AI Stack

AI Stack Latest News

India is anchored in the vision of AI for Humanity and building a robust AI stack is both a technological priority and a social commitment for India.

About AI Stack

• An AI stack is the complete set of tools and systems that work together to build and run AI applications. 
• It makes artificial intelligence work in the real world, from the apps people use every day to the data, computing power, networks etc.
• It is made up of five layers
  • Application Layer
    • It represents the user-facing component of the AI stack.
    • It includes AI-powered apps and services such as health diagnostic tools, farming advisory platforms, chatbots, and language translation applications.
    • This layer turns complex AI processes into simple, user-friendly services that people can easily use.
  • AI model layer
    • It acts as the brain of AI systems. AI models are trained on data to recognize patterns, make predictions, and take decisions.
    • It is the core intelligence that determines how effectively applications can understand, predict, and respond to real-world needs. 
    • Example: They help detect diseases from X-rays, predict crop yields, translate languages, or answer questions through chatbots.
  • Compute layer
    • It provides the computing power required to train and run AI models. During training, computers process vast amounts of data so the model can learn and improve. 
    • It is the critical enabler that determines the scale, speed, and sophistication of AI innovation.
  • Data Centres and Network Infrastructure Layer
    • Data centres are where AI systems are stored and operated, while networks like the internet, broadband, and 5G move data between users, computers, and AI models.
    • The data centres and network infrastructure layer provides the foundational backbone that enables AI systems to operate at scale and in real time. 
  • Energy Layer: It keeps the entire AI stack running. AI data centres consume large amounts of electricity because powerful computers are needed to train and operate AI systems.

Source: PIB

‹ ›