Comment by Riaz Haq on September 28, 2026 at 3:32pm

Building localized AI Models

By Nasir Jamal

https://www.dawn.com/news/2033093


Agriculture provides another practical use. AI systems could combine weather information, crop data, satellite imagery and local knowledge to provide farmers with timely advice. Such tools would not replace agricultural experts. They could extend their reach to communities that have limited access to them.

Healthcare and education also offer large potential gains. AI-assisted systems could help process medical records, support clinical workflows and expand access to basic information. In education, affordable AI tutors could provide personalised assistance to students while teachers could use the technology to prepare lessons and identify learning gaps.

The public sector should not be overlooked. Pakistan’s government generates enormous volumes of regulations, statistics, court documents, administrative records and public information. AI systems could make these materials easier to search and analyse. They could also reduce routine administrative work. Tax administration, for example, could use AI to identify anomalies and improve communication with taxpayers.

The economic case may be even stronger. Pakistan has struggled to raise productivity and expand exports fast enough to support sustained growth. AI offers a chance to sell higher-value digital services to international markets. Instead of competing mainly on labour costs, Pakistani firms could develop specialised solutions for finance, logistics, agriculture, education, manufacturing and other sectors. That requires investment.

Pakistan needs reliable electricity, better connectivity, cloud infrastructure, data centres and affordable access to computing. Universities need stronger links with industry. Engineering and computer-science programmes must give greater weight to machine learning, data engineering, AI systems and related fields.

This is an area where cooperation with China and BRICS countries could be useful. Pakistan could seek joint computing facilities, technical training, research partnerships and AI applications for local industries. It could also use BRICS platforms to secure access to infrastructure and expertise that would otherwise be expensive. Chinese technology can give Pakistan an alternative to systems dominated by Western companies. Data governance will be central to that effort.

Pakistan needs clear rules on where sensitive data is stored, who can access it and how AI systems can use it. Critical government and infrastructure data should receive particular protection. At the same time, regulation should not become so restrictive that it prevents researchers and businesses from building useful applications.

India has invested heavily in AI capacity while remaining attentive to questions of data security, strategic autonomy and technological dependence. Pakistan faces a similar policy challenge. This is also why AI governance matters.

For Pakistan, the test should be practical. Can a university train and fine-tune a model without prohibitive computing costs? Can a Pakistani company build an AI product for global customers? Can a student use an effective tutor, and a farmer receive useful information, in a local language? Can a government department process thousands of documents without exposing citizens’ private data?

Pakistan should engage with the BRICS initiative, but it should do so with a clear national strategy. It should seek access to models and computing, demand skills transfer, support local developers and establish rules that protect data without strangling innovation.

Most importantly, Pakistan should stop viewing AI simply as another imported technology. The strategic objective is to become a builder, adapter and exporter of AI solutions. China’s proposal may open the door. Pakistan’s task is to walk through it with its own engineers, its own companies, its own languages and its own priorities.

Comment by Riaz Haq on September 29, 2026 at 8:56am

India readies $25 billion for Deep Tech investment as U.S. and China race ahead

https://www.cnbc.com/2026/09/29/india-investment-billions-tech-ai-u...


India is planning to invest $25 billion in its nascent deep tech industry as it tries to catch up to China and the U.S.
Deep tech is a broad term used for start-ups in artificial intelligence, semiconductors, advanced manufacturing, drones and space tech.
Deep tech is becoming “very critical” to every country amid rising geopolitical tensions, leading to an investment push from the Indian government.
While venture capital and private equity leaders in India are confident that a global Deep Tech champion would emerge out of India, they were hesitant to put a timeline on it.

India is preparing to deploy $25 billion to support Deep Tech start-ups in the country, as it tries to catch up with the U.S. and China in the global tech race to cut its reliance on frontier foreign technology.

In the last decade, India invested $11.6 billion in Deep Tech, Rajat Tandon, president of the Indian Venture and Alternative Capital Association, told CNBC.


But now, the government alone has committed to invest $11 billion under the Research Development Infrastructure Fund, which will be matched by venture capital and private equity fund managers, he said.

Three or four billion more will be added, making ”$25 billion available to be spent on Deep-Tech investments,” Tandon said.

Deep Tech is a broad term used for start-ups in artificial intelligence, semiconductors, advanced manufacturing, drones and space tech. Experts told CNBC that the funding push comes as the government has realized that frontier tech is becoming “very critical” amid rising geopolitical tensions.

China and the U.S. currently lead the global AI race, and while Chinese tech is seen with suspicion in India, Washington’s curbs on tech exports make it unreliable.

“Tariffs from the U.S. actually help this [Deep Tech] segment a lot,” Anandamoy Roychowdhury, managing director of Crane Venture Partners, said, adding that India will develop local Deep Tech companies as it fears that “important technology can get cut off at any point.”

Some of these fears were realized earlier in June when Anthropic disabled access to its new models — Fable 5 and Mythos 5 — for foreign nationals, complying with an export-control directive from the U.S. government.

The start-up funding leaders who had gathered for SuperReturn Asia said that India’s Deep Tech companies were at a nascent stage but added that the ecosystem was poised to create global champions eventually given the quality of ideas and talent in the country.

There is a “dramatic acceleration of innovation” in the Deep Tech space in India, Shweta Rajpal Kohli, president and chief executive of Startup Policy Forum, said, adding that some companies are moving from prototype to “real commercialization.”

Earlier this year, Vibe-coding startup Emergent, space tech company Skyroot and India’s full-stack sovereign AI company Sarvam became unicorns as their valuations crossed $1 billion during the latest fundraising round.

‘Candy store’

“I feel like a kid in a candy store,” Roychowdhury of Crane Ventures said, describing his experience of scouting for Deep Tech start-up investments in the country. About 80% of his $150 million APAC fund is currently concentrated in India, he shared.

In August, the IVCA, in a report, said that its survey across 100 funds in the country showed that nine out of 10 funds in India were deploying capital in Deep Tech startups, with 37% holding a stake in 11-20 such companies,

The report added that in 2025 the sector received nearly $3 billion in funding—its highest ever, even as overall start-up funding in India fell. However, it also underscored the massive funding gap with the U.S., where $136 billion was raised.

One of the key disadvantages for Indian Deep Tech start-ups over those in the U.S. is the lack of domestic capital required to scale these long-gestation, innovative investments.

Comment

• ‹ Previous
• 1
• 2
• 3
• 4
• Next ›
• Page