Рубрика: AI

“Last quarter I rolled out Microsoft Copilot to 4,000 employees. Thirty dollars a seat, per month. A neat, board-pleasing line item of around $1.4 million a year.”

The punchline in security engineer Peter Girnus’s now-viral, satirical post is not the spend, but what followed.

Three months in, usage reports showed just 47 employees had opened Copilot, only 12 had used it more than once, and Girnus himself mostly used it to summarise emails he could have read faster on his own, while also fixing hallucinations.

The project was still declared a “pilot success”, carried forward by familiar phrases such as “digital transformation” and “10x productivity”, even as day-to-day reality looked closer to a stalled experiment than a breakthrough in workplace AI.

It is into this gap between promise and practice that Microsoft has introduced its new “Frontier Firms” narrative, with four of the world’s largest IT services companies—Cognizant, Infosys, Tata Consultancy Services (TCS) and Wipro—now positioned as flagship examples.

In Bengaluru on December 11, Microsoft chairman and CEO Satya Nadella announced strategic partnerships with the four companies, under which each will deploy more than 50,000 Microsoft Copilot licences, collectively crossing two lakh seats.

Microsoft said this would “set a new benchmark for enterprise-scale AI adoption.”

Puneet Chandok, president of Microsoft India & South Asia, described the companies as organisations “moving beyond experimentation to full-scale deployment,” embedding Copilot into everyday work and “setting the global pace” on agentic AI.

‘Frontier Firms’

The firms are explicitly labelled “Frontier Firms”, enterprises that redesign workflows around human-agent collaboration across delivery, sales, finance, HR and customer engagement.

Each company, however, presents a different picture of what that frontier looks like.

Cognizant, which Microsoft has positioned as “client zero” for Copilot, has tied the partnership to its identity as an “AI builder company.”

CEO Ravi Kumar S has said Cognizant’s role is to bridge “hundreds of billions” of dollars in AI infrastructure investment with real business value, using Copilot and agentic solutions to rewire how organisations access data and make decisions.

Infosys, which Microsoft calls one of the world’s largest Copilot deployments, is integrating Microsoft’s Intelligence Layer with its own Topaz Fabric and Cobalt platforms to operationalise “multi-agent workflows” and a “human+agent powered AI-first enterprise,” according to CEO Salil Parekh.

TCS is using Microsoft 365 Copilot and GitHub Copilot internally across sales, HR and finance, equipping tens of thousands of employees with AI tools, while running a company-wide hackathon involving over 2.81 lakh participants and providing all employees with a personalised AI coach, CEO K Krithivasan has said.

Wipro, through a three-year strategic partnership and a Microsoft Innovation Hub at its Bengaluru Partner Labs, is embedding Copilot and agentic AI into Wipro Intelligence, with CEO Srini Pallia saying the collaboration is reshaping how enterprises work, and accelerating adoption across sectors including financial services and healthcare.

For Indian IT, which has spent the past year positioning itself as the global delivery engine for enterprise AI, the “Frontier Firm” label is both an endorsement and a test.

Lowered Growth Expectations

These companies have large internal workforces to deploy copilots at scale, deep client relationships, and long experience in managing technology change.

On paper, that makes Microsoft’s choice of reference customers logical, and the scale, over two lakh Copilot licences, places them among the largest real-world testbeds for agentic workflows.

The unresolved question, however, is whether these deployments translate into deep, sustained usage, or whether some will echo the Girnus experience of expensive licences with limited everyday impact.
Microsoft did not respond to questions from AIM, probing whether Copilot’s enterprise deployments translate into sustained, everyday usage and measurable return on investment, or remain headline-driven rollouts that struggle to deliver real value after launch.

TCS, Wipro, Cognizant and Infosys did not respond either to queries whether “Frontier Firm” Copilot deployments translate into actual usage, use cases, and measurable productivity or delivery impact, rather than announced scale or partnership optics.
The unresolved question has sharpened as Microsoft itself has moderated expectations around enterprise spending on AI agents.

Entering 2025, executives had positioned the year as a turning point in which autonomous agents would move from demos into production budgets.

But, according to a report by The Information, followed by television and wire coverage, several Microsoft divisions lowered internal growth expectations for newer AI offerings after many Azure sales teams missed sales-growth goals for the fiscal year ended June.

In one US Azure unit, a target to grow customer spending on the Foundry service by 50% was met by fewer than one-fifth of sales staff, prompting a reduction in the growth target to about 25% for the current fiscal year; another unit cut an earlier goal of doubling Foundry sales to a 50% growth target.

Microsoft said that “aggregate sales quotas for AI products have not been lowered,” arguing that some reporting had confused internal planning targets with compensation-linked quotas.

Copilot Versus Cursor

However, Microsoft’s Copilot family of AI assistants surpassed 100 million monthly active users across commercial and consumer versions in FY2025, according to the company’s annual report. Copilot Studio is now used by more than 230,000 organisations to build custom agents and GitHub Copilot has reached over 20 million users as a peer programmer for software development.

That is where the “Frontier Firms” idea will ultimately be tested. For Cognizant, Infosys, TCS and Wipro, Copilot deployments are framed as both internal transformation and proof-points for clients.

The decisive evidence will come not from announcements or licence counts, but from engineers, consultants and delivery teams who use these tools daily.
P S Ranjith Kumar, VP of AI & tech at Khetika, an Indian healthy food brand, said he switched to Cursor for its convenience and deeper contextual understanding.

“Cursor understands the entire repository as context, while Copilot mostly reacts to the current file or cursor position,” he explained.

To him, Cursor feels like a true pair programmer, whereas Copilot often comes across as an intelligent autocomplete tool.

He added that Cursor’s key strength lies in being built from the ground up around AI, unlike Visual Studio Code with GitHub Copilot, which in his view remains “a traditional editor with AI bolted on.”

Worst Thing to Happen?

Jon Pressnell, a partner at Blue Point Capital Partners and self-described AI evangelist, declared on LinkedIn two months ago that “Microsoft Copilot is the worst thing to happen to enterprise AI adoption,” citing sky-high hype undone by disappointed customers, wasted budgets, and eroded industry trust.

He pointed to low usage rates in deployments “gathering dust,” security and governance fears around data privacy, high licensing costs paired with inconsistent performance and dubious ROI, and a resulting “Copilot fatigue” that has bred market skepticism toward all enterprise AI services, even as industry leaders voice public frustration.

Pressnell ended with a half-question: “Is Copilot really the worst thing to happen to enterprise AI, or am I just bitter because I thought we would get the product they demoed?”
Replying to Pressnell, Debasish Bhattacharjee, director of engineering and operations (AI) at SAP, said Copilot’s real sin isn’t bad performance, it’s teaching enterprises that AI adoption can skip proper change management.

“We see 80% better results when teams start with narrow, measurable AI pilots rather than company-wide rollouts,” he shared.

Copilot Seeks Comprehensive Adoption

Edyta Gorzoń, a Microsoft MVP (Most Valuable Professional) and Copilot adoption specialist, outlined two recurring challenges she hears from enterprise customers implementing Microsoft 365 Copilot. The first is “cautious investment” — companies buy a small set of licenses (10–20 for organisations with 100+ employees) to pilot the tool, but then get stuck when few users actively use it, or when dashboards show low engagement.

Gorzoń said assigning licenses alone doesn’t drive adoption; organisations need to study user behaviour, identify repetitive tasks, run pilot studies, and align expected ROI with business goals before scaling.

The second is “ineffective training” — clients assume a few standard training sessions will increase usage, but see little change.

