People and sources

Not mathematics-specific; retained as a labeled comparison.

Into the Impossible interview with Terence Tao on AI in mathematical research, reliability, verification, and proof assistants.

Curriculum reform and computational thinking.

Theories of Everything interview with mathematical physicist Yang-Hui He on AI-assisted discovery, formal proof, LLM limits, and changing research practice.

AI, mathematical understanding, and conceptual frameworks.

Mathematics-specific proposal for testing meaningful AI contribution.

Humanistic argument; direct page access was limited during initial review.

Marcus argues that LLMs can fail to generalize recursive reasoning tasks and that plausible output is not robust abstraction. Directly relevant to mathematical reasoning, but not a systematic mathematics-education study.

Use a specific talk or transcript if added in a revision.

Future-facing assessment reported by secondary journalism.

Lanier identifies immersive mathematical visualization as a promising educational use of VR, but the source is not about current generative AI.

Formalization and responsible response to AI.

Quanta Magazine report and podcast synthesizing mathematicians' views of AI-assisted proof, verification, creativity, and changing research practice.

Governance and policy proposal.

Submitted through the site scoring tool by James B. Wilson. Scores are provisional heuristic estimates pending review.

Argues for total opposition to AI use in mathematics.

Dedicated explainer on claims of AI mathematical breakthroughs. Full public transcript was unavailable at review, so the assessment is intentionally conservative and should be revised when a citable transcript is available.

Mindscape interview with science journalist Anil Ananthaswamy on the mathematical structures of neural networks, sample efficiency, and limits of current models.

Argues for computation-augmented generation, combining LLMs with exact computation and current knowledge for mathematics and other technical domains.

Quanta's The Joy of Why interview with Emmanuel Candès on black-box models, uncertainty, prediction, and trustworthy statistical inference.

Conditional analysis of mathematical goals and practice.

Reports a recent working paper on Chinese secondary students.

Add a specific post when reviewing its scores.

Secondary reporting; avoid treating as independent evidence.

Not mathematics-specific; included only as a broad AI comparison.