Sector Update

Institutional Research
India I IT
9 March 2025

IT Services NIFTY 50: 22,552

Artificial Intelligence to be a key growth driver for IT Services companies BSE Sensex: 74,333

Artificial Intelligence (AI) has transitioned from a buzzword to a core growth driver for the Stock price performance (%)*
IT Services industry. For these firms, AI is both a tool to improve their own operations and Company Name 1 Mth 3 Mth 6 Mth 1 Yr
a major service line, which can add value to their offerings. IT companies that successfully
TCS (11.7) (18.8) (19.3) (12.1)
weave AI into their offerings are likely to see higher productivity, faster growth besides
Infosys (11.1) (12.3) (12.8) 4.3
gaining access to larger transformation deals as clients accelerate investment in AI. AI can
Hcl Technologies (9.0) (19.0) (13.0) (4.9)
drive non-linear growth for IT services companies as more volume of work can be carried
Wipro (10.2) (4.2) 8.5 10.5
out without a commensurate increase in manpower. Conversely, failing to embrace AI
Tech Mahindra (10.2) (16.3) (9.0) 15.9
could render traditional outsourcing services less competitive (as automated solutions
Ltimindtree (21.6) (26.0) (23.2) (7.8)
undercut labor-intensive work). The consensus among IT industry leaders is clear - AI is not
L&T Technology Serv. (14.1) (9.3) (15.0) (9.0)
optional, it’s an imperative.
Mphasis (19.0) (24.6) (25.4) (7.6)
IT sector’s adoption of AI will pave the path to future growth. Key aspects to watch Coforge (9.3) (13.4) 17.6 23.2
out for: Happiest Minds Tech (0.7) (8.2) (13.2) (16.4)
 Productivity metrics: IT companies should be able to increase revenue, led by Source: Bloomberg, *as on 7th March 2025
productivity growth as revenue per employee increases amid growing adoption of
Rating and target prices
AI tools. Firms that manage to expand output without commensurate workforce
Company Name Rating TP (Rs)
growth (through AI augmentation) should reap the benefits of non-linear growth.
TCS ADD 4,589
 Deal wins & pipeline: There has been an increase in the mention of “AI-led deals”
Infosys BUY 2,218
in earnings calls of IT companies. Companies reporting a high proportion of new
deals involving AI/Automation likely have stronger long-term positioning. Also, HCL Technologies ADD 2076
verticals such as Banks, Healthcare, Manufacturing, Retail etc. will heavily invest in Wipro REDUCE 291
AI – IT partners entrenched in them with proven solutions will capture these spends. Tech Mahindra ADD 1,868

 Internal IP & Platforms: IT services companies can be increasingly viewed as LTIMindtree BUY 7,188
platform/IP owners. Those with successful AI platforms (like TCS Ignio, Infosys Topaz L&T Technology Serv. REDUCE 5,050
use cases, etc.) can potentially create subscription-like revenue streams, a departure Mphasis REDUCE 3,071
from pure headcount-based revenue. This could lead to valuation re-rating if Coforge BUY 10,540
executed well (as it suggests more scalable, non-linear growth). Happiest Minds ADD 769
 Talent & Training: In the short term, training thousands of staff in AI looks like an Source: Centrum Broking, *as on 7th March 2025

expensive affair. But, it’s a necessary investment. The companies doing it

Nifty 50 vs Nifty IT
aggressively (Infosys, Wipro and TCS) are basically preparing an AI-ready army –
130
which should pay off in delivery quality and innovation. This is akin to building Nifty IT
120
capacity for future; those skimping on training might face skill gaps later.
110
 Acquisitions/Partnerships: We might see acquisitions of AI startups by these IT firms 100 NIFTY 50
to bolster capabilities (similar to cloud acquisitions a decade ago). Already, Wipro 90
has an AI startup fund; Infosys invested in OpenAI (small stake) indirectly, etc. If a
80
company keeps acquiring niche AI firms, it could accelerate their expertise (though Mar-24 May-24 Jul-24 Sep-24 Nov-24 Jan-25 Mar-25
integration is key). Partnerships with hyperscalars also matter: those in deep Source: Bloomberg
partnerships (e.g., co-developing solutions with Microsoft or Google) might get
preferential access to tech and clients.
 Monetization of efficiency: Over time, as AI becomes the default industry standard,
pricing may abate (deflationary effect on some outsourcing rates). However, volume
should increase because lower costs spur demand (elasticity) and new types of
projects as clients utilize savings to further drive tech transformation. Net-net,
expect cost per unit service to drop, but volume to rise – good quality IT firms will
manage this to still expand absolute margins by taking on more work.
 Leaders with AI advantage: TCS and Infosys appear well-positioned to use AI to
cement their dominance. They have scale to invest and broad client trust. They may
enjoy higher win rates in the evolving deal mix, supporting steady revenue growth
even if the macro environment is tough. However, smaller companies also have
IT Services

been proactive with regards to their AI related investments.