She explained that Copilot adoption requires more than training; it demands cultural and behavioural change management, building internal “Copilot Champion” networks, ongoing communication, and contextual learning.

Without a comprehensive adoption strategy, she warned, investments in Copilot licenses are unlikely to pay off or deliver measurable returns.

The post Two Lakh Copilot Licences, One Big Question: Does Indian IT Actually Use Them? appeared first on Analytics India Magazine.

In 2025, a growing cohort of Indian startups earned global recognition by securing spots in international accelerator programmes, raising capital from marquee global investors, and building AI products with cross-border relevance. From satellite intelligence and enterprise AI agents to voice automation and adaptive learning platforms, these startups reflect India’s shift from an AI services hub to a product-led innovation ecosystem.

Their success signals rising global confidence in Indian-built AI technologies that are scalable, commercially viable, and increasingly central to enterprise, government, and consumer AI adoption worldwide.

1. SatSure

SatSure gained significant global visibility in 2025 after being selected for the IndiaAI Startups Global Accelerator, co-hosted by Station F and HEC Paris. The selection placed SatSure among a small cohort of Indian AI startups introduced to European investors, policymakers, and enterprise partners, reinforcing its position as a globally relevant geospatial intelligence company rather than a region-specific analytics player.

The startup uses AI and satellite imagery to deliver decision intelligence across agriculture, climate risk, infrastructure planning, and financial services. Its platforms are used by governments, insurers, agribusinesses, and enterprises to monitor crop health, assess climate exposure, and manage large-scale assets.

2. NeuroPixel.AI

NeuroPixel.AI drew global attention in 2025 through its inclusion in the IndiaAI global accelerator programme in Paris, signalling international confidence in India-built generative AI tools for the creator economy. Its recognition stemmed from a precise positioning in a crowded GenAI market: speed, workflow efficiency, and practical deployment for real commercial use cases.

The startup builds AI-powered image-editing and visual workflow tools tailored for fashion, e-commerce, and content teams that need rapid, repeatable creative output. By focusing on production-ready tooling rather than experimental generative art, NeuroPixel.AI has attracted interest from global brands and agencies seeking scalable design automation.

3. CoRover.ai

CoRover’s global recognition in 2025 was driven by its selection into the IndiaAI-Station F accelerator, placing it in front of international enterprise buyers and government stakeholders. Already known in India for deploying conversational AI at a population scale, the accelerator marked its transition from domestic success to global ambition.

The company builds multilingual conversational AI platforms used for customer support, enterprise workflows, and citizen-facing digital services. Its systems are designed to handle high volumes and complex interactions across languages, making them especially relevant for emerging markets globally.

4. Ringg AI

Ringg AI emerged as one of the most visible Indian voice-AI startups globally in 2025 after raising a $1 million seed round led by Capital 2B, alongside support from Microsoft for Startups and NVIDIA Inception. The backing validated Ringg’s focus on voice as a primary AI interface in markets where chat alone is insufficient.

The startup builds AI voice agents to automate inbound and outbound calls across use cases, including lead qualification, appointment scheduling, loan collections, logistics coordination, and customer support. Supporting more than 20 languages, Ringg AI has found adoption beyond India, particularly in the Middle East and Latin America. Its technology addresses a global demand for scalable, human-like voice automation in high-volume operations.

5. Nurix AI

Nurix AI gained strong global recognition in 2025 after raising $27.5 million from Accel and General Catalyst, positioning it among the most well-funded Indian startups focused on enterprise AI agents. The scale and profile of the investors signalled confidence in Nurix’s approach to agent-driven enterprise automation.

The company builds custom AI agents that integrate into sales, support, and internal business workflows, automating repetitive and decision-heavy processes. Its solutions are designed for enterprise-grade deployment, focusing on reliability, compliance, and measurable productivity gains. With international customers and pilots underway, Nurix AI reflects the growing global demand for agentic AI beyond consumer applications.

6. Staqu Technologies

Staqu Technologies’ global visibility in 2025 was strengthened by its inclusion in the IndiaAI global accelerator programme, introducing its AI-driven surveillance and analytics platforms to European markets. The selection highlighted the international relevance of its computer-vision capabilities beyond Indian public-sector deployments.

Staqu develops AI solutions for video, image, and audio analytics used in public safety, urban governance, and enterprise monitoring. Its systems help organisations detect anomalies, manage crowd behaviour, and improve real-time situational awareness.

7. VolarAlta

VolarAlta gained international recognition through the IndiaAI-Station F accelerator, reflecting rising global interest in AI-enabled aerial intelligence. Its selection aligned the startup with European climate-tech and industrial monitoring ecosystems increasingly reliant on drone-based data.

The startup integrates AI with drones to deliver aerial analytics for emissions monitoring, infrastructure inspection, and environmental compliance. These use cases are highly relevant across energy, utilities, and climate-focused industries worldwide.

8. Smartail

Smartail’s global recognition in 2025 came via its participation in the IndiaAI global accelerator, signalling international confidence in adaptive learning technologies built in India. The programme connected Smartail to global EdTech networks at a time when AI-driven personalisation is reshaping education worldwide. Smartail builds AI systems for personalised learning, student assessment, and adaptive content delivery. Its tools help educators tailor instruction to individual learner needs while generating real-time insights on progress and gaps.

Disclaimer: This list is not a ranking. The order does not reflect size, impact, valuation, or investment performance.

The post The Indian Startups the World Took Notice Of in 2025 appeared first on Analytics India Magazine.

Even at 81, Larry Ellison is still in builder mode.

As Oracle’s executive chairman and CTO, he is betting on AI infrastructure with the urgency of a founder chasing a new gold rush, pouring billions into data centres to train and serve large models.

Ellison has ridden many such cycles at Oracle before, turning databases, enterprise software and cloud into growth engines through conviction and timing.

This time, however, as the economics of AI and data centres come under strain, the latest bet that carries his imprint is being tested and increasingly questioned.

In December, Oracle’s stock fell more than 10% in a single session, wiping out tens of billions of dollars in market value after the company warned that AI-related capital spending would weigh on earnings.

Shares reached an all-time high of ~$328 in September but have since fallen to ~$197 as of December end, representing a drop of ~40%.

The sell-off followed reports of uncertainty about Oracle’s funding for multi-billion-dollar data centre projects.

Oracle has committed ~$300 billion to cloud services and AI data-centre capacity for OpenAI, leasing massive GPU clusters inside data centres.

Internal documents showed Oracle generated ~$900 million in revenue from renting servers powered by NVIDIA chips, but only about $125 million in gross profit, a ~14 % gross margin on that business.

So, where do we go from here?

A Fragile Boom Built on Fast-Moving Capital

“LLMs can get better by leaps and bounds, and you could still have a very financially fragile sector,” Advait Arun, senior associate for capital markets at the Centre for Public Enterprise (CPE), told AIM in an interaction.

Arun recently authored Bubble or Nothing at CPE, a comprehensive report that examines how the AI boom is being financed.

Capital continues to densify and round-trip around a small set of players.

Oracle’s commitment to OpenAI is only one example. NVIDIA and OpenAI have announced plans to deploy at least 10 gigawatts of NVIDIA systems, with up to $100 billion in planned investment.

OpenAI has committed up to $1.4 trillion in long-term infrastructure and compute investments with partners including Microsoft, Oracle and NVIDIA, even as it remains unprofitable—with those commitments explicitly intended to be funded out of future revenues paid back to the same firms.

