Promised land of AI lies in science, not commerce

At a time when most AI conclaves have become marketplaces for dream merchants selling models and apps that promise less work and more profit, the narrative has begun to feel predictable.

Since ChatGPT’s arrival in 2022, investments worth billions have poured into the sector. The gold rush continues and investors, caught in the fear of missing out, have brushed aside warnings from analysts about a possible bubble.

Against this backdrop, it was refreshing to hear a group of top scientists who gathered at the CSIR-NIIST centre in Trivandrum (aka Thiruvananthapuram) to discuss the advantages of using artificial intelligence (AI) and machine learning (ML) – not to make millions, but to deepen discovery.

This brings a new perspective to the wider debate between those who praise AI’s promised benefits and those who dismiss it as hype. That debate is almost entirely framed in commercial terms, but scientists argue that science is where AI can deliver real breakthroughs.

At the conclave “AI & ML for Transformative Innovation: Bridging R&D Frontiers and Industry”, this was the thrust of the speakers from India’s scientific establishment. They stressed that AI can open up new research frontiers and reach targets that have so far remained beyond human reach.

From VSSC director A Rajarajan to Niist director C Anandharamakrishnan, the message was consistent: AI in science research can not only sustain the economy by creating jobs instead of eliminating them — as feared with commercial AI — but also mitigate the environmental stresses of unsustainable growth.

Planetary stakes: Rajarajan framed the case in stark, planetary terms. “Human beings are the only animal species that overconsume, far beyond their needs,” he observed, warning that unchecked growth will exhaust Earth’s resources and force humanity to look beyond our planet for survival.

The implications are clear. Research into long-distance space travel, the physiological effects of gravity, in-space refuelling, and new materials for interplanetary permanence is no longer optional. These are large, difficult, expensive problems – precisely the kind of challenges where AI, properly deployed, can accelerate progress.

He also turned the usual job-loss narrative on its head. “The loss of jobs is not going to happen because developments in scientific areas from space to medicine will expand exponentially. We will be creating so many different things. It will need a lot of people and their efforts, and that is where we should be looking,” he said.

When AI is harnessed for discovery, it expands the scope of human labour and specialised expertise rather than shrinking it.

R&D in action: Anandharamakrishnan gave a series of concrete examples of this marriage of AI and science. Detection of microplastics in water, genome sequencing, early cancer detection, food safety – all stand to gain.

The globally noted food technologist recalled how computational fluid dynamics problems in his field once took a month to solve, involving complex formulae and billions of molecules. Today, optimised with AI and ML, such results can be obtained in just three days. Time-to-insight collapses, research cycles shorten, and innovation speeds up.

Niist’s own projects illustrate this vividly. The institute has built India’s first food 3D printer and is planning to integrate AI to personalise nutrition. The vision is futuristic but grounded: machines that can “print out your breakfast, lunch and dinner according to your dietary needs and taste”, driven by genomic, health, and lifestyle data.

The market for such food technology was valued at 298 million US dollars last year and is projected to hit 7.6 billion US dollars by 2034.

Building capacity: The path ahead is not easy, especially in India, where scientific research has often been an afterthought in policy-making. But over-reliance on global supply chains and geopolitical shifts has forced New Delhi to rethink about its priorities.

As part of new initiatives to boost research and development, several areas are being given funding and computing power by the government.

Gopalkrishna Patra, who heads the newly created CSIR Fourth Paradigm Institute (formerly the CSIR Centre for Mathematical Modelling and Computer Simulation), says that his centre was set up to integrate physics-based mathematical modelling, high-performance computing (HPC), and AI for research in climate science, urban resilience, and ground motion.

“The primary goal is to become India’s foremost data-driven science institute,” he said. Concentrated infrastructure of this kind allows India to set research agendas rather than rely on imported solutions.

The human bottleneck: Despite a series of well-funded initiatives by New Delhi – from India AI and the Semiconductor Mission to Productivity-linked Incentives and Design-linked Initiatives – the response from academia and research centres remains underwhelming.

Engineering colleges and much of the technology sector remain locked in the service economy and commercial projects. Private funding, too, rarely prioritises deep tech and R&D. Unlocking existing resources through training and access programmes could be India’s fastest route to AI-powered discovery.

