The Union Ministry of Skill Development and Entrepreneurship (MSDE) announced the “Skill the Nation” challenge under President Droupadi Murmu with the goal of developing advanced artificial intelligence (AI) skills and preparing India’s workforce for the future.
President Murmu Launches Skill the Nation Challenge
At the challenge’s launch on Thursday at the Rashtrapati Bhavan Cultural Center here, President Murmu said that artificial intelligence is changing society and economies throughout the world.
This challenge aims to promote widespread involvement in AI learning and innovation. Millions of students will benefit from this program, which will also help them develop their skills and create a workforce that is ready for the future in India, she said.
AI’s Role in Society and Economic Growth
She said that AI is revolutionizing education, employment, service delivery, and problem-solving globally and that it presents a significant chance for good change for a young country like India. According to Murmu, India has always believed that technology should empower people, encourage inclusivity, and provide possibilities for everyone.
“The use of AI should strive to bridge technical, societal, and economic gaps.” She added, “It is crucial to make sure that its benefits reach people of all backgrounds and ages, especially those from marginalized communities. AI is emerging as a key growth driver for India’s economy, contributing to GDP, jobs, and productivity.” She urged government, business, and academia to work together to create a responsible, skilled, inclusive, tech-driven India.
Skilling for AI Readiness Initiative
Additionally, under the skill development and entrepreneurship’s “Skilling for AI Readiness” (SOAR) initiative, President Murmu awarded AI certificates to students and Members of Parliament (MPs) who finished a variety of online modules. These modules include short 10- to 15-hour introductory courses as well as extensive 30- to 60-hour programs spread over several days to weeks, covering AI fundamentals, data science, analytics, and applied AI skills.
Murmu expressed gratitude to MPs and students, saying, “You are preparing yourselves for a future full with possibilities and chances.” You must keep in mind that you should utilize technology, together with your knowledge and abilities, to advance society, solve problems, and assist others.
As China Advances AI Efficiency, DeepSeek Announces New Training Technique
Despite not having unrestricted access to Nvidia Corp. processors, DeepSeek issued a paper explaining a more efficient strategy to building AI, demonstrating the Chinese AI industry’s attempt to compete with companies like OpenAI.
Manifold-Constrained Hyper-Connections Framework
Liang Wenfeng, one of the document’s co-authors, presents a framework known as Manifold-Constrained Hyper-Connections. The authors claim that it is intended to increase scalability while lowering the computational and energy requirements of training sophisticated AI systems.
In the past, DeepSeek’s articles have predicted the release of significant models. A year ago, the Hangzhou-based business, which built the R1 reasoning model for a quarter of the cost of its Silicon Valley competitors, shocked the industry. Since then, DeepSeek has launched a number of smaller platforms, but excitement is growing for the R2, the company’s next flagship system, which is anticipated to launch around February’s Spring Festival.
Innovation Under Chip Restrictions
Due to US restrictions on access to the most cutting-edge chips needed for AI development and operation, Chinese firms are still operating under severe limitations. These limitations have compelled researchers to use novel techniques and structures.
This week, DeepSeek, which is renowned for its unconventional inventions, released its most recent work via the open-source platform Hugging Face and the arXiv repository. Liang’s name appears last among the 19 authors listed in the study.
Future of Large-Scale AI Systems
The creator, who has continuously guided DeepSeek’s research agenda, has pushed his team to reconsider the design and construction of large-scale AI systems.
The most recent study notes that the new approach integrates “rigorous infrastructure optimization to assure efficiency”, addressing issues like training instability and restricted scalability. Building on ByteDance Ltd.’s 2024 study on hyper-connection architectures, tests were carried out on models with 3 billion to 27 billion parameters.
According to the authors, the method has potential “for the development of basic models.”
