Top Stories
India AI Impact Summit Wraps Up With $200B+ in Investment Pledges
The India AI Impact Summit concluded in New Delhi after five days that transformed the country’s AI ambitions from aspiration to infrastructure reality. The numbers are staggering: over $200 billion in committed investments, plus $17 billion in venture capital for the Indian AI ecosystem.
Indian conglomerates led the charge. Reliance announced $110 billion for data centers and AI infrastructure. Adani outlined a $100 billion decade-long AI data center buildout. On the international side, Microsoft committed $50 billion for AI in the Global South by decade’s end. Amazon pledged ₹2.9 lakh crore for cloud infrastructure and AI-driven digitization by 2030. Google is building a 1 GW AI hub in Vizag.
The partnership announcements were equally significant. OpenAI signed a deal with Tata Group to become TCS’s first data center customer. Anthropic partnered with Infosys. The Indian government confirmed plans to add 20,000 GPUs through the IndiaAI Compute initiative.
This was the first AI summit hosted by a Global South nation, and the message was clear: India intends to be an AI superpower, not an AI consumer. Whether these pledges translate to actual infrastructure remains to be seen - summit commitments often shrink in execution - but the scale of ambition is unprecedented.
Sources: CNBC, Open Magazine, Digit
Spotify’s Best Developers Haven’t Written Code Since December
In a revelation that captures where AI-assisted development is heading, Spotify co-CEO Gustav Söderström disclosed during Q4 earnings that the company’s top developers “have not written a single line of code since December.”
They’re using an internal system called “Honk” built on Claude Code. The workflow sounds like science fiction from two years ago: an engineer on their morning commute can tell Claude to fix a bug or add a feature via Slack on their phone. Claude writes the code, and a new version of the app gets pushed back to the engineer’s phone for review. They can merge to production before arriving at the office.
The system helped Spotify ship over 50 features in 2025, including AI-powered Prompted Playlists, Page Match for audiobooks, and About This Song. Söderström framed this as “just the beginning” of AI development at the company.
This is the clearest evidence yet of how AI coding tools are changing the role of senior engineers. They’re becoming reviewers, architects, and decision-makers rather than line-by-line coders. The implications for the industry - hiring, salaries, skill requirements - are still playing out.
Sources: TechCrunch, Let’s Data Science
Disney Characters Now Available in OpenAI’s Sora
The OpenAI-Disney deal announced in December is now live. Users can generate short videos featuring over 200 characters from Disney, Marvel, Pixar, and Star Wars using Sora. ChatGPT Images can create stills from the same IP library.
The deal is substantial: Disney made a $1 billion equity investment in OpenAI and received warrants for additional equity. OpenAI is paying an unspecified licensing fee for access to 250 characters. The three-year exclusive agreement doesn’t include any talent likenesses or voices - you won’t be generating videos with actors’ faces.
Disney+ subscribers will see curated Sora-generated videos on the platform. Disney is also becoming a major OpenAI API customer, deploying ChatGPT internally and building new products for Disney+.
This is a test case for how major IP holders will engage with generative AI. Disney chose licensing over litigation - a billion-dollar bet that controlled access beats courtroom battles. Other studios and IP owners are watching closely.
Sources: OpenAI, TechCrunch
Quick Hits
-
Mistral AI makes its first acquisition: The French AI company is buying Koyeb, a Paris-based serverless deployment startup, to accelerate its Mistral Compute cloud offering. Koyeb’s 13 employees join Mistral’s engineering team. This signals Mistral’s ambition to become a full-stack player - not just models, but infrastructure. TechCrunch
-
Apple’s March 4 event looms: Bloomberg reports Apple is accelerating work on three AI wearables: smart glasses targeting 2027 launch, an AI pendant, and camera-equipped AirPods. All built around Siri with visual context. No display on the glasses - Apple is betting on audio and contextual AI rather than AR overlays. The March 4 “special Apple Experience” event may hint at this direction. Bloomberg
-
OpenAI’s $100B round nears close: The record-breaking funding round is approaching finalization, with Amazon ($50B), SoftBank ($30B), Nvidia ($20B), and Microsoft among backers. Valuation could exceed $850 billion. OpenAI revised its 2030 revenue target to $280 billion and compute spending to $600 billion. TechCrunch, CNBC
-
AI finds 25 new magnetic materials: University of New Hampshire researchers used AI to analyze scientific literature and build a database of 67,573 magnetic compounds. They identified 25 new materials that stay magnetic at high temperatures and don’t require rare-earth elements - potentially reducing EV dependence on Chinese-controlled supply chains. ScienceDaily
-
Google’s Gemini can now take over Search: Bloomberg reports Gemini has “matured to the point where it can take over Search.” AI Mode is reaching 75+ million daily users. LinkedIn says non-brand B2B traffic has dropped 60% as AI search experiences reduce clickthrough behavior. The shift from links to AI answers accelerates. Bloomberg
Worth Watching
The Spotify disclosure deserves attention beyond the headline. If senior engineers at a major tech company are no longer writing code directly, what does the career ladder look like in five years? The skills that got you promoted - deep language knowledge, debugging intuition, architectural expertise - still matter, but they’re expressed through AI orchestration rather than direct coding. The junior engineer path, traditionally built on grinding through tedious tasks to build intuition, may be disappearing.
India’s AI summit pledges will face the execution test. Summit announcements tend to include best-case scenarios and friendly math. But even at 50% delivery, the scale would transform India’s AI infrastructure. The political will is there. The question is whether regulatory, grid, and talent constraints can be overcome.
The Disney-OpenAI deal establishes a template: major IP holders licensing to AI companies rather than fighting them. If this works financially for Disney, expect other studios, publishers, and rights holders to follow. The “fair use” legal battles may become moot if the economics of licensing win out.