DEVOURED
How AI does, and does not, change the way I do math
Mathematician Terence Tao argues that AI will accelerate mathematical research, but human collaboration remains vital for the social and structural aspects of the field.
What: Mathematician Terence Tao and guest author Rachel Webb discuss the impact of LLMs on math research. They suggest that while AI makes the process of drafting and searching for lemmas cheaper, the core value of math—finding 'interesting' problems and building community—remains a human-centric activity.
Why it matters: This perspective emphasizes that 'fluency' in generating text and code is becoming a commodity, shifting the primary role of the mathematician from drafting to deciding what is strategically or artistically worth pursuing.
Deep dive
- Mathematical writing is a proxy for clarity of thought; automating the writing process creates a risk of obscuring poor definitions.
- AI speed-ups allow mathematicians to explore 'side roads' of research that were previously too time-consuming, while retaining the freedom to choose manual exploration for deeper understanding.
- The social fabric of math (tea-times, conferences, collaborations) faces a risk of atomization if researchers choose to consult machines over colleagues.
- There is a growing concern that research is turning into a 'black box' as the most powerful models, and their underlying prompts, remain proprietary and opaque.
Original article
Full article content is not available for inline reading.
Explore all 57 articles in today's digest
Fresh dev news, AI-digested daily. Never miss a trend at devoured.fyi