Alphabet, Meta, Microsoft and Amazon are expected to spend over $350 billion on AI infrastructure in 2025, with that figure projected to cross $400 billion in 2026.

HSBC estimates that OpenAI alone will require at least $207 billion in computing capacity through 2030, even as it expects the company to remain unprofitable.

CPE’s report states that hyperscalers invested more than $560 billion into AI technology and data centres between 2024 and 2025, while generating just $35 billion in associated revenues.

Analysts at Bain & Company estimate the sector needs $2 trillion in new revenue just to fund the data centres profitably planned for 2030.

The circular ecosystem pushes losses around rather than recognising them—so when one node weakens, stress propagates through everyone connected.

Image: Center for Public Enterprise (CPE)

Usage is Not Demand

Part of the disconnect between technical advancement and financial fragility lies in how demand is measured. AI revenue is rarely separated cleanly from broader cloud or software sales.

Usage is bundled into consumer products and enterprise suites, making it difficult to tell whether customers are willing to pay prices that justify the capital locked in upstream.

Many companies now claim unprecedented levels of AI adoption, often comparing it to the uptake of automobiles, smartphones or the internet.

But being considered a “user” does not necessarily mean a product is delivering sustained value or changing how work is actually done.

If the AI assistant embedded into Gmail gives a user an auto prompt—or simply a cue to use AI—does hovering the mouse over it, or accidentally clicking on it, count as a ‘use’?

Of ChatGPT’s roughly 800 million weekly active users, only about 5% subscribe to a paid plan.

Leslie Joseph, a principal analyst at Forrester Research, explained to AIM that much of what is being counted as “AI adoption” today is shallow and fragile.

“Most companies just woke up yesterday, [integrated a] Copilot tool — and that’s it,” he said, pointing to how many deployments stop at surface-level assistants instead of re-engineering data pipelines, workflows, and decision systems that would be required for AI to deliver durable productivity gains.

Joseph’s point is not that enterprises lack interest in AI, but that adoption has been reduced to surface-level tooling rather than structural change.

Adding an AI layer to existing workflows, he argues, is not the same as rebuilding how work actually gets done. “It’s a failure of imagination and execution,” he said.

Infrastructure spending, however, is being justified as if that transformation has already occurred.

“When so much use is also personal rather than enterprise, it’s not really clear that any of this can be sustained at a price point that’s reasonable for the sector’s health,” Arun explained.

Why the ‘Productive Bubble’ Analogy Breaks Down

Defenders of the boom often argue that even if AI is in a bubble, it is a productive one.

Capital overshoot, they say, builds infrastructure that remains useful long after weaker firms fail—much like fibre laid during the dot-com era or railroads constructed in the nineteenth century.

All the fibre and railroads laid down 25 years ago are still in use.

But will the GPUs being laid down today be used even five years from now? And, more importantly, will they be expected to continue to grow the revenue they help an entity earn?

CPE states that the economic life of modern AI GPUs is estimated at two to four years, even as new designs arrive annually.

That matters because GPUs are increasingly used as collateral for the loans financing AI expansion.

In 2023, high-end GPUs such as NVIDIA’s H100 were rented for about $8 an hour. By 2024, rates for that same class of hardware had fallen to close to $1 per hour, with an 80-85% price drop within a year.

Jordan Nanos, an analyst at SemiAnalysis, told AIM that GPU lifecycle decisions are governed more by economics than by capabilities.

“If the performance per dollar [achieved] GPU is higher than the cost to keep it running, people will keep using it,” he added. “GPUs get replaced when the power and floorspace in the datacenter can be used for something else”

That threshold is being reset every year by NVIDIA’s rapid upgrade cycle.

“If new GPUs are so much more performant than the old ones that it no longer makes economic sense for buyers to rent the old ones,” said Nanos.

As a result, demand drains from older variants even if they still work. Meanwhile, buyers chase the newest chips because, in an intensely competitive market, marginal gains in speed and efficiency often decide who can train faster, serve cheaper and win customers.

In practice, contracts and depreciation schedules, not hardware failure, determine when capacity exits the system.

That depreciation dynamic sits directly underneath how today’s AI infrastructure is financed.

Debt’s Role in Shaping the Boom

Today’s AI build-out is being financed through two distinct debt stacks, each with its own risks. Both are also increasingly intertwined.

Data centre developers raise capital much like commercial property firms.

Short-term construction loans and mini-perm facilities fund builds, which are then refinanced into longer-term debt once tenants are secured and cash flows stabilise.

These longer-term loans are bundled into asset-backed securities (ABS) or commercial mortgage-backed securities (CMBS) and sold to institutional investors.

Data centres already account for roughly 61% of the $79-billion digital-infrastructure securitisation market, embedding AI build-out risk into pensions, insurers and banks.

Most AI-focused data centres built in 2024-25 rely on structures that require refinancing in 2027-28, even though the debt can run 10 to 20 years while tenant leases typically last just three to five.

And what if, at renewal, tenants downsize, renegotiate aggressively, or simply walk away because GPU economics have shifted or their AI workloads no longer justify the cost?

Buildings underwritten as long-lived assets are left chasing shorter, weaker cash flows, just as mini-perm loans roll into term refinancing. This forces owners either to inject fresh equity, accept punitive terms or hand the keys to lenders.

Sitting on top of this is a second layer of leverage, GPU-backed loans taken by neoclouds and AI operators.

For example, the triangle between Google, TeraWulf, and Fluidstack is highlighted in CPE’s report.

TeraWulf, a data centre developer, issued $3.2 billion in high-yield bonds to build GPU-heavy facilities that would be leased to Fluidstack, a neocloud operator.

Fluidstack’s ability to pay those rents, in turn, was effectively guaranteed by Google, which committed to minimum payments for the capacity.

So while Google did not issue the debt itself, its promise of revenue is what made TeraWulf’s ‘junk’ bonds viable in the first place.

The bonds sit on TeraWulf’s balance sheet. FluidStack runs the GPUs. But the cash flows ultimately depend on Google.

Image: Center for Public Enterprise (CPE)

The result is a three-layer structure where leverage is raised against Google-backed demand, even though Google keeps the debt off its own books.

Neoclouds such as CoreWeave and Lambda have taken on multi-billion-dollar GPU-backed loans, often through special purpose vehicles. This includes pledging NVIDIA chips as collateral on facilities priced well above investment-grade credit, despite the GPUs’ estimated economic life of just two to four years.

NVIDIA’s relationship with CoreWeave goes further than supply.

It has committed to buy back up to $6.3 billion of unused compute capacity through 2032 if demand falls short. In effect, NVIDIA is acting as a buyer of last resort for capacity built on its own chips. This helps CoreWeave raise debt and expand today, while limiting its own downside if customers fail to materialise.

For CoreWeave’s lenders, that promise improves near-term confidence in utilisation. For the system as a whole, it means risk is being propped up by the chipmaker itself.

But even with that backstop, depreciation still bites: GPUs pledged as collateral can lose economic value long before the loans written against them mature.

Arun calls the resulting pattern an “extend and pretend” dynamic, where refinancing delays recognition of losses rather than resolving them, as long as lenders keep rolling exposure forward.

The model only works if two things hold at once: that utilisation stays high enough to service debt, and that lenders remain willing to refinance even as hardware ages.

Accounting choices further soften the picture. CoreWeave depreciates GPUs over six years, Nebius over four, even though engineers and project-finance lawyers put true economic life closer to three to four.

