India just topped global enterprise AI adoption. Sarvam is raising ₹2,900 crore. Microsoft’s Hyderabad data centre goes live this summer. Five developments from this week — with the India context that actually matters to you.
Five AI stories. All happened in the last 7 days. All have a direct implication for Indian professionals, founders, and creators. No global fluff — only what changes something for you.
Deloitte’s 2026 State of AI report surveyed 3,200+ senior leaders across 24 countries. The headline: 40% of Indian enterprises report significant or full AI deployment, against 28% globally. India ranked first.
The detail that did not make the headline: deep AI specialist talent in India is described as almost nonexistent. Enterprises are deploying AI faster than they are building the people to run it. 39% of Indian enterprises now cite regulatory and compliance demands as their top obstacle — up from a footnote two years ago.
Sarvam AI — the Bengaluru startup building LLMs trained from scratch on Indian languages — is closing a $300–350 million round at a $1.5 billion valuation. This comes after the Indian government selected Sarvam under the IndiaAI Mission to build India’s first sovereign large language model.
Sarvam already beat Gemini 3 Pro and ChatGPT on the olmOCR benchmark for Indian language document reading (84.3% vs 80.2% vs 69.8%). Their Bulbul V3 voice model covers 11 languages with 35+ voices. Their ASR model handles all 22 official Indian languages.
Microsoft’s $17.5 billion India AI infrastructure commitment is now materialising. A new hyperscale data centre in Hyderabad goes live mid-2026 with three availability zones. Microsoft is also doubling its pledge to train 20 million Indians in AI skills by 2030.
For Indian developers and startups, Azure AI services will have dramatically lower latency — models running in India, data staying in India. The Hyderabad centre will be one of the largest in Asia.
Google rolled out a major Gemini-powered update to its shopping ecosystem in India this week. The assistant now handles queries like “laptop under ₹60,000 for video editing with same-day delivery in Bhopal” — pulling real-time inventory from Flipkart, Amazon and local retailers simultaneously.
The bigger story for small businesses: Gemini-powered merchant tools now let local shop owners generate product descriptions and catalogue images from smartphone photos. No designer, no agency. Google projects 10 million+ new local merchants entering Google Shopping by end 2026 through this tool.
67% of talent acquisition professionals plan to increase investment in AI recruitment screening in 2026, with 46% saying candidate quality got worse even as application volumes surged. AI pre-screening is now standard at scale across Indian hiring.
Separately, India’s AI talent market is tightening fast. With Sarvam, Krutrim, and Microsoft all scaling India teams simultaneously, ML engineer salaries are climbing. Krutrim already launched Kruti — an agentic assistant supporting 13 Indian languages — competing for the same talent pool.
Sarvam’s Indus app — free on web and mobile. If you deal with any regional language content — client documents, WhatsApp messages, vendor invoices in Hindi or Gujarati — this is the fastest way to test what Indian-built AI feels like on Indian text. Go to sarvam.ai, try the free chat, and compare it to ChatGPT on the same Hindi input. The accuracy difference on regional text is real.
India topped global enterprise AI adoption this week. The infrastructure is arriving — Hyderabad data centre, Sarvam’s capital, government backing. The tools are getting better for Indian use cases specifically.
The gap that remains is not ambition or capital. It is people who can actually implement the tools. That gap is where every reader of this newsletter sits — and it is closing faster than most people realise.
Next issue: a deep cut on one profession. Reply and tell us which field you want covered.