Piyush Pandey
Research Analyst, IT Services
+91-022-4215 9864
piyush.pandey@centrum.co.in

Please see Appendix for analyst certifications and all other important disclosures.

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

 Global vs Indian: Accenture in particular has a very strong AI practice and is directly
competing with Indian firms for many digital deals. It has already committed US$3bn to AI
and has a robust partnership with OpenAI & others. Some investors might prefer Accenture
as it can package strategic consulting with AI implementation (plus it’s not purely
headcount-based).
 AI ecosystem stands to gain: Cloud providers (Microsoft Azure, AWS & Google Cloud) and
chip makers (Nvidia) also benefit as AI adoption by client firms increases as IT Services firms
use their infrastructure to deliver AI. For instance, Microsoft (via Azure/OpenAI) is leveraged
to the AI services boom (many AI projects will drive Azure consumption). Nvidia’s GPUs are
essentially “picks & shovels” for this gold rush. So, beyond IT sector, the AI infrastructure
stocks stand to gain.
In conclusion, AI is set to redefine the business model of IT Services. The Indian IT sector has
generally shown agility in past technology waves (e.g., cloud, digital) and is doing so again for AI
– with the added advantage of a huge skilled talent pool they are rapidly upskilling. Investors
should look for companies that have put in place a clear AI strategy, early results (case studies
or productivity metrics) and a culture of innovation as those will likely emerge as long-term
winners.
While there may be short-term disruptions (e.g., reskilling costs, some legacy service revenue
stagnation etc), the overall effect of AI should be to expand the pie of services and increase the
value they deliver to clients, thus strengthening client relationships and opening new revenue
streams. Gartner’s prediction that 80% of enterprises will use GenAI by 2026 implies a robust
pipeline of AI projects for IT Service partners in the next few years. Firms ready to capture that
demand will enjoy sustained growth.
In our coverage universe, our top picks in IT Services are Infosys followed by TCS in tier 1
category. While we prefer LTIMindtree and Coforge in tier 2 category.

Centrum Institutional Research 2

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Brief history of AI
AI ecosystem growing at fast pace

Source: Centrum Broking

AI isn't new—it started in the 1950s. Alan Turing introduced the idea of testing machine
intelligence in 1950 and John McCarthy formally launched the field in 1956 at Dartmouth.
But, the recent AI boom began ~2010, fuelled by GPUs (Graphical Processing Units)—
originally made for gaming.
Here’s how it unfolded.
Timeline of recent defining developments in AI
Year Event
2012 AlexNet showed big improvements in image recognition.
2014 GANs let AI generate realistic images and data.
2016 AlphaGo beat a Go champion using reinforcement learning.
2017 Google introduced Transformers, improving language AI.
2018 BERT improved AI’s language understanding.
2019 GPT-2 created convincing text content.
2020 GPT-3 produced human-like text at a huge scale.
2022 ChatGPT made advanced AI accessible worldwide.
2023 GPT-4 combined text and images, becoming even smarter.
Source: Centrum Broking

The past decade’s timeline is one of accelerating AI capability, powered by bigger models,
big data and better hardware.
2024 Physics Nobel Prize was awarded to developments in AI/Machine Learning

Source: NobelPrize.org

Centrum Institutional Research 3

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Timeline showing improved capabilities of AI

Source: Centrum Broking

There has been significant improvements in capabilities of AI, upgrading it from narrow
perception tasks to fluent language and creativity.

The AI landscape: Hierarchy of AI subfields

AI is about machines doing tasks that usually need human thinking.
AI includes:
 Machine Learning (ML): AI systems learn patterns from data instead of using fixed
rules.
 Deep Learning (DL): A special kind of ML using complex neural networks; behind recent
breakthroughs in vision and language tasks.
 Generative AI: Models creating original content like text or images (e.g., GPT-3, DALL-
E).
An emerging frontier is “Agentic AI”, referring to AI systems (or agents) that can
autonomously act and make decisions to achieve goals. Agentic AIs leverage generative
models (often LLMs) with reasoning and tool-use abilities to carry out multi-step tasks (for
example, an AI agent that plans and executes a series of web searches & actions to fulfill a
complex user request).
Different subfields of AI

Source: Centrum Broking

Prominent AI models (LLMs) and their comparison

Today’s AI landscape is defined by a handful of frontier models – primarily large language
models – developed by leading research labs. Given below is an overview of five prominent
AI models and comparison of their characteristics:

