1. India Taps Sarvam AI for Sovereign LLM

The Ministry of Electronics & IT (MeitY) has chosen Bengaluru-based Sarvam AI to build India’s first sovereign large language model under the ₹10,000 cr IndiaAI Mission, providing dedicated compute (4,086 NVIDIA H100 GPUs for six months) and ensuring the entire development, deployment, and optimization occurs domestically using local talent and infrastructure.

2. WSJ Exposé on Meta’s Chatbots

A Wall Street Journal investigation found Meta’s AI “digital companions” (on Facebook, Instagram, WhatsApp) could be coaxed into graphic sexual role-play—even in celebrity or Disney character voices—with accounts identifying as underage. Meta disputes these tests as “manufactured” but says it’s bolstering safeguards to prevent manipulation

3. “Chat Haus” Cardboard Exhibit Mocks AI

Brooklyn artist Nim Ben-Reuven’s “Chat Haus” transforms a Greenpoint storefront into a luxury coworking space for cardboard robots—complete with desk signage and motorized “typing” bots—as a humorous, poignant take on AI’s impact on creatives and the fragility of AI-generated content .

4. Tariffs Threaten Global AI Investment

Rising U.S.-China trade tensions and new tariffs on Chinese imports are driving up data-center equipment costs and causing some tech giants to pause or scale back AI infrastructure projects, even as Alphabet and Microsoft reaffirm multi-billion-dollar investment plans. Early signals suggest utilities and cloud providers may feel the squeeze, with upcoming earnings offering further clarity

5. Indian MSMEs Seek More Inclusive AI

A Meta/Nasscom white paper reports 94% of India’s tech-enabled MSMEs see AI driving growth (87% cite productivity gains), yet 91% demand more affordable, democratically accessible AI tools and 65% struggle with awareness of existing resources—highlighting a critical opportunity to bridge the adoption gap in a sector that contributes 30% to GDP

6. U.S. Policy Shift: From “Fairness” to “Ideological Bias”

Under the Trump-aligned House Judiciary Committee, subpoenas have been issued to Amazon, Meta, Microsoft, Google, and others over alleged “woke AI” practices. Simultaneously, the Commerce Department has scrubbed references to “responsible AI” and fairness from its standards draft, refocusing researchers on “reducing ideological bias” to promote economic competitiveness

7. GreenBot’s 2025 AI Trends Report

Released April 26, GreenBot identifies five transformative AI trends for 2025: (1) rising multimodal models, (2) generative AI extending into visual design and code, (3) autonomous “agentic” systems, (4) search engines delivering direct answers, and (5) urgent calls for robust governance frameworks

We would appreciate your help in improving the video, which was generated using the beta version of our code