If there was any proof needed that we are just getting started on AI, the news this week underscored it yet again. At Google I/O 2025, AI occupied centre-stage and a slew of announcements were made as Google pulled out all stops to try and inch ahead in the AI race. Google CEO Sundar Pichai used the term ‘AI’ 92 times in his keynote. The only term used more was ‘Gemini’, Google’s family of AI models.
Here are just some of the announcements that I thought were exciting, and truth be told, there are many more:
One of the most useful announcements was real-time translation with multimodal AI in Google Meet. Especially in a country like India where many languages are spoken, this would be useful in many scenarios. Obviously, the underlying technology would at some point go beyond Meet. Tools like these could end the outrage driven videos where some customer gets upset because a bank officer cannot speak his regional language.
Then there’s Stitch by Google Labs, which uses AI to generate designs and UIs—just tell Stitch what you want—say a shopping app with a pets theme and it will throw up a number of great options. The end of UI/UX designers? You tell me!
Then there’s Deep Think in 2.5 Pro. The LLM now features a deeper reasoning mode which gives it the capability to test multiple hypotheses before responding to your query. Another interesting tool is Veo3, a new generative video model—with the Flow filmmaking tool you can add sound effects to clips you make.
Some features are US-centric for now but may not be too far away from our shores. It may be a bit gimmicky, and many have tried variations earlier in non-AI avatars, but virtual apparel shopping gets a big boost with a new Labs feature where you upload a full body photo of yourself and you can virtually try on clothes to see how they look on you.
I’m also excited about AI mode in Google Search which will provide ChatGPT like answers directly in Google Search—you can still choose regular search but there’s no compulsion—the days of blue links on search may not be over yet, but there’s a new choice. There’s also a feature called ‘Deep Search.’ I hope Google brings this to India soon.
Here are just some of the announcements that I thought were exciting, and truth be told, there are many more:
One of the most useful announcements was real-time translation with multimodal AI in Google Meet. Especially in a country like India where many languages are spoken, this would be useful in many scenarios. Obviously, the underlying technology would at some point go beyond Meet. Tools like these could end the outrage driven videos where some customer gets upset because a bank officer cannot speak his regional language.
Then there’s Stitch by Google Labs, which uses AI to generate designs and UIs—just tell Stitch what you want—say a shopping app with a pets theme and it will throw up a number of great options. The end of UI/UX designers? You tell me!
Then there’s Deep Think in 2.5 Pro. The LLM now features a deeper reasoning mode which gives it the capability to test multiple hypotheses before responding to your query. Another interesting tool is Veo3, a new generative video model—with the Flow filmmaking tool you can add sound effects to clips you make.
Some features are US-centric for now but may not be too far away from our shores. It may be a bit gimmicky, and many have tried variations earlier in non-AI avatars, but virtual apparel shopping gets a big boost with a new Labs feature where you upload a full body photo of yourself and you can virtually try on clothes to see how they look on you.
I’m also excited about AI mode in Google Search which will provide ChatGPT like answers directly in Google Search—you can still choose regular search but there’s no compulsion—the days of blue links on search may not be over yet, but there’s a new choice. There’s also a feature called ‘Deep Search.’ I hope Google brings this to India soon.
On the other side of the AI war, Google’s biggest competitor in AI, OpenAI, had its own exciting announcement to make as former Apple design genius Jony Ive and his design firm will be taking creative and design control at OpenAI. They’ve already been working on something for two years now, and the idea is to develop a whole new generation of consumer devices—think beyond screens—that will leverage AI in ways that may seem unimaginable today. Are we seeing the end of the smartphone as we know it? It’s possible, because as we mentioned some weeks ago, Apple is also working on projects that could ultimately kill its biggest product, the iPhone. The kid in me who’s always thrilled to try out new gizmos couldn’t be happier. Google, OpenAI, Apple, Microsoft and many more—all working on changing how we interact with technology. And we’re just getting started.
Meanwhile, there’s lot else happening in the world of AI and here’s some of our coverage of it from recent days:
Apple To Open AI Models To Developers, Betting That It Will Spur New Apps
Technology-First Logistics: Building Resilience And Driving Innovation
Will AI Take Your Job? Experts Weigh On Its Growing Impact As Big Firms Slash Workforce Globally
Frightening Lack Of Cybersecurity Readiness In India Even As AI Raises Threat Levels: Cisco Study
AI's Fast-Moving Ecosystem Tops Gen AI-Related Security Risks For Indian Organisations: Thales Report
Amid AI-Powered Cyberattacks, Critical Need For Connected Security Operations: Splunk Report
Sify Announces Pay-Per-Use Model At Nvidia-Certified AI-Ready Data Centres In India
Responsive Launches No-Code Environment For AI Agent Creation
Microsoft Is Bringing Elon Musk’s AI Models To Its Cloud
Hewlett Packard Enterprise Deepens Integration With Nvidia On AI Factory Portfolio
HCLTech, IIT Guwahati To Upskill Employees In AI
Kore.ai Collaborates With Microsoft To Accelerate Enterprise AI Transformation
Yotta, Nvidia Launch Shakti Cloud On DGX Cloud Lepton To Power India’s Sovereign AI Ambitions
Shiprocket Unveils India's First AI-Integrated MCP Server
China Tech Earnings Expose AI Growing Pains
Google Rolls Out New AI, Accessibility Updates Across Android, Chrome
Meta Delays Launch Of Llama 4 Behemoth AI Model Over ‘Capability Concerns’
The excitement around AI doesn’t stop—see you next week!
-Ivor Soans
RECOMMENDED FOR YOU

What Is Perplexity Comet And How Does It Compare To Browsers Like Chrome, Safari, Firefox, And Edge?

Airtel Offers One-Year Perplexity Pro To 36 Crore Users At No Cost


Tesla Arrives In India — But Why Analysts Think Mahindra & Mahindra, Tata Motors Can Relax For Now


BSE Bomb Threat Email Sparks Police Search, But No Impact On Operations
