Your question
What does Grant Sanderson think AI means for mathematics?
Start with a question
01 Ask about podcasts
Name a person, topic, or episode and tell us what you want to know. You do not need to find the episodes yourself.
Podcast research, made simple.
Get answers
from podcasts.
Ask about a person, topic, or episode. We find relevant podcasts and write an answer to your question, with source links and timestamps so you can check what was said.
Free examples need no account. Your own answers require a plan.
How it works
You ask the question.
We research the answer.
Find a recommendation, understand someone's views, or compare what they said across episodes. Start with what you want to know.
Find relevant episodesOn one show or across several
Review what was saidCheck the available recordings
Write your answerInclude sources and timestamps
02 We research the episodes
We find relevant episodes, review what was said, and write an answer focused on your question.
Your written answerRead the answer.
Check the sources.
Check the sources.
Open source episode↗
03 Read and ask follow-ups
Check the source links and timestamps, then ask follow-up questions about the episodes reviewed. Your answer is saved privately in your account.
See an example
Here is what
an answer looks like.
This question about Grant Sanderson's views on AI and mathematics draws on 2 podcast episodes. Read the answer, then open a source link to check the context.
Read the full answer2source episodes
65source links
TLDR PodsExample answer
Grant SandersonContinue reading
What does Grant Sanderson think AI means for the future of mathematics?
Answer in brief
Grant Sanderson expects AI to become extraordinarily capable at mathematics, but he does not think an Olympiad medal, a famous proof, or any other single result will mark the arrival of general intelligence. The deeper transformation begins when machines can choose valuable questions, invent productive definitions, build explanatory theories, and compress sprawling results into intelligible ideas. He expects AI eventually to outperform most people at both proving and explaining mathematics. Human mathematicians would then shift toward judging significance, curating an overwhelming supply of machine-generated work, and connecting mathematics to worthwhile problems. His central uncertainty is not whether machines will produce more mathematics, but whether their output will be trustworthy, comprehensible, and directed toward anything people actually value. 6:259:2028:3535:0037:201:25:45
Free examples
Real questions. Published answers.
Find the part you need
Get to the answer.
Keep the context.
Looking for a book someone recommended? Want to understand their advice on building a business or changing a habit?
Ask your question instead of searching through hours of audio. Read the answer and use the source links to listen to the relevant parts yourself.
About TLDR PodsUse your answer in other apps
Bring the answer into your next project.
Use Codex, Claude Code, or Cursor? Open a finished answer there, with its source links included, and use it in your own work. You can also copy or print your answer.
Codex
Claude Code
Cursor
Choose how many answers you need.
Pro includes 10 answers a month. Research includes 40. Both include source links and follow-up chats. Each new research question uses one answer, even when it uses several episodes. Follow-ups do not use this allowance.
For occasional questions
Pro
Start here
Get written answers to your podcast questions, with links to the episodes.
£15/mo
billed monthly
- Monthly allowance
- 10 answers
- If you use every answer
- £1.50 / answer
Eligible new customers get 7 days free, then the monthly price shown in Stripe. Card required. Cancel anytime.
For frequent questions
Research
Ask more questions each month, with the same features as Pro.
£39/mo
billed monthly
- Monthly allowance
- 40 answers
- If you use every answer
- £0.98 / answer
Eligible new customers get 7 days free, then the monthly price shown in Stripe. Card required. Cancel anytime.
Prices per answer assume you use the full monthly allowance at the listed GBP price. Failed requests do not use your allowance.
Every paid plan includes
- Research a person, topic, or episode
- Up to 8 relevant episodes considered per question
- Privately saved answers and follow-up chats
- Episode and timestamp source links
- Copy, print, or save your answers
- Open answers in Codex, Claude Code, or Cursor
- API access
What do I get?
A written answer to your question, based on relevant podcast episodes. It includes source links and timestamps so you can check what was said. Your answer is saved in your account, where you can read it again or ask follow-up questions.
Can I read something before I sign up?
Yes. Our example answers are free to read without an account or a card. Creating your own answers requires a current paid plan or an eligible introductory trial.
Does it work across different podcasts?
Yes. Ask about a person or topic across podcast appearances. We consider relevant episodes, subject to source availability and your plan’s episode limit.
Who can see my answers?
Only you. Your answers and follow-up conversations stay in your account. They are never added to our public answer collection.
Will I get the original transcript?
No. You get an answer written for your question, with source links and timestamps. Open a source episode to hear the surrounding conversation. We do not provide transcripts.
Your next question
What do you want to know
from podcasts?
Ask a question A written answer, with source links and timestamps.