Analysts at Cerno Capital estimate that if Microsoft, Alphabet, Meta and Amazon stretched data centre asset lives to six years, reported depreciation would fall 54%, from $51 billion to $28 billion. That flatters earnings today, but does nothing to stop asset values eroding.

In short, debt, which belongs with predictable cash flows and durable assets, is being layered onto uncertain demand and fast-depreciating hardware.

The Neocloud Squeeze

Neoclouds are specialised cloud providers built almost entirely around renting high-performance GPUs.

Unlike hyperscalers, they do not bundle compute with profitable software, advertising or enterprise services.

CoreWeave, a leading neocloud, derives over 60% of its revenue from just two customers—Microsoft and NVIDIA.

That concentration matters because those customers can shift workloads, renegotiate pricing or build capacity elsewhere, exposing most of CoreWeave’s cash flows to decisions it does not control.

At the same time, hyperscalers can afford to undercut the market.

AWS has cut GPU pricing by 30-45%, absorbing margin pressure through higher-margin services that neoclouds lack. In debt terms, that price pressure directly hits coverage ratios.

With GPU-backed loans sized for much higher rental assumptions, these price cuts can push projects below debt-service thresholds, forcing renegotiation or restructuring.

When Private Risk Becomes Public Exposure

Household leverage is not the epicentre. But the exposure is not benign.

Roughly 30% of American wealth sits in equity markets. AI-related capital expenditure accounted for over 40% of the US GDP growth in 2025.

A sharp correction would still hit consumption and growth. That exposure is already visible. Microsoft disclosed that its OpenAI investment reduced net income by $3.1 billion in a single quarter, implying total OpenAI losses of roughly $11.5 billion.

Losses at the frontier do not stay contained. They move outward—into earnings, pensions and portfolios.

Barclays analysts cut earnings estimates for Alphabet, Microsoft and Meta by as much as 10%. They argue that GPU depreciation was being materially underestimated and that consensus models were overpricing these companies by 5-25%.

Source: Coatue

Risk also travels through credit.

With data centres dominating digital-infrastructure ABS and CMBS issuance, AI exposure now sits on the balance sheets of insurers, pension funds and regional banks—often concentrated among a small number of developers.

Take for example the Meta-Blue Owl Capital deal to finance, develop and operate a massive data centre campus called Hyperion in Richland Parish, Louisiana.

Meta owns only 20% of the SPV, while Blue Owl owns 80%.

The debt (financed by PIMCO) stays off Meta’s books, but it provides a residual value guarantee.

This means if the data centre’s value drops due to tech obsolescence, Meta is still on the hook to reimburse investors

Besides, a recent report from Coatue, a finance research firm, noted that the Nasdaq-100 next 12 months (NTM) price-to-earnings (P/E) multiple in 2025 stands at 28x, compared to 89x in 1999.

This indicates a more reasonable valuation, rather than the wild overpricing seen in the dot-com bubble.

However, that doesn’t change the fact that a majority of these companies—from Oracle to Blue Owl, Meta, Microsoft and others—are publicly traded, and a correction will undoubtedly affect household capital.

What a Correction Looks Like

Tenants consolidate workloads, renewals slip and capacity that looked full on paper slides into partial utilisation.

Besides, cash flows weaken just as debt rolls over. Projects that cannot refinance disappear, and others survive at lower margins.

The likely survivors are hyperscalers with diversified cash flows who can cross-subsidise losses.

The casualties are neoclouds and GPU-heavy tenants whose only product is compute.

At the macro level, the pattern is familiar. Tomohiro Hirano, an economist at Royal Holloway, University of London, told AIM that bubbles often arise during periods of uneven technological growth.

“I can easily expect that if stock bubbles collapse, it will generate a recession,” he said. “Then, to stimulate the economy, the central bank will lower the policy rate, as the [Federal Reserve Board] did after the burst of IT bubbles.”

Lower interest rates often follow, pushing capital toward leverage-heavy sectors such as housing. One of the hidden spillovers of the AI build-out is its impact on energy.

Utilities are committing to 15-year cost-recovery contracts for gas turbines and grid upgrades sized for AI data-centre loads, even though tenant demand may only be contractually secure for three to five years.

If utilisation falls, Arun warns, the gap does not vanish—it is socialised onto ratepayers.

So, are we in a reckless wave of financing?

“It will only look reckless in hindsight,” Arun said.

The post The Blind Spots of The AI Bubble appeared first on Analytics India Magazine.

In the past few months, the Big Tech firms, including Microsoft, Google and Amazon, have announced multi-billion-dollar investments to build data centres in India. But in all three cases, Karnataka is notably missing from their expansion plans.

During Microsoft CEO Satya Nadella’s recent India visit, the company announced a $17.5 billion investment, its largest in the region. The company will expand cloud and AI infrastructure, add new data centre regions, and scale capacity across existing sites in Pune, Chennai and Mumbai. A new India South Central cloud region in Hyderabad is scheduled to go live by mid-2026.

Earlier, Karnataka felt the loss when Google announced a $15 billion AI data-centre project in Visakhapatnam.

Notably, Andhra Pradesh IT and industries minister Nara Lokesh met Google CEO Sundar Pichai in San Francisco on December 10. In a post on X, he said, “We reviewed progress of their landmark $15 billion investment in the Visakhapatnam AI Data Centre, which is set to be one of the largest FDI projects.”

Pichai called the Vizag AI hub a landmark development during a call with Prime Minister Narendra Modi. He said the project will have large-scale computing power, a new subsea cable gateway, and strong energy infrastructure to support AI growth in India.

At the Bharat AI Shakti event, he stated that this would be Google’s largest AI investment outside the US, with $15 billion to be invested over five years.

On the policy and incentives front, Lokesh said that the state has created a framework that enables companies like Google to scale quickly while ensuring long-term economic benefits for the region.

Similarly, Microsoft’s investment is the biggest in Asia. During his recent tour to Bengaluru, Nadella spoke about multiple cloud regions in India, including Central and West India, but the company does not operate a data centre in Bengaluru or anywhere else in Karnataka.

What is Karnataka upto?

The IT capital did not secure any comparable hyperscale projects in 2025. The Google and AWS commitments in neighbouring states are significantly larger than any recent announcements made in Karnataka.

Karnataka’s IT Secretary N Manjula and IT Minister Priyank Kharge were unavailable for comments.

Meanwhile, Datasamudra, the data centre arm of Teleindia Datacenter, plans to invest ₹300–500 crore over the next few years to expand beyond Bengaluru. The company will build a 35–40 MW AI-focused data centre in Mangaluru, along with 5 MW edge facilities in Mysuru and Hubballi–Dharwad.

NTT announced a ₹2,400-crore investment for a new multi-data centre campus in Devanahalli near Bengaluru airport on December 3, 2025. The 8.5-acre site will house three facilities with over 60 MW capacity, including Bengaluru 4A, already operational at 22.4 MW, supported by a 220 kV substation and 48-hour backup.

The state’s other notable investments include Burkhan World Investment (BWI), which announced a ₹1,500 crore plan to set up a GPU and AI server manufacturing facility near Devanahalli, on the outskirts of Bengaluru.

Karnataka has also signed an MoU with Taiwan-based Allegiance International to establish an India–Taiwan Industrial Technology Innovation Park (ITIP). Under the agreement, Allegiance will invest ₹1,000 crore over five years to develop the park, aimed at strengthening electronics and semiconductor manufacturing.

The project is expected to create around 800 direct jobs as Taiwanese firms set up advanced manufacturing, chip design and R&D units.