Centrum Institutional Research 4

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Key models:
 OpenAI’s GPT-4
 Anthropic’s Claude 2
 Google’s Gemini (next-gen model by Google DeepMind)
 Meta’s LLaMA 2
 Mistral (from startup Mistral AI).
Each AI model has unique strengths and understanding their differences is crucial for
enterprise adoption decisions.
Comparison of select AI models

Source: Artificialanalysis.ai
Comparison Table – Key Model Specs:
Model (Developer) Parameters Notable Features Primary Strengths & Use Cases
Multimodal (text+image); up to 32k- Best-in-class accuracy and reasoning; excels at complex tasks
GPT-4 (OpenAI) ~1 Trillion (est.) 128k context; Closed API model (coding, legal, etc.). Used via API/ChatGPT for high-stakes
content generation and analysis.
100k token context; Alignment- Handles long documents exceptionally well; good
focused (friendly & safe) coding/math skills. Great for enterprises needing deep
Claude 2 (Anthropic) 130B (est.)
analysis of long texts and conversational agents with a polite
tone.
Multimodal; Google AI integration; Cutting-edge contender aimed at GPT-4. Tight integration
2M context exp. with Google’s ecosystem (Cloud, Search). Expected to shine in
Gemini Ultra (Google) “Trillions” (reported)
multi-modal understanding and to be widely used via GCP in
apps (once fully released).
Open-source; Fine-tunable; 4k Open & adaptable: can be self-hosted. 70B model is strong
context (expandable with tricks) general AI, while smaller ones enable AI on edge devices. Used
LLaMA 2 (Meta) 7B / 13B / 70B
for custom LLM solutions where data control is key (finance,
gov, etc.).
Open-source (Apache 2.0); Efficient Lightweight powerhouse: 7B model matches some 13B+
training (outperforms bigger models) peers. Ideal for cost-efficient deployments and as a starting
Mistral 7B (Mistral AI) 7.3B point for domain-specific fine-tunes. Showcases that smart
design can beat sheer size. (Mistral’s upcoming larger models
aim at top-tier performance).
Source: Centrum Broking

Centrum Institutional Research 5

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Sharp improvement in technical performance of AI

Source: Stanford AI Index Report 2024

Notes on transformative models:

 GPT-4 is often considered the most transformative due to its leap in capability – e.g.
passing bar, medical exams and enabling complex automation that was considered
science fiction a few years ago. It has demonstrated “emergent abilities” (qualitatively
new skills at scale) that have convinced many of AI’s disruptive potential.
 LLaMA (and similar open models) have been transformative in another sense –
democratizing AI by allowing anyone to experiment and innovate without needing a
mega-budget.
 This open-source wave (Meta LLaMA, Falcon, Mistral, etc.) means AI tech isn’t locked
to a few firms, broadening its adoption.
 Claude has shown that AI models can be made much more context-aware (100k
tokens), which opens up new enterprise uses (feeding entire manuals to AI, etc.) - a
trend now being followed by others.
 Gemini is highly anticipated – if it delivers even partial AGI-like capabilities as hinted, it
could spur another jump in productivity.

Centrum Institutional Research 6

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Key Trends in AI (2024–2025)

Enterprise adoption surges:
AI – especially Generative AI – has moved from research hype to real enterprise pilots and
deployments.
 Companies now use AI in around two business functions (marketing, sales, software
development etc).
 59% of firms saw revenue growth and 42% cut costs through AI (McKinsey, 2024).
 AI tools like ChatGPT and Copilot boost productivity (developers code 55% faster).
 Global AI spending surged (US$13.8bn in 2023 - 6× growth vs 2022).
Enterprise AI budgets are rising accordingly – worldwide spending on Gen AI is estimated at
~US$13.8bn in 2023, up 6× from 2022, signaling that pilots are turning into production
investments.
AI for automation & productivity:
Businesses increasingly use AI to automate skilled tasks.
Growing use cases of AI
Use Case Examples

