💼 Indian e-commerce expands gig economy. Google spent $200M on Gemini 1.0 Ultra. Indian city pollution.
Chart of the Day #525 looks at Indian e-commerce, air quality in Indian cities, and AI model training costs.
Pryvit 👋
Maine to upskill HVAC workforce. Rockwell Automation commits $1M to train AM workforce. AWS expands cloud skills initiative in Malaysia. UK to upskill future health workforce with science education.
Google and the University of Oxford are partnering to offer AI tools to support learning, research, and AI literacy. AWS and Exabytes launch cloud skills training initiative in Malaysia. OpenAI launches 'Education for Countries' program to support national AI education strategies and capacity building.
Today's Topics
💼 Workforce. Growing Indian e-commerce prompts gig economy expansion
🍃 Green Skills. Air quality in major Indian cities drop to record low levels
🤖 Artificial Intelligence. Google’s early AI training spending mirrored industry trends
Thanks for reading!
Not a subscriber yet?
💼 Growing Indian e-commerce prompts gig economy expansion
India's e-retail market emerged as the second largest globally with nearly 300M shoppers in 2025, driven by quick commerce, with platforms like Zomato, Blinkit, Zepto, and Swiggy Instamart delivering groceries in 10 minutes via extensive dark store networks, while multiple firms prompt competition. This rise in popularity fueled the gig economy, as the workforce expanded owing to growing demand. As the gig economy has helped increase the economic condition of many Indians, these delivery services face pressures with wages and welfare concerns, delivery volume pressure, and state-level regulatory oversight. Addressing these issues will be critical for sustaining growth in India’s evolving digital economy.
🍃 Air quality in major Indian cities drop to record low levels
Byrnihat and Delhi recorded the highest PM2.5 concentrations with deterioration stemming from industrial emissions, vehicle exhaust, construction dust, and biomass fuel use. Lifestyle disruptions, including air travel cancellations, train delays, school closures, halted construction, work-from-home mandates, and hospitals were flooded with respiratory cases. Key initiatives such as the NCAP, targeting up to a 40% reduction in PM10 levels by 2025-26, along with GRAP emergency protocols and restrictions on vehicle movements. Rapid action is needed for air quality improvement, so that disruptions to lifestyles are minimised.
🤖 Google’s early AI training spending mirrored industry trends
Google spent over $190M training Gemini 1.0 Ultra in 2023, reflecting the escalated training costs, from a mere $670 on Transformer in 2017 to $78M on GPT-4 and $191M on Gemini Ultra. Costs have continued to grow annually, driven by growing compute demands and scaling laws requiring larger models. Extended big company partnerships created tight-knit ecosystems. Microsoft invested $13B in OpenAI while partnering with Meta for Llama, and OpenAI secured deals with Oracle ($300B). However, concerns about an AI investment bubble are intensified as circular investments such as Nvidia-OpenAI-CoreWeave loops, OpenAI's $1T commitments against $4.3B revenue, and debt-financed data centers potentially raise concerns of ROIs and over-reliance on AI.
Like getting this newsletter? Request a demo for unlimited access to over one million charts
Thank you for reading. Have a great week ahead!
Have some feedback or want to sponsor this newsletter? Let us know at qs.com.