Taiwanese electronics manufacturer Hon Hai Precision Industry (Foxconn) hired nearly 30,000 workers at its new iPhone assembly facility near Bengaluru over the last nine months, marking one of the fastest factory ramp-ups ever seen in India.

Besides that, the Karnataka government recently announced an investment of ₹967 crore in incentives under the new IT Policy for 2025-2030 to attract large-scale investments in AI, quantum computing, cybersecurity, and other emerging technologies.
However, during Invest Karnataka 2025, the government secured Rs 10.27 lakh crore (approximately $120 billion) in total investment commitments, including Rs 4.03 lakh crore in announced/recognised investments and Rs 6.23 lakh crore via signed MoUs.

Andhra and Telangana Firepower

Lokesh has been at loggerheads with Karnataka’s IT minister Priyank Kharge over investment claims. The two have exchanged barbs on X over Bengaluru’s infrastructure problems.

Lokesh also made headlines this month for a global roadshow. In San Francisco, he met OpenAI, AMD, Intel, NVIDIA, Canva, Adobe, Autodesk, Salesforce and others to pitch AP as a future tech hub. He proposed an AI University and free ChatGPT access for students, and invited OpenAI to base its data centre operations in Visakhapatnam.

The state has notched another major win in IT services. In December 2025, Cognizant laid the foundation for a 22-acre mega campus at Kapuluppada IT Hills in Visakhapatnam. With a total investment of around ₹1,583 crore, the three-phase project is expected to accommodate 20,000 employees, including about 8,000 direct jobs eventually.

Meanwhile, in Telangana, AWS has signed a framework to invest $7 billion in Hyderabad’s cloud infrastructure by 2030, building on its existing region launched in 2022.

The investment builds on its Hyderabad region launch in 2022 and is part of Amazon’s broader $12.7 billion commitment to expand cloud and AI infrastructure in India, announced in May 2023.

Recently, Google and the Telangana government launched the Google for Startups Hub at T-Hub in Hyderabad, featuring a dedicated space to support the state’s growing startup ecosystem.

The state recently concluded the Telangana Rising Global Summit 2025 at Bharat Future City, where it announced investment commitments totalling ₹5.75 lakh crore.

Several of the significant announcements were concentrated in capital-intensive data centre and AI infrastructure projects. Leading these was Infrakey Datacenter Parks, which signed an agreement to set up a 1 GW AI data centre with an investment of ₹70,000 crore, one of the largest single technology commitments announced in the state so far.

Duddilla Sridhar Babu, IT minister of Telangana, told AIM that both large and small tech companies are investing in the state, supported by government policies. “Hyderabad offers a strong value proposition in terms of talent depth, cost efficiency, cross-vertical domains for engineering innovation, and a strong government focus on capacity building and co-innovation,” he said.

Still hope for Karnataka

Mohan Das Pai, former CFO and board member at Infosys remains optimistic about Karnataka’s potential. “Karnataka has to define the areas it wants to dominate, like AI, quantum computing, semiconductors, and biotechnology,” he had earlier told AIM.

He urged the government to set up dedicated innovation funds and data centre infrastructure. “Set up a data centre in Mangaluru, it has a coast and could become a hub,” he suggested.

Pai suggested a spike in investment towards innovation and research. “Karnataka should spend at least ₹10,000 crore a year on innovation. With a ₹4 lakh crore budget, that’s small.”

The contrast playing out across southern India highlights how decisively infrastructure readiness, policy execution, and speed now shape big-tech investment outcomes.

As AI and data centres become increasingly capital-intensive and power-hungry, states that align land, energy, talent, and regulatory clarity into a single operating framework are pulling ahead.

For Karnataka, the challenge is no longer about brand or legacy—it is about whether it can recalibrate fast enough to compete in an AI-first investment cycle that is already moving elsewhere.

The post Why Karnataka is Missing out on Big Tech Investments appeared first on Analytics India Magazine.

In the era of digital transformation, a structural shift is underway in how economies grow and how industries contribute to national output. Gross domestic product (GDP) has been the standard candle for measuring a country’s economic output, factoring in government and consumer spending, manufacturing, exports, and services. Increasingly, however, digital infrastructure, particularly data centres and cloud computing platforms, is emerging as a foundational driver of economic value creation.

The scale of this shift is massive. Recent analysis by Harvard economist Jason Furman showed investments in data centres and cloud-related infrastructure in the US in the first half of 2025 accounted for roughly 4% of GDP. However, they contributed over 90% of GDP growth.

Earlier, Renaissance Macro Research also estimated that the dollar value contributed to GDP growth by AI data centres had surpassed US consumer spending.

For India, the story is quite different. Data centres rarely appear as consumer-facing industries. Instead, they enter national accounts as capital formation, construction, servers, networking equipment, and software.

But at a sufficient scale, the conversation changes. Market research firm Grand View Research estimated that the Indian data centre market generated around $9.17 billion in 2024, based on comparative industry analysis, and this figure is expected to reach nearly $22 billion by 2030.

Now, with cloud adoption accelerating across enterprises and government in India, this raises a key question: Is the economic contribution of digital infrastructure being systematically undercounted?

India’s Digital Infrastructure Inflexion Point

Industry leaders argue that India is already at a pivotal moment. “India’s rapid expansion in cloud adoption and digital infrastructure marks a pivotal inflexion point for economic value creation,” Dhiraj Udapure, CTO of IT firm SCS Tech India, told AIM.

“At SCS Tech, we see data centres and cloud services evolving from backend enablers into core contributors to India’s GDP over the next decade,” he added.

Multiple structural forces are driving this transition. Data localisation mandates, AI-led workloads, and the scale of India’s digital public infrastructure, such as UPI, Aadhaar, and ONDC, are increasing demand for domestic compute and secure cloud environments.

At the same time, enterprises across BFSI, manufacturing, healthcare, telecom, and government are migrating mission-critical workloads to hybrid and multi-cloud models.

“From a GDP perspective, the impact will be both direct and indirect,” Udapure explained. “Directly, data centres generate high-value capital investment, skilled employment, and recurring service revenues. Indirectly, they act as multipliers, fuelling innovation in AI, SaaS, IoT, and digital services that are exportable at scale.”

This distinction is critical. While data centre revenues alone may appear modest relative to India’s overall GDP, their downstream effects on productivity, exports, and innovation are significantly larger.

GDP Input or Economic Output?

One reason data centres remain underrepresented in GDP conversations is that much of their economic value materialises over time. The construction of a data centre contributes immediately through capital expenditure and employment. The larger impact, however, comes later, through the digital services, platforms, and productivity gains.

Productivity improvements driven by cloud computing, faster analytics, automation of routine processes, real-time decision-making, and AI deployment are diffused across sectors and only partially captured in traditional GDP accounting.

“Hybrid multi-cloud is creating impact far beyond traditional data-centre revenues by enabling modernisation, automation, and AI readiness across industries,” said Faiz Shakir, VP and MD (India and ASEAN) of cloud computing company Nutanix.
“While we don’t assign GDP figures, its role as a digital foundation is clear, helping sectors like BFSI, healthcare, manufacturing, and public services scale securely, optimise costs, and accelerate time-to-market,” he added.

Shakir emphasised that cloud infrastructure should be understood as a productivity multiplier, not merely an IT expense. As organisations adopt containerisation, AI platforms, and cloud-native architectures, new skill clusters and partner ecosystems emerge, feeding into longer-term economic competitiveness.

According to JLL, India’s installed data centre capacity has increased from roughly 350 MW in 2019 to over 1 GW by 2024, driven by cloud adoption, data localisation requirements, and rising AI workloads.

