Chennai, August 16
Tamil Nadu’s plan to train five lakh people in artificial intelligence by 2031 could face a major practical hurdle as colleges struggle to secure regular access to the advanced computing resources required for hands-on learning.
The State’s AI Economy Mission, announced in the 2026-27 Budget, will use existing engineering colleges, polytechnics and Industrial Training Institutes to provide AI training. However, many institutions lack research-grade Graphics Processing Units, which are essential for training models, processing large datasets and conducting practical exercises.
A single advanced GPU costs around ₹20 lakh, placing dedicated infrastructure beyond the financial reach of many colleges. Practical AI courses may require between 15 and 20 such units to provide meaningful training to students.
The Union government has said colleges and students can apply for subsidised GPU access under IndiaAI Compute. The programme allows eligible users to remotely access shared computing infrastructure through approved cloud-service providers, removing the need to purchase costly equipment.
Around 10,571 GPUs have been allocated under the initiative. Additional capacity is available through the AIRAWAT facility and the PARAM Siddhi-AI system, which together offer nearly 48 petaflops of processing power and 656 GPUs.
However, no fixed share of this capacity is reserved for universities or undergraduate students. Academic applicants must establish their eligibility, explain the project and justify their technical requirements before receiving approval.
An application seeking access to around 15 GPUs has been submitted, with a proposal to distribute the capacity to students through a university partnership.
The application-based system may work for research projects, but colleges require continuous, large-scale access for routine classroom training.
Institutions also face shortages of curated datasets, secure testing platforms, advanced AI tools and teachers with industry experience. Unless these gaps are addressed, AI education may remain largely theoretical despite rising employer demand for graduates with practical skills.