Content & Coding Automation Microsoft Copilot, Amazon CodeWhisperer

Research & Reporting JPMorgan summarizing financial reports

IT Automation Anomaly detection, automated incident fixes

Data Analysis Natural-language database queries, real-time visualizations

Source: Centrum Broking

A major trend is using AI to automate knowledge work. Beyond well-known use cases like
customer support chatbots, AI is being applied to write code (Microsoft’s Copilot & Amazon
CodeWhisperer), generate reports & slide decks, summarize meetings and draft marketing
copy – tasks usually done by skilled employees.
This trend of AI-augmented workforce is expected to continue, effectively making human
employees “centaurs” (human+AI teams) for higher efficiency.
Ethical and regulatory developments:
As AI adoption improves, AI ethics and regulation have become front-and-center.
Regulators worldwide are concerned about issues like bias, transparency and security of AI
systems. The number of AI-related regulations and laws is sharply increasing – e.g.,
 In the US, the count of active AI-related bills has gone from practically none a few years
ago to 25 laws in 2023
 The EU is moving forward with the AI Act.
 Companies like Salesforce and Microsoft have published ethical AI principles and
conduct audits of their models for bias.
On the data privacy front, Gen AI has raised concerns because models trained on web data
might leak sensitive info or copyrighted content. This has led to guidelines – e.g.,
 OpenAI and others now allow companies to opt-out their data from training
 Italy temporarily banned ChatGPT in 2023 until privacy measures were improved
We are likely to see requirements for AI transparency, bias testing and usage disclosures
becoming standard in the coming years, especially in sensitive applications (finance,
healthcare, hiring, etc.).
Other notable trends
 Democratization: Open-source AI (LLaMA & Mistral) helps smaller firms innovate.
 Multi-modal AI: GPT-4 and Gemini combine text, images and audio.
 Agentic AI: Autonomous AI workflows (AutoGPT & BabyAGI).

Centrum Institutional Research 7

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

 AI Upskilling: Companies actively train employees ("prompt engineers"), boosting

productivity and AI literacy.
This human capital development is crucial to actually realize the productivity potential
behind the current AI hype.

5. AI hype cycle: Where are we now?

The AI hype cycle, a framework by Gartner, illustrates technology adoption through five
phases: innovation trigger, peak of inflated expectations, trough of disillusionment, slope
of enlightenment and plateau of productivity.
AI has experienced multiple cycles:
 1950s-1960s: Early neural nets and expert systems triggered excitement, peaking with
predictions like Simon’s 20-year forecast, but unmet expectations led to
disillusionment.
 1980s: Expert systems, like R1, reached peak expectations, but the first AI Winter
(1974-1980s) marked a trough due to funding cuts, followed by enlightenment with
neural net revivals like Hopfield nets.
AI’s history of being “The Next Big Thing”

Source: Gartner

 2010s-present: Deep learning and big data kicked off the current cycle, with Gen AI (like
ChatGPT) at the peak of inflated expectations in 2023, projected to reach mainstream
adoption in 2-5 years.
Gen AI at peak hype:
According to Gartner (2023), Gen AI is now at the "Peak of Inflated Expectations," meaning:
 GenAI is currently highly publicized, with huge expectations
 Vendors widely promote "AI-inside," creating potential disappointment ahead
 Next stage: "Trough of Disillusionment" with practical hurdles (cost, integration &
data).
Yet, genuine productivity gains already exist—developers and workers use AI to boost
efficiency.
Bottom line:
AI might be overhyped in the short term, but it is underestimated in its long-term potential;
the most valuable applications will emerge clearly after the current hype settles.

Centrum Institutional Research 8

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Current AI hype cycle projections

Source: Gartner

The timeline of AI models

The last 5–6 years have seen an explosive evolution of large AI models, particularly in
natural language processing. Given below is the timeline of OpenAI major model releases
and innovations globally, showing how we went from early LLMs to today’s advanced
systems.
Evolution of OpenAI’s models over the years
Model Parameters Key Milestone

2018 GPT-1 117M First transformer-based LLM, introduced pre-training

Advanced coherent text generation, initial release limited due to
2019 GPT-2 1.5B
misuse concerns
Zero-shot learning, significantly raised public awareness of LLM
2020 GPT-3 175B
potential
2020 DALL-E - First multimodal model from OpenAI (text-to-image generation)

2021 Codex 12B Specialized GPT-3 for coding; basis for GitHub Copilot
ChatGPT (GPT-
2022 175B Massive adoption (100M+ users); sparked global AI acceleration
3.5)
2023 GPT-4 ~1T (est.) Multimodal, enhanced reasoning, industry-leading capabilities

2024 GPT-4 Turbo ~1T+ Increased context (128K tokens), optimized speed, advanced features
Source: Centrum Broking

Centrum Institutional Research 9

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Exponential parameter growth of GPTs

Source: OpenAI

OpenAI’s GPT models exemplify the broader AI trajectory – bigger & better, but also more
efficient over time.
Parameter growth & performance:
The GPT series has grown exponentially; GPT-1 (2018) had 117mn parameters; GPT-2 (2019)
1.5bn; GPT-3 (2020) 175bn ; and GPT-4 (2023) is estimated at 1 trillion (some sources even
suggest 1.7T).
This ~10× increase year-on-year from 2018–2023 has led to breakthroughs in capability –
e.g. GPT-3’s jump has enabled coherent long-form text and few-shot learning and GPT-4’s
jump brought human-level exam scores.
The chart below illustrates this exponential growth of model parameters in AI over time.
Exponential growth of data points used to train notable AI systems

Source: Epoch 2024

(blue line shows pre-2010 slower growth, pink line shows the post-2010 explosion). GPT models (e.g. “GPT-1” on chart) ride this
curve, culminating in trillion-plus parameter systems by 2023.

