Over 8.5 million developers use Google AI models monthly, plans $190 billion capex: Sundar Pichai
New Delhi, May 20
US-based Alphabet Inc.-backed technology firm Google CEO Sundar Pichai has said that more than 8.5 million developers are now building apps and experiences every month using its artificial intelligence models.
On the AI infrastructure investment front, Pichai said Google expects its capital expenditure to reach nearly $190 billion this year, largely directed toward AI data centres, custom silicon and model training.
Pichai further said the company is witnessing rapid growth across its AI ecosystem, with demand increasing among developers, enterprises and consumers.
According to the California-headquartered firm, its model APIs are now processing nearly 19 billion tokens per minute, while more than 375 Google Cloud customers processed over one trillion tokens each during the past year.
Highlighting the growing adoption of Google's AI products, Pichai said the company's Gemini app has crossed 900 million monthly active users, more than doubling from last year. AI-powered search features are also seeing strong traction, with AI Overviews reaching over 2.5 billion monthly users and AI Mode surpassing one billion monthly active users.
The company also unveiled a range of new AI tools and upgrades, including Gemini 3.5 Flash, a faster AI model focused on coding and real-world workflows, and Gemini Spark, a new personal AI agent designed to perform tasks on behalf of users.
Google further introduced updates to its AI infrastructure, including its latest generation of Tensor Processing Units (TPUs), aimed at supporting faster AI training and inference at scale.
Pichai said Google remains committed to its AI-first strategy, describing AI as the most transformative technology for advancing the company's mission and improving lives globally.
Additionally, Google introduced 'Ask YouTube' - an AI-powered feature designed to help users find the most relevant moments in videos through conversational queries - and 'Docs Live', which allows users to create and organise documents using voice-based prompts.
Moreover, the company expanded its AI transparency initiatives through SynthID, its watermarking technology for AI-generated content. It said more companies, including OpenAI, Kakao and Eleven Labs, are adopting the technology to improve verification of AI-created images, videos and audio.
— IANS
Reader Comments
Great to see an Indian-origin CEO leading this AI revolution! Pichai's vision is clear—AI is the future. But I hope some of this investment also reaches Indian startups and academia. We have so much talent, but the infrastructure gap is real. Gemini 3.5 Flash for coding sounds exciting for our engineering colleges! 👩💻
Bro, 900 million monthly active users for Gemini? That's insane growth! I've been using it for translation and coding help—works well for basic stuff. But sometimes it struggles with Indian regional languages. Google should focus on Hindi, Tamil, Telugu support. Make AI accessible to the masses, not just English speakers.
As a developer in Bangalore, I can confirm Google's AI tools are becoming essential. The 19 billion tokens per minute processing is mind-boggling. However, pricing for API calls needs to be reasonable for Indian devs—dollar rates hurt us. But credit where due: SynthID watermarking is a needed step against deepfakes.
'Ask YouTube' feature sounds useful! As someone who watches cooking and DIY tutorials all the time, finding exact moments in long videos is a pain. But privacy concerns? Google already knows too much about us. At least they're transparent with SynthID. Let's hope this doesn't become another data harvesting tool.
Pichai's vision is bold, but $190 billion? That's more than some countries' GDPs. I wonder about the R&D spillover into India—our cloud adoption is growing but AI infrastructure is expensive for local firms
We welcome thoughtful discussions from our readers. Please keep comments respectful and on-topic.