This infrastructure build-out is expected to continue as India’s AI data centre market expands rapidly. According to Grand View Research, India accounted for roughly 4.3% of the global AI data centre market in 2024, generating approximately $588.6 million in revenue in 2024.

From Support Layer to Strategic Infrastructure

Operators running India’s data centre backbone argue that the sector’s role now closely resembles that of traditional infrastructure such as power or transport.

“India’s digital infrastructure is increasingly becoming a foundational layer of the economy, and data centres sit right at the core of that transformation,” said Vipul Kumar, VP of Edge and Network, CtrlS.

Kumar noted that focusing only on direct revenues obscures the broader picture. “While data centre revenues today may appear as a relatively small line item within India’s overall GDP, their multiplier effect is already significant and will become far more visible over time.”

That multiplier spans nearly every digital sector. “Every major digital activity, cloud services, fintech, e-commerce, AI, digital public platforms, telecom, and enterprise IT, ultimately depends on resilient, scalable data centre infrastructure,” he added.

As digital adoption spreads beyond metros into tier-2, 3 cities, the role of data centres as economic enablers is also intensifying. Edge infrastructure, regional cloud availability, and domestic capacity expansion are linking digital growth more closely to local economies.

“As India’s digital economy expands across metros and non-metros alike, the economic contribution of data centres will grow not just through direct revenues, but through job creation, capital investment, energy ecosystems, and the enablement of downstream digital services,” Kumar highlighted.

Reducing the Measurement Gap

Unlike the US, where data centre investment has been identified as a visible driver of GDP growth, India’s contribution remains embedded within broader categories such as IT services and digital services. This makes the sector’s economic importance harder to quantify, even as its real-world impact accelerates.

Yet the signals are increasingly difficult to ignore: rising enterprise cloud spend, growing IT and digital services exports, expanding startup ecosystems, and sustained investment from hyperscalers and domestic players alike.

At SCS Tech, this evolution is already visible on the ground. “As cloud becomes the backbone for startups and global capability centres, India’s digital services exports are likely to see significant upside,” Udapure noted.

“With the right policy support and continued private investment, data centre and cloud ecosystems can emerge as a measurable, high-growth contributor to India’s GDP, much like IT services did in earlier decades,” he added.

As AI workloads scale, hybrid cloud adoption deepens, and data localisation strengthens, the economic footprint of data centres and cloud infrastructure in India is set to expand further. The challenge ahead lies not only in building capacity, but in recognising digital infrastructure as a long-term growth asset rather than a background utility.

The post Is Data Centre Impact Being Undervalued in India’s GDP Calculations? appeared first on Analytics India Magazine.

While there are many reasons why AI has become indispensable in today’s society, most notably productivity and efficiency, one critical dimension is often overlooked: its potential to build a more inclusive world. At its core, AI reflects the data it is trained on. What we choose to include, or exclude, ultimately determines who benefits from technological progress.

Yet, despite rapid innovation, deep structural gaps remain in the distribution of opportunity. One of the most persistently overlooked groups in this race is people with disabilities. In India, less than 1% of the corporate workforce comprises persons with disabilities, according to HR and DEI consulting firm Marching Sheep’s PwD Inclusion Index 2025, the Economic Times reported. This is a stark indicator of systemic exclusion rather than lack of talent.

The picture becomes even more troubling on closer examination. Nearly 38% of companies report having no employees with disabilities. Progress, where it exists, is slow. There has been only a 4.1% increase over the past year in organisations that have employed even a single disabled person. These numbers point not to isolated failures, but to a broader absence of enabling ecosystems.

At the heart of this problem lies a critical gap: the lack of inclusive technology capable of supporting the education, employability, and growth of people with disabilities, both within corporate environments and in public systems. Without tools designed for accessibility from the ground up, inclusion remains aspirational rather than actionable.

This exclusion is also reflected in workforce participation rates. Only about 36% of people with disabilities participate in the workforce, compared to nearly 60% of those without disabilities, a disparity that continues to define today’s labour landscape.

Additionally, national surveys show that among individuals with disabilities aged 15 and older, the labour force participation rate is around 23.8%, and the worker population ratio is about 22.8%.

Most employed individuals with disabilities work in the informal sector, often in roles like agriculture, home-based work, or casual labour. There is a notable gender gap: 47% of men with disabilities are employed, compared to just 23% of women, highlighting the compounded disadvantages of gender and disability.

From Charity to Equity

Against this backdrop, a growing number of innovators are challenging a long-held assumption: that disability inclusion is primarily a matter of charity. Instead, they argue that inclusive AI, when designed with lived experience at its centre, can become a structural equaliser, embedding access into systems rather than retrofitting it as an afterthought.

“For a long time, people with disabilities have been invisible in our cities, our schools, our offices,” Prateek Madhav, founder and CEO of AssisTech Foundation (ATF), India’s leading assistive technology (AT) innovation ecosystem, told AIM. “The question is not whether talent exists. It’s whether the system was ever built for them,” he said.

Madhav’s own journey into the disability ecosystem began long before ATF. After two decades in the corporate world, including years at Accenture in the US, he returned to India and began volunteering with disability-focused organisations. What struck him was not a lack of aspiration among people with disabilities, but their absence from public life. “We don’t see them in schools, malls, or workplaces,” he said. “That invisibility is systemic.”

ATF was born of a conviction that technology could serve as a force multiplier if built differently. “Technology doesn’t discriminate,” Madhav noted. “One application can reach millions. The same leverage that powers consumer apps can power independence and dignity.”

Lived Experience as a Design Imperative

That philosophy has shaped a new generation of assistive-technology startups that do not frame disability as a deficit, but as a design constraint worth solving for, often with benefits that extend far beyond the original user group.

For Akshita Sachdeva, co-founder and director of Trestle Labs, an assistive technology firm making education and employment digital and inclusive, that shift began with a single question asked by a young blind student. As a college student, Sachdeva had built a prototype glove that could read and describe objects for visually impaired users. When she tested it at a school in Delhi, a boy excitedly told his father he had read a newspaper independently for the first time. He asked her, “Didi, when can I get this?”

“That question stayed with me,” Sachdeva recalled. “It wasn’t about a prototype anymore. It was about access.”

Trestle co-founder Bonny Dave arrived at the same conclusion from a different starting point. Trained as a mechanical engineer, he gained early exposure to blind schools during college, which challenged his assumptions about access and privilege.

What began as a technical prototype soon evolved into a deeper realisation: meaningful innovation was not about scaling what engineers found interesting, but about identifying the right problem worth solving.

“We had to unlearn our own biases,” Dave explained, describing months spent meeting visually impaired users across cities before committing to a solution that prioritised access to content over novelty.

Trestle Labs’ approach would soon challenge a common assumption in assistive tech, that products must be niche, subsidised, and perpetually dependent on grants. Instead of selling directly to individuals, the company worked with schools, universities, and public institutions, embedding accessibility into shared infrastructure.

Designing for the Margins, Benefiting the Mainstream

“We never branded it as a ‘disability product,’” Sachdeva explained. “If it helps someone with a visual impairment, but also helps a student who doesn’t speak English fluently, that’s real inclusion.”

At Translead Medtech, the focus is not on software but on physical independence, specifically, the everyday act of sitting down and standing up safely. What began as a decade-long research project at IISc evolved into a mechanically engineered chair that requires no electricity, motors, or sensors.

