Fei-Fei Li Warns AI's Biggest Risk in Schools Isn't Cheating—It's Eroding Student Motivation

AI Pioneer Identifies Core Threat to Student Learning

Fei-Fei Li, Stanford computer science professor and AI pioneer widely recognized as the Godmother of AI, has identified what she believes is AI's biggest risk in schools—and it's not the cheating epidemic that has dominated headlines. Speaking on the Huberman Lab podcast released Monday, Li warned that the technology's most dangerous impact is the erosion of students' motivation and agency to learn1

1

2

2

. "The absolute bad outcome is that our young generation, their agency and human-level motivation of learning and living is taken away by tools," Li stated. "It should not be taken away by humans nor should it be taken away by machines."1

1

The Real Cost of Over-Reliance on AI Tools

While cheating has captured most attention since ChatGPT went mainstream—reaching endemic proportions by August 2024—Li argues students losing the desire to learn represents a far graver threat1

1

. The ImageNet inventor expressed concern that graduates who rely completely on AI in schools without actively engaging their own thinking risk leaving education without having "properly developed the brain"1

1

2

2

. This cognitive development crisis stems from a fundamental question many students now ask: why bother learning something when AI provides instant answers?1

1

Research Reveals Troubling Patterns in AI-Assisted Learning

Emerging evidence supports Li's concerns about AI's real school risk is lost motivation and depth. Oxford University Press research found students gaining speed while losing depth of thinking when using AI tools2

2

. MIT researcher Nataliya Kosmyna discovered people given generative AI for essay writing performed worse over time compared to those using Google or no assistance—though four researchers later questioned aspects of this study's methodology, sample size, and EEG analysis2

2

. Vivienne Ming, chief scientist at the Possibility Institute, told Business Insider that most AI users she studied were deploying the technology to avoid thinking independently1

1

2

2

. Wharton researchers have labeled this pattern "cognitive surrender," observing it affects workplaces too, where juniors never learn essential skills like debugging2

2

.

Rejecting Both Extremes: Bans and Unchecked Access

Li is equally wary of blanket bans and unrestricted access to AI tools. "Both things worry me," she explained. "Either denying the tool or taking away agency and motivation."2

2

The Stanford professor believes AI in schools can genuinely help struggling students who are already engaged in learning. She recalled her own experience battling organic chemistry as a premed student, when teaching assistants and professors lacked sufficient time to answer every question. "I know where I'm stuck," Li said. "I have the motivation to learn. I just need guidance."2

2

This distinction—between students seeking deeper understanding versus those avoiding thought entirely—should shape educational AI policy1

1

.

Emerging Approaches to Balance AI Access and Critical Thinking

Several initiatives attempt to address students losing the desire to learn while preserving AI's benefits. OpenAI introduced a study mode specifically designed to encourage critical thinking rather than simply delivering answers1

1

. Computer science professors have shifted focus from coding syntax to higher-level problem-solving. The UK government has backed AI-assisted homework, arguing it could particularly benefit children struggling in traditional learning environments1

1

. Li advocates for finding ways to preserve student motivation while providing access. "Let's find a way to give them the access and the right way of using these tools," she urged1

1

. She believes proper implementation could make future students "way smarter than us because they are superpowered"1

1

2

2

.