Along with scale, new emerging capabilities have surfaced –

 GPT-3 suddenly could translate languages or do arithmetic better than smaller
predecessors do.
 GPT-4 exhibited multimodal reasoning (e.g. describing why a meme image is funny – a
task requiring knowledge + vision + commonsense).

Centrum Institutional Research 10

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

The GPT trajectory has been one of scaling and refinement, leading to improved
performance on benchmarks and real tasks, now reaching a point where GPT-4 can
outperform most humans on many standardized tests.
Cost per token decline:
A less visible but crucial trend is the massive drop in the cost of AI computing.
OpenAI’s own API pricing illustrates this: at launch in 2023, GPT-4’s API cost was US$0.06
per 1K output tokens; by late 2024, OpenAI introduced optimized models (“GPT-4 Turbo /
GPT-4o”) that cut this to ~US$0.012 per 1K tokens – a drop of ~80% in price.
Projected token usage and cost per 1mn tokens over 10 years

Source: Chris Mann (Medium)

In fact, after a recent OpenAI price cut, GPT-4’s tokens cost just US$4 per million (input+
output blended), down from US$36 per million in March 2023. This is ~9× cheaper in under
1.5 years, roughly a >70% cost reduction per year. Driving this are efficiency gains like model
optimizations, better hardware utilization and economies of scale.
OpenAI’s CEO Sam Altman has noted they expect the cost of intelligence (in silico) to “fall
exponentially” – akin to a Moore’s Law for AI.
Energy and efficiency improvements:
Large models are power-hungry, but efficiency per computation is improving. For instance,
NVIDIA’s new AI chips (A100 & H100 GPUs) deliver more FLOPS per watt and techniques
like mixed-precision training cut energy use.
For GPT-type models, OpenAI and Microsoft built custom supercomputers on Azure to
maximize training efficiency. However, absolute energy usage for the largest models is still
huge – the GPT-4 training run likely used tens of thousands of gigawatt-hours, with
estimated compute cloud cost of ~US$50-100mn.

Demand for power by data centres is expected to rise significantly

Source: McKinsey (2023)

Centrum Institutional Research 11

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Yet, each model often brings inference optimizations: GPT-4’s successor (or optimized
versions) can run faster or on specialized accelerators (like AWS’s Inferentia or Google’s
TPUv5), which do the same work with less electricity.
OpenAI specifically has been refining how their models use power: they trained GPT-4 on
Microsoft Azure, which uses a significant amount of renewable energy and they’ve likely
improved utilization (so GPUs aren’t idle).

The crucial role of Cloud Computing in AI

AI’s Heavy Compute Needs = Cloud to the rescue
Training and deploying cutting-edge AI models is incredibly resource-intensive – it requires
clusters of GPUs/TPUs, high-speed networking and specialized software. The major cloud
providers (AWS, Azure & Google Cloud) have become indispensable in providing this
infrastructure at scale. For example, OpenAI’s GPT-4 was trained on Microsoft Azure’s AI
supercomputers (co-designed with Microsoft). Such a supercomputer comprised thousands
of GPUs networked together – something few companies could assemble on premise.
By leveraging Azure, OpenAI accessed virtually unlimited computing power on demand.
Similarly, Anthropic partnered with AWS as its primary cloud, using Amazon’s GPU instances
and even custom AI chips (AWS Trainium) for training models like Claude. Cloud platforms
offer ready access to petaflops of compute without organizations needing to invest tens of
millions upfront in hardware.
Increasing revenue from AI servers

Source: Bloomberg, Centrum Broking

Scalability and elasticity:

Cloud allows AI workloads to scale up or down dynamically. Training an AI model might need
1,000 GPUs for a few weeks – a burst that’s easy on cloud (just spin up more instances), but
impractical if one had to own those GPUs year-round.
Likewise for inference: an e-commerce site might see spiky usage of its AI recommendation
model during holiday sales – cloud lets it auto scale the serving backend to handle peaks
and then scale down to save cost.
This elasticity ensures optimal resource use and cost efficiency, rather than sizing on-
premise infrastructure for peak load and having it idle otherwise.
Managed AI services & ecosystem:
Cloud providers also offer a rich ecosystem of AI services that accelerate development.
 AWS has services like
○ Sagemaker (managed ML training/deployment)
○ Amazon Bedrock (access to foundation models via API)
○ Pre-built AI APIs (for vision, speech, etc.).
 Google Cloud offers Vertex AI, with tools to build and tune models besides accessing
models like PaLM and soon Gemini too.