“We were looking at something as basic as sit-to-stand,” Sanchit Jhunjhunwala, Translead’s co-founder, told AIM. The simple act becomes a challenge with age or disability, he said.

Translead’s innovation lies in compliant mechanisms, structures that derive function from geometry rather than electronics. While the product was initially designed for older adults and people with mobility impairments, it is now being trialled in hospitals and rehabilitation settings, extending its relevance far beyond its original use case.

“What we’re seeing is that accessibility-driven design often solves a broader problem,” Jhunjhunwala added. “It’s not charity. It’s engineering.”

Sustainability Beyond Grants and Sympathy

This shift, from charity to equity, is perhaps most visible in how founders talk about sustainability. Many reject the idea that assistive technology must be loss-making or perpetually subsidised.

“We didn’t want to be seen as a not-for-profit people feel sorry for,” Sachdeva emphasised. “We wanted to be transactional, because value was being created.”

Dave reinforced this position: “Assistive technology is often seen as charity first and business later, if at all. That framing limits scale.” From the outset, Trestle Labs chose to operate as a for-profit company, not to dilute its mission, but to strengthen it. “If a product creates value, people should be willing to pay for it, just like they do for any other technology,” Dave argued.

This approach, he believes, is what allows accessibility-first solutions to move into mainstream institutions rather than remain dependent on grants or goodwill. “Inclusion cannot survive on sympathy alone. It has to be engineered for scale.”

The same principle applies in AI-driven solutions for deaf and hard-of-hearing communities. Jayasudan Munsamy, founder and CEO of DeepVisionTech.AI, traces his motivation to personal experience.

“There was this assumption that captions solve everything,” he said. “That may be true in countries where sign language is taught formally. In India, it’s not.”

What followed was a deeper realisation: communication was only the surface problem. Education gaps, digital inaccessibility, workplace isolation, and lack of interpretation services all compounded exclusion.

DeepVisionTech’s work on sign-language interpretation tools emerged from this broader understanding, but Munsamy is careful not to overstate the role of technology. “AI is only useful if it works on the devices people actually have, in the environments they live in,” he noted.

Across these stories, a common thread emerges, which is that lived experience is not an optional input; it is the foundation of design. Whether it is a blind student struggling to access textbooks, a deaf employee unable to communicate at work, or an older adult fearful of falling, these realities shape the technology itself.

Pilot Projects to Policy and Public Infrastructure

For inclusive AI to translate into real equity, it must move beyond pilots and proofs of concept into policy-backed adoption and public infrastructure. In India, this transition is critical.

Slowly, but surely, India has strengthened its policy framework to uplift persons with disabilities, shifting from welfare-based approaches to a rights-based model of inclusion. The cornerstone of this effort is the Rights of Persons with Disabilities (RPwD) Act, 2016, which recognises 21 categories of disability and mandates non-discrimination, accessibility, and reasonable accommodation in education and employment.

The Act provides 4% reservation in government jobs, alongside obligations for inclusive infrastructure and digital access. Supporting its implementation is the Scheme for Implementation of the RPwD Act (SIPDA), which funds accessibility projects, skill development, and awareness programmes across states.

Complementing this is the ADIP scheme, which subsidises assistive devices to enhance functional independence. Other laws, such as the National Trust Act, and initiatives promoting Indian Sign Language and accessible digital services, address specific needs across disability groups.

This is where ecosystem builders such as AssisTech Foundation (ATF) intervene. “Innovation is meaningless if it doesn’t reach people,” said Madhav. Since its inception, ATF has supported over 65 startups across seven cohorts, helped develop 120+ assistive products, and worked with governments, CSR bodies, and various institutions.

For founders, policy engagement changes the conversation. Instead of being asked whether a solution can be subsidised, they are asked whether it can scale, integrate, and sustain.

When inclusive AI becomes part of procurement policies, education systems, and public infrastructure, it reshapes funding flows, accountability, and expectations. Accessibility stops being optional and begins to function as what it always should have been: essential infrastructure for a more equitable society.

The Future of AI Is Intentional

As AI continues to shape the future of work, education, and public life, we have to make sure that innovation is intentional. The data we choose, the users we centre, and the problems we deem worth solving will determine whether AI deepens existing divides or helps dismantle them.The startups and ecosystem builders working at this intersection offer a compelling alternative narrative: one where disability is not an exception to be accommodated, but a lens through which better systems are built. In doing so, they move the conversation beyond charity, towards equity, agency, and sustainable change.

The post Building an AI Economy That Includes Everyone appeared first on Analytics India Magazine.

For a country known for its IT and software talent, India’s absence in databases stands out. While Oracle, Microsoft Azure SQL, MongoDB, and Redis shaped how modern applications store and process data, no comparable database company has emerged from India. This raises deeper questions about funding, risk-taking, and the country’s innovation ecosystem.

In a wide-ranging conversation with AIM, Jayant Haritsa, professor at the Indian Institute of Science (IISc), explained why this gap exists and why closing it is far harder than it appears.

Drawing on history, systems engineering realities, and industry behaviour, Haritsa made a point that databases are not just software products but long-term infrastructure bets that demand deep expertise, patience, and trust.

A Historical Asymmetry in Knowledge

According to Haritsa, the roots of the problem lie in how the global software industry evolved before 2000.

“Until about 2000, all the major database companies were located in the US or Europe,” he explained. “The kind of work that they would send to India was work at the periphery.”

Indian engineers were tasked with testing, documentation, localisation, or building applications around databases and not with building the database engines themselves. As a result, the core intellectual capital behind database internals never fully developed in India during those formative years.

According to him, what was missing was not talent, but exposure to the deepest layers of systems engineering.

Building a Database Is Not Using One

Haritsa used a striking analogy to explain the difference between consuming technology and creating it. “Everybody in India knows how to drive a car,” he said. “But can I build a Ferrari car in India? That’s much, much harder.”

Database engineers don’t just work with queries and schemas. They need a deep understanding of operating systems, hardware architecture, memory management, concurrency, and distributed systems.

“Building database systems requires understanding not just databases, but operating systems and computer architecture, because we deal with both the hardware and software platforms,” Haritsa said.

This kind of systems-level thinking only began to spread in India after 2000, when faculty trained in the US returned, and Indian universities started teaching what was “under the hood.”

Over the past two decades, that deficit has narrowed.

Haritsa mentioned examples where India now plays a central role in global database development. Teams in Pune worked on early query optimisers, later absorbed by SAP. Microsoft’s Azure SQL development, he shared, is now largely done out of Bengaluru.

But trust inside multinational corporations does not automatically translate into Indian product companies.

The Ecosystem Problem

Even when Indian teams attempted to build database systems in the 1990s, most efforts failed not because the core engine didn’t work, but because databases are ecosystems rather than standalone products.

“The tricky part with database systems is that even if you build the engine, users expect a lot of supporting tools,” Haritsa explained.

These include schema designers, index advisors, query optimisers, monitoring tools, backup systems, and performance guarantees. Without them, an engine is unusable in real-world deployments.

“It’s like saying I give you the Ferrari engine—build the car around it,” he said. “You need the steering wheel, the brakes, the wheels—everything.”

That means millions of lines of code, years of testing, and sustained investment. Unlike consumer apps or SaaS tools, database companies cannot be built quickly.

“You need deep pockets,” Haritsa said plainly. “This is not something that can be done over a few years.”

Enterprise databases power air traffic control systems, banking infrastructure, and national payment rails. Failure is not an option.

Risk Aversion Against New Entrants

Databases sit at the heart of mission-critical systems, and CIOs are deeply conservative in choosing them.

“If I say I’m a new database system and something goes wrong tomorrow,” Haritsa noted, “everybody will criticise me and say you made a mess by using untested software.”

Choosing Oracle or Microsoft SQL Server is seen as a safe decision, even if it is expensive, because no one can be blamed for following global norms.

He explained that if an Indian database succeeds, it is applauded, but if it fails, the decision to trust it is quickly questioned. This asymmetry makes customer acquisition extraordinarily difficult for new database vendors.

Selling From India to the World Is Still Hard

Haritsa said that most database buying decisions still happen in the US.

“For an Indian company to be able to sell in the US is very difficult,” he said, adding that India historically produced services giants rather than product companies.

While this is changing slowly, breaking into global infrastructure markets requires ironclad guarantees, long-term support, and credibility built over many years—not startup-style speed.

A Decade-Long Marathon, Not a Sprint

So, when will India have its own database company? “It will take at least a decade to develop a very good database system with all the peripheral tools and robustness guarantees,” Haritsa said.

He stressed that success requires engineers who understand operating systems, architecture, distributed protocols, and end-to-end customer behaviour—combined with leadership from industry, not just academia.

“This is a marathon, not a 100-metre dash,” he added.

Looking Ahead: Quantum and LLMs as Game Changers

Interestingly, Haritsa believes the next decade could offer India a rare reset opportunity.

“There are two big game changers,” he said. “Quantum is a hardware game changer, and LLMs are a software game changer.”

Rather than copying existing systems, India could leapfrog by building databases designed from the ground up for quantum architectures and AI-driven interaction, while still preserving the non-negotiable guarantees of performance and robustness.

In Haritsa’s view, India finally has what it lacked for decades: expertise, capital, academic depth, and industry experience.

“All the ingredients are coming together,” he said. If India succeeds, it will not be because it built a faster database, but because it is committed to building infrastructure that the world can trust, over the long term.

The post Why India Still Doesn’t Have Its Own Database Company appeared first on Analytics India Magazine.

Coforge has announced a definitive agreement to acquire Encora in an all-stock transaction valued at $2.35 billion, describing the acquisition as a “defining moment” for the company as it builds capabilities in AI-led engineering, data and cloud services.

On a conference call with analysts, Coforge chief executive officer Sudhir Singh said the transaction would establish a scaled, AI-native engineering capability for the firm at a time when enterprise technology is increasingly being shaped by artificial intelligence, cloud, and data.

Singh said Coforge’s current leadership team, which came together about eight and a half years ago, has delivered one of the highest growth rates among mid- and large-cap technology services firms, driven by execution intensity, hyper-specialisation in select industries and deep capabilities in emerging technologies.

Over this period, the company’s revenue run rate has increased nearly fivefold, while its market capitalisation has grown almost twentyfold, he said.

The Encora acquisition, Singh said, is intended to build on that track record. He described the firm as becoming an approximately $2.5-billion technology services company, with a $2-billion enterprise core of AI-led engineering, data and cloud services.

Encora, founded in Silicon Valley, provides AI-native software engineering services to digital-native companies and Fortune 1000 enterprises. Its offerings span intelligent process design, agent-native product engineering, core modernisation, AI foundations, data readiness, and AI operations.

Singh highlighted Encora’s composable AI platform, AIVA, which he said enables agentic orchestration and allows organisations to compose intelligent workflows across engineering and business functions.

He outlined five attributes that Coforge believes differentiate Encora: an AI-native internal agentic platform, long-tenured relationships within large enterprises, a human-plus-agent delivery model, a talent composition aligned with AI-native engineering rather than labour arbitrage, and a services-plus-software platform model.

According to Singh, the acquisition is expected to create material scale across service lines by fiscal year 2027, with AI-led product engineering projected to become a $1.25-billion business, cloud services a $500-million business, and data engineering contributing about $250 million in revenue.

He added that the deal would immediately expand Coforge’s presence in high-tech and healthcare verticals, with each expected to reach a $170-million run rate following the acquisition.

Singh said Encora brings AI-led healthcare solutions, including biomedical research assistants, AI-enabled patient monitoring, multi-omics data ingestion, and AI foundations for clinical trials.

The acquisition will also expand Coforge’s near-shore delivery footprint in Latin America, where Encora has more than 3,100 delivery professionals, and significantly increase the company’s presence in the western and midwestern United States.

Coforge’s North America business is expected to increase by about 50% to $1.4 billion after the transaction, Singh said.

The firm will have 45 client relationships, each generating more than $10 million in annual revenue.

Encora contributes 11 such relationships, with its top 10 client relationships averaging more than a decade in tenure.

Singh cited Coforge’s track record of expanding acquired client relationships, including its earlier Cigniti acquisition.

Providing transaction details, chief financial officer Saurabh Goyal said Coforge will acquire 100% of Encora from Advent International, Warburg Pincus, and other minority shareholders.

The transaction has an enterprise value of $2.35 billion, with equity consideration of $1.89 billion to be paid through a preferential allotment of Coforge shares. The company’s board has approved raising the remaining amount through a qualified institutional placement.

Encora is estimated to generate revenues of about $600 million in FY26, with an adjusted EBITDA margin of around 19%, Goyal said.

In a separate statement, Shweta Jalan, managing partner at Advent International, said the investment reflects Advent’s approach of backing businesses and management teams to build industry-leading companies.

The post Coforge to Acquire Encora in $2.35 Bn All-Stock Mega Deal to Strengthen AI Engineering appeared first on Analytics India Magazine.

Infosys has raised entry-level salaries and is offering compensation packages of up to ₹21 lakh per annum for fresh graduates hired into specialised technology roles, making it the highest starting pay currently offered in India’s IT services sector.

The move was confirmed by Infosys Chief Human Resources Officer Shaji Mathew in a statement to The Financial Express. The story was first reported by Moneycontrol.

The company said the higher pay reflects its push towards an AI-first operating model and the need for deep digital skills at the entry level.

“Infosys continues to lead the industry in championing an AI-first approach across everything we design and deliver for our clients,” Mathew said.

“Achieving this vision requires not only upskilling our existing workforce but also infusing the organisation with digitally native talent with deep expertise and high learnability.”

He added that the company has expanded opportunities within its Specialist Programmer track. “While we have always offered multiple roles for early-career professionals, we have now expanded opportunities within the Specialist Programmer track, including compensation packages of up to ₹21 lakh per annum,” Mathew said.

Infosys is hiring freshers for roles such as Specialist Programmer and Digital Specialist Engineer (trainee).

According to Moneycontrol, compensation is structured as follows: specialist programmer L3 (trainee): ₹21 lakh per annum; specialist programmer L2: ₹16 lakh; specialist programmer L1 (trainee): ₹11 lakh; and digital specialist engineer (trainee): ₹7 lakh.

These roles are open to graduates with BE, BTech, ME, MTech, MCA and integrated MSc degrees, primarily from computer science, IT, and select circuit branches such as ECE and EEE.

The top-end package stands out at a time when fresher salaries across the IT industry have largely remained stagnant.

The company has onboarded 12,000 freshers in the first half of FY26 and expects to meet its target of 20,000 entry-level hires for the year.
Infosys CFO Jayesh Sanghrajka shared the update during the company’s Q2 earnings call on October 16. The firm also reported its fifth consecutive quarter of net headcount growth, adding 8,203 employees in Q2. Total headcount now stands at 3,31,991.

The post Infosys To Offer ₹21 lakh CTC For Freshers With Specialised Skills appeared first on Analytics India Magazine.