YouTube Algorithm Basics (Cristos Goodrow, VP Engineering at Google) | AI Podcast Clips

Lex Fridman interviews Cristos Goodrow about the foundations of the YouTube recommendation and search systems. Goodrow explains collaborative filtering and how watching videos together creates a 'related graph' that naturally clusters videos by language and topic. They discuss the evolution of quality signals from views, to watch time, to post-watch satisfaction surveys, plus likes, shares, subscribes, and dismissals. Goodrow also covers clickbait, gaming the system, A/B testing with hundreds of variables, and the idea that the 'algorithm' is really code married with the collective behavior of all viewers.

• The first real attempt at 'watch next' recommendations used collaborative filtering, observing which videos get watched close together by the same person.
  00:01:32
• The resulting 'related graph' automatically groups videos by language and topic without anyone programming language detection.
  00:02:36
• YouTube moved from optimizing for views to watch time because views can be inflated by misleading clicks people quickly abandon.
  00:10:28
• YouTube surveys users the next day, asking whether they'd rate a video four or five stars, and feeds that into the ML system.
  00:11:31
• Content analysis of video itself is still 'somewhat crude' - it can tell soccer is being played but not whether it's Manchester United.
  00:15:46
• What outsiders call 'the YouTube algorithm' is really code plus the live behavior of every viewer; without people the algorithm wouldn't work.
  00:32:38
• An early diversity rule was a simple heuristic: no more than three videos in a row from the same channel.
  00:34:11

• A researcher couldn't believe YouTube recommended English academic videos but Turkish baklava videos, matching her bilingual habits automatically.
  00:03:40
• YouTube sees each user essentially as the full history of videos they've watched, like a 'DNA strand' or a vector in video space.
  00:06:48
• Some users don't subscribe because they think it costs money; others subscribe just to give a creator a 'high five' without watching more.
  00:12:34
• A major World of Warcraft livestream was hard to find because the title was just 'match 478 - A team versus B team,' not the game name.
  00:16:48
• If thumbnails are too racy or have too many exclamation points, YouTube suppresses them because users report being offended.
  00:24:10
• Nearly every change to YouTube is shipped only after an A/B experiment running one week to months, measuring literally hundreds of variables.
  00:35:15
• The simplest success signal is whether a user returns on another day - people don't come back to things they don't value.
  00:29:28
• Some people and machines deliberately try to associate unrelated videos together because it's profitable, so systems must resist gaming.
  00:21:31

Books, products and media the guest or host genuinely endorsed here — with the buy link.

Affiliate link — we may earn a commission at no extra cost to you.