Centrum Institutional Research 12

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

 Azure’s OpenAI Service provides easy deployment of OpenAI models with enterprise
security.
This means IT Services firms or enterprise development teams can integrate state-of-the-
art AI without needing to manage underlying infrastructure. Essentially, cloud is
democratizing AI – giving even smaller companies the compute & models needed via a
utility model.
Cost considerations – Cloud vs On-Prem:
There’s an ongoing debate: rent compute from cloud or invest in on-premise AI hardware.
 For most, cloud is preferred for AI because of the reasons given above (scalability, no
CapEx & managed support). Cloud also enables access to specialized hardware (TPUs,
Graphcore IPUs, etc.), which one might not easily procure.
 However, if a company has very large, constant AI workloads (e.g. running a stable
model 24/7 at high volume), at some point on-prem could be more cost-effective (no
provider markup). Large AI players like Meta and Google run their own datacenters for
this reason.
 However, many enterprises lack the expertise to maintain AI clusters – the cloud
providers have that expertise and offer reliability and security out of the box.
A hybrid approach is emerging too: critical workloads on-prem for data governance or cost
reasons, and cloud for burst or experimental workloads. However, even on-prem
deployments increasingly use cloud-like managed stacks (Nvidia sells “DGX Cloud” which
can be on Azure or on-prem but managed similarly).
Broadly, cloud has been a key enabler for AI boom – without it, the deep learning revolution
would be contained to only those who could afford massive compute installations. With
cloud, a lone graduate student can train a model on AWS with a credit card and a startup
can deploy an AI API globally without building datacenters.
Cloud and AI go hand-in-hand:
All three big cloud companies (AWS, Microsoft Azure & Google) are also leaders in AI –
reflecting how intertwined these domains are. Microsoft’s US$10bn+ investment in OpenAI
was as much about supercharging Azure’s AI credentials as it was about ROI on OpenAI.
Google’s AI researchers develop on Google Cloud TPUs. Amazon is integrating models into
AWS offerings and even investing in startups (like their US$4bn into Anthropic) to ensure
AWS is the “home of enterprise AI.”
For clients (like an insurance or retail company wanting AI), using cloud AI means benefiting
from this cutting-edge innovation immediately via services. It also means lower barriers to
entry: no need to hire 100s of PhDs to build a model from scratch when you can call an API
for sentiment analysis, for example.
Cloud vs On-Prem in Numbers:
Comparison of Cloud vs On-premise
Cloud On-Premise

Low upfront cost Cheaper at very large, stable scale

Elastic scaling No provider markup

Always latest-gen hardware Hardware becomes outdated quickly

Easy to manage Requires internal management

Source: Centrum Broking

Security and Data:

A consideration is data sensitivity – some firms initially hesitated to put sensitive data on
cloud for AI processing (fear of leaks or compliance issues). But, cloud providers have
addressed this with VPC isolation, encryption and even on-prem extension (Azure Stack &
AWS Outposts) for cases where data absolutely can’t leave premises. In practice, most large
enterprises are now comfortable with using cloud for AI after seeing robust security and
compliance measures.

Centrum Institutional Research 13

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

9. AI’s impact on global and Indian markets

Global AI market is soaring: The business market around AI has been experiencing rapid
growth and is poised for an even greater expansion this decade.
In 2023, the global AI market (covering software, hardware and services related to AI) was
estimated ~US$200bn in size. Forecasts vary, but all agree on a high growth trajectory. For
instance, Grand View Research projects a ~36.6% CAGR from 2024 to 2030, reaching
roughly US$1.8 trillion by 2030. Some estimates are even more bullish – Straits Research
pegs the 2030 AI market size at ~US$1.6 trillion (37% CAGR from 2021) while Fortune
Business Insights estimates its size at ~US$1.77 trillion by 2032.

Increasing revenue from AI in IT Services

Source: Bloomberg, Centrum Broking

In short, AI is one of the fastest-growing tech sectors. To put that in perspective: the AI
market is growing roughly 2–3× as fast as the broader tech sector.
Key drivers are:
 widespread enterprise adoption
 government investments
 consumer AI applications
By 2030, AI could be a multi-trillion dollar general-purpose technology market – on par with
cloud computing or perhaps even exceeding it, given AI’s reach across industries.
Demand trends globally: Every industry is exploring AI, but some are leading in spend.

