Overall, we are seeing query growth, including across Apple's devices, and the growth from AI overviews is healthy and continuing over time. YouTube has continued to grow despite the rise of TikTok, as people are engaging more and we are improving our products.

Our primary metric is the number of queries per day, and we analyze weekly and monthly growth rates to understand our performance better.

With the rise of GDPR, companies are shifting budgets from paid advertising to organic SEO efforts, as tracking users has become more challenging.

Transformers, particularly BERT and MUM, have significantly improved search quality, leading to new use cases and query types.

We initially focused on searching over specific verticals or datasets, but pivoted to a more general search approach when we realized the limitations of that strategy.

AI Overviews have scaled to over 1.5 billion users every month in more than 200 countries, driving user satisfaction and increased search frequency.

People are consuming a lot more information, and there are more creators putting out content, making this an expansionary moment for the web.

The web is evolving profoundly, with the number of web pages increasing by 45% in the last two years, indicating a significant growth in content availability.

At Exa, we apply a scaling hypothesis for search, believing that increasing data and compute will enhance our search engine's capabilities, unlike traditional search engines that have stagnated.