Google’s latest AI flex came with Silicon Valley’s new favorite word

• Sundar Pichai touted a 7x increase in Google’s AI products at the I/O developer conference today.
• The stunning increase drew audible gasps from the audience.
• Pichai cracked a joke about ‘tokenmaxxing,’ but said the stat still tells an important story.

Google CEO Sundar Pichai isn’t afraid of touting big AI usage numbers, despite the ‘tokenmaxxing‘ controversy.

At Google’s annual developer conference today, Pichai said that monthly usage of its AI products has increased sevenfold to 3.2 quadrillion tokens since last year.

The huge increase drew audible gasps from the packed Shoreline Amphitheater near Google’s Silicon Valley headquarters — and a joke from Pichai.

“Some out there might call it tokenmaxxing, and there’s probably some truth to it,” Pichai said to some laughter from the audience. “I still think it tells an important story about our products and how others are building on it as well, especially our developers.”

Tokens are the basic building block of AI chatbots, with one ‘token’ making up about ¾ of a word. Tokenmaxxing, or flexing about how many tokens you’re using, has become a lightning rod in tech this year, as some argue developers are using up unnecessary tokens just to brag about it.

Despite the controversy, the clear use of Google’s AI products is surging, something Pichai was keen to highlight.

“I never imagined I’d say quadrillion at a keynote, but here we are,” he said.

Google has been seen as a winner on the AI front in recent months, thanks to its Gemini 3 AI model and AI chips known as TPUs. Its stock has more than doubled since last year’s I/O.