Increasing capex of major tech companies in recent quarters ($mn)

Source: Bloomberg, Centrum Broking

1. Tech and Telecom companies were early and heavy investors (they built AI into their
platforms – e.g. search engines & social media feeds).
2. Financial Services is another big spender – banks and insurers use AI for fraud
detection, algorithmic trading, credit scoring, claims automation, etc. Retail and e-

Centrum Institutional Research 14

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

commerce companies invest in AI for recommendations, supply chain optimization and

customer service bots.
3. Healthcare is seeing rapid AI growth too (diagnostics, drug discovery assistance &
patient triage bots), although regulatory hurdles mean deployments can be slower.
4. A McKinsey study (2024) has found the highest value potential for Gen AI in marketing
& sales and product development functions; so, expect heavy demand there.
5. Manufacturing and Automotive sectors leverage AI for predictive maintenance, quality
control (vision inspection systems) and autonomous systems (robots & self-driving).
Even the Public Sector is increasing AI spend – smart city initiatives, defence (AI for
surveillance or logistics) and administrative process automation.
A notable trend: AI-as-a-Service – instead of bespoke projects, many companies prefer to
consume AI via APIs or platforms (like hiring an NLP service or a vision service). This makes
AI adoption easier for non-tech firms and drives market growth for those platform
providers. We also see industry-specific AI solutions: e.g., AI packages tailored for call
centers or AI analytics for retail foot traffic. These targeted solutions accelerate adoption in
mid-tier firms that wouldn’t develop AI in-house.

Impact of GenAI in time reduction

Source:McKinsey

AI’s economic impact:

Beyond direct market size, AI is expected to have a large macroeconomic impact. Various
studies (PwC & McKinsey) estimate AI could add trillions to global GDP by 2030 – PwC
famously projected ~US$15 trillion by 2030 in cumulative contribution.
However, capturing this value requires investment – which is reflected in the market growth
figures. Global private investment in AI was ~US$90bn in 2022 (Stanford AI Index) and while
VC funding cooled slightly in 2023 (after the 2021 peak), corporate investment is
accelerating.
Tech giants are pouring capital (the US$10bn+ OpenAI investment by Microsoft, Google’s
billions into AI research, etc.), and countless start-ups worldwide are focusing on AI
solutions for everything from Agriculture to Education. In sum, the global AI market growth
is a combination of direct spend on AI solutions and an amplification effect where AI
augments many other markets (cloud services, hardware sales for AI, etc.).
India’s AI market and adoption:
India, being a major IT Services and Technology talent hub, stands to benefit hugely from
AI. The Indian AI market (value of AI industry in India) was ~US$10.3bn in 2023. Although
India’s AI market is taking off from a smaller base, it is driven by both export demand and
domestic adoption. The Government of India is also pushing AI adoption as part of its

Centrum Institutional Research 15

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

“Digital India” initiative – from Smart Governance to Agriculture AI applications – which will
stimulate local market growth.
In terms of use cases in India, a few stand out.
 IT Services and BPO sector is adopting AI for automation – for example, call center AI
to augment human agents or AI to automate back-office processes (which historically
have been done by large workforces in India).
 The e-commerce and start-up ecosystem in India (Flipkart, Reliance Jio, etc.) is
embedding AI in customer experience and logistics.
 In the Manufacturing sector, companies in India use AI for predictive maintenance
given the focus on improving factory uptime and quality in the emerging “Make in
India” drive.
Market size vs adoption:
While the market revenue (spend on AI) is one measure, another is AI penetration.
NASSCOM has reported that AI investments in India in 2023 were only ~1.5% of global AI
spend. However, this is changing – many global firms’ AI R&D is being carried out in India
(e.g., Google’s AI center in Bengaluru, Microsoft Research India, etc.), and domestic start-
ups like Fractal Analytics, Postman, etc., are building AI products. Indian AI market growth
is expected to mirror global trends with a short lag – NASSCOM forecasts India’s AI growth
to be ~25–35% annually in the next few years, matching the global pace.
Industry-specific in India:

GenAI use cases for Indian IT companies

Source: Statista, Centrum Broking

The IT Services (export) sector is perhaps the single largest contributor to India’s AI revenue
– as TCS, Infosys, Wipro etc. roll out AI solutions, those count towards India’s “AI market”.
Another segment is AI services for domestic sectors like Finance and Healthcare. The
government’s use of AI is also growing – from using computer vision to monitor traffic and
compliance, to using AI in schools for personalized learning pilots.
For investors, these market trends indicate a huge growth opportunity: companies
enabling AI (cloud providers, chipmakers & AI software firms) are positioned for strong
demand, and sectors that effectively use AI should outperform peers in efficiency. In India’s
stock market, for instance, IT firms with strong AI capabilities could capture a larger share
of the expanding AI spend. The rapid CAGR is also a warning that those who fail to invest or
adopt AI could be left behind in competitiveness.