But institutional support alone will not suffice. C-Dac director Dr Subodh PS highlighted the shortage of trained talent. “In the global AI talent rankings, India is number one. But there is a dearth of skilled workforce, and training on high-power computing is something all research centres in India should take up as a priority.”

He pointed out that the idea of establishing centres of excellence was to connect academia and industry. But so far, the centres established have been limited to some IITs. “No one has taken up the call to establish centres of excellence for education.”

The moral case: This is not just about strategy and ethics but also about an opportunity for India to take a key global role. As Anandharamakrishnan argued, India has the strength and resources to shape how research and data analysis proceed. That means setting standards and ethical guardrails for how AI is used in medicine, climate adaptation, and even space exploration.

If private capital alone sets the agenda, the public interest risks being an afterthought. If public science leads, markets will still follow – but with products that are socially useful as well as profitable.

AI’s story in the coming decade need not be a binary between jobless futures and billionaire founders. It can be one of accelerated discovery – of microscopes and telescopes, of models that predict monsoons, materials that withstand heat, food printers tailored to our needs, and centres advancing quantum simulations.

India has both the human talent and the beginnings of the infrastructure to lead this pivot. The question now is whether funders and policymakers will recognise that the surest commercial gains come when commerce follows scientific discovery – not the other way round.

When AI takes the red carpet

Ever heard of Tilly Norwood? She doesn’t actually exist – she’s an AI-generated persona who lives only on social media. The Instagram account that introduced her described Norwood as an “AI art and aspiring actress,” but controversy erupted when her creator, Dutch artist Eline Van der Velden, revealed in September that several talent agents were negotiating movie roles for her. The account now shows Norwood digitally inserted into glamorous red-carpet scenes, prompting outrage from stars like Emily Blunt and Ralph Ineson. As in other industries, it’s not far-fetched to imagine profit-driven studios seeing AI as a cheaper replacement for real talent.

In India, Bollywood stars Abhishek Bachchan and his wife, Aishwarya Rai Bachchan are also on war path against AI generated reels and suing YouTube. Many show them in controversial settings. The most popular video on the now-deleted channel was a video with 4.1 million views showing an AI animation of Salman Khan and Aishwarya in a swimming pool. India is YouTube’s largest market globally with 600 million users and in May, it disclosed that it had paid more than 2.4 billion US dollar – yes, Billion – to Indian creators over the last three years. So, anyone lucky enough to create a viral video stands to make some good money. Given that expect more AI slop coming your way.

Anthropic woos Indian coders

It’s not just YouTube eyeing India’s massive market – AI platforms are lining up too. After reports of OpenAI planning to set up an office in the country, Anthropic has announced an invite-only hackathon for Indian developers. Scheduled for October 11 in Bangalore and held in partnership with Accel, the event challenges participants to build projects using Claude Code or the Claude API. Entries must include a three-minute demo video, a 300-word description, and a GitHub repository or app link. The focus areas: fintech, healthcare, logistics, and compliance.

Google grows up in Singapore

Children accessing dangerous websites and watching inappropriate videos are a global concern. Singapore has taken that up with Google and the tech giant is putting in some prevention measures. The steps will come into effect in first quarter of 2026, and if the user is deemed under 18, safeguards will be applied automatically. Features include disabling timeline in Google Maps, restricting adult-only apps in Playstore, activating YouTube’s Digital Wellbeing tools, bedtime reminders, and other content safeguards. If Google’s system mistakenly flags an adult as under 18, they’ll need to verify their age – which can involve uploading a government-issued ID or a selfie for confirmation.

Thirty silver cents for your privacy

Everyone loves to rant about privacy – right before uploading their breakfast, their cat, and their latest purchase on social media. Offer a few cents, and suddenly privacy concerns take a back seat. How else do you explain the rise of Neon, an app that literally pays you to let it record your phone calls and sell the audio to AI companies? It recently hit No 2 on Apple’s US App Store under Social Networking, reports TechCrunch. Neon proudly claims to pay 30 cents per minute for chatting with other Neon users – up to 30 dollars a day. We suspect some users would be even complaining about “data theft” while cashing in on a Neon call.