Simon Willison’s Weblog

Subscribe

When I wrote about how good ChatGPT with GPT-5 is at search yesterday I nearly added a note about how comparatively disappointing Google's efforts around this are.

I'm glad I left that out, because it turns out Google's new "AI mode" is genuinely really good! It feels very similar to GPT-5 search but returns results much faster.

www.google.com/ai (not available in the EU, as I found out this morning since I'm staying in France for a few days.)

Here's what I got for the following question:

Anthropic but lots of physical books and cut them up and scan them for training data. Do any other AI labs do the same thing?

Screenshot of search results page with navigation tabs "AI Mode" and more. Main search query reads "Anthropic but lots of physical books and cut them up and scan them for training data. Do any other AI labs do the same thing?" Results show "11 sites" and include text: "In 2024, Anthropic paid millions of dollars to buy physical books in bulk, which they then cut up, scanned, and discarded to train their AI models. This method was a major component of a copyright lawsuit settlement that Anthropic paid in September 2025. While Anthropic took this destructive, large-scale approach, other major AI labs have taken different routes to acquire book data." Followed by "Here's how other AI companies have sourced book data:" with bullet points for "OpenAI and Microsoft: In June 2025, the companies announced a partnership with Harvard University libraries to train AI models on digitized, public-domain books. This non-destructive method provides high-quality text while preserving the original physical copies." "Meta: Meta has been accused of relying on large-scale piracy to build its datasets for models like LLaMA. In 2025, a search tool revealed that Meta had copied more than 7.5 million books from the pirate site LibGen for its AI systems." "Google: Decades before the current AI boom, Google pioneered non-destructive scanning methods for its Google Books project. This method became a successful legal precedent for digitizing copyrighted works for internal purposes. In 2024, Anthropic hired the former head of this project, Tom Turvey, to help with its own book-acquisition strategy.

I'll be honest: I hadn't spent much time with AI mode for a couple of reasons:

  • My expectations of "AI mode" were extremely low based on my terrible experience of "AI overviews"
  • The name "AI mode" is so generic!

Based on some initial experiments I'm impressed - Google finally seem to be taking full advantage of their search infrastructure for building out truly great AI-assisted search.

I do have one disappointment: AI mode will tell you that it's "running 5 searches" but it won't tell you what those searches are! Seeing the searches that were run is really important for me in evaluating the likely quality of the end results. I've had the same problem with Google's Gemini app in the past - the lack of transparency as to what it's doing really damages my trust.

Posted 7th September 2025 at 10:08 am

Recent articles

Monthly briefing

Sponsor me for $10/month and get a curated email digest of the month's most important LLM developments.

Pay me to send you less!

Sponsor & subscribe