Why AI is critical for IT Services firms

AI isn’t just another tech offering for IT Services – it’s a make-or-break transformative force
for the industry’s future. Indian IT Services companies built their businesses on delivering
human talent to global clients. Now AI automation and intelligence threatens to upend
that labor-arbitrage model, but also offers huge opportunities for those who harness it
properly.
Productivity Boost = Survival

Centrum Institutional Research 16

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Key comparison of traditional and AI-driven approach in IT Services

Area Traditional Approach AI-driven Approach
Non-linear (AI boosts revenue per
Revenue Model Linear (more people = more revenue)
employee)
Client Expectations Manually intensive services AI-automated, efficient solutions
AI platforms, automation tools, specialized
Service Offerings Human talent (software coding, testing)
IP
Source: Centrum Broking

Traditionally, IT Services revenue grew linearly with headcount – more people on a project
meant more billing. However, AI can automate many tasks those people do (coding, testing,
system maintenance etc).
In other words, firms must embrace AI to boost productivity per engineer or competitors
(or clients themselves) will do so and undercut them. Embracing AI is critical not only for
efficiency but for maintaining profitability in an environment where clients expect more for
less.
Why AI is critical now?
AI directly impacts productivity and competitiveness.
 Productivity leap:
IT firms must deliver more with fewer resources. As HCL Tech's CEO says: "Companies
must now deliver twice the revenue with half the people".
 Meeting client demand:
Enterprises increasingly demand AI solutions like predictive maintenance, intelligent
automation and generative AI. IT companies without proven AI capabilities risk losing
deals to competitors or big tech providers.
 New revenue streams & differentiation:
Firms (e.g., Infosys with Topaz and TCS with Ignio) now develop reusable AI platforms,
creating higher-margin "services-as-a-software" products. This helps IT companies
move up the value chain, attract strategic clients and command higher valuations.
 Strategic cannibalization:
While automating tasks may initially reduce revenue, it positions firms to capture
larger strategic deals in the long term. IT Services companies have now become
trusted innovation partners rather than just staffing providers.
Essentially, AI is the new battleground – similar to how digital transformation was a decade
ago. If an IT firm misses the digital wave (cloud, mobile & analytics), it would have lost out;
AI is even bigger.
Additionally, there is a risk of disruption. Thus, AI is critical to IT Services because it impacts
how they deliver (productivity), what they deliver (new AI solutions) and changes the
competitive landscape. If done right, AI can significantly enhance profitability (by
automating low-value work and enabling more projects with fewer people) and unlock new
revenue (AI consulting & IP). If done wrongly (or ignored), an IT firm could see its low-value
services commoditized by AI and lose market share to more advanced competitors. Thus,
embracing AI is imperative for IT Services firms – it is the key to driving the next wave of
growth and staying indispensable to clients in the AI-first world.

Centrum Institutional Research 17

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Impact of AI on Indian IT sector

Workforce transformation
AI is significantly reshaping the Indian IT workforce. Traditionally, growth meant hiring
thousands of fresh graduates each year. Today, AI automates basic tasks such as coding,
testing and ticket handling, reducing the need for entry-level staff.
Impact radar of GenAI

Source: Gartner

For example, Indian IT companies are now focused on driving employee productivity
through the use of AI tools.
Workforce readiness for AI adoption by major Indian IT companies
Company Employees Trained in AI (FY2023-24)

TCS ~350,000 employees (~80%)

Wipro ~220,000 employees (~70%)

Infosys ~270,000 employees (~84%)

Source: Company data

This paradigm shift is not causing mass unemployment, but it is redefining roles. Employees
previously performing routine tasks are being retrained as data scientists, AI specialists or
prompt engineers, moving towards higher-value work.
Changing competitive landscape
AI capability now differentiates leaders from laggards in IT Services. Large companies (TCS,
Infosys and Accenture) with significant AI investments are pulling ahead. Platforms like
Infosys’s Topaz and TCS’s Ignio give them advantages in winning AI-centric contracts.
Smaller firms must rapidly build or partner for AI capabilities. Consulting firms like
Accenture and Deloitte are also intensifying competition, expanding aggressively into AI
implementation services.
New business models and delivery approaches
AI is driving a shift away from billing by manpower (time & material) to outcome-based
models. Clients increasingly prefer paying for delivered results rather than billed hours.
Companies leveraging AI effectively can deliver outcomes more profitably, capturing
greater market share.
National-scale upskilling and cultural shift
Indian IT firms and government initiatives are jointly driving massive AI training programs.
NASSCOM emphasizes national-level AI literacy, partnering with IITs and industry to train

Centrum Institutional Research 18

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

professionals. Such broad upskilling represents one of the largest workforce

transformations in recent history.
Critically, the industry is shifting culturally from fearing AI job loss to embracing AI as a
productivity tool that makes work more interesting and valuable.
Redefining the global delivery model
The classic offshore delivery model (client site + India offshore) is evolving to include AI bots
in the workflow. Future projects might combine onsite teams, offshore engineers and AI
systems collaborating seamlessly, significantly boosting productivity.

How AI is transforming IT service offerings

AI is reshaping IT services, transