AI research, a pattern observed in over 70 percent of studied scientific fields.

The issue is attributed to institutional incentive structures rather than the technology itself. AI increases the capacity for exploration, but current academic systems reward researchers for focusing on deep, well-defined problems that are easy to evaluate, rather than venturing into unknown territories. This tendency accelerates a long-term trend of decreasing innovation and disruption seen over the last six decades.

To mitigate this, experts suggest reforming research funding to reward the creation of new datasets and new measurement capabilities rather than just the exploitation of existing data. While AI can lead to breakthroughs—such as Google’s GNoME tool identifying 381,000 potentially stable mineral crystals—the current academic framework may inadvertently encourage researchers to follow familiar paths because they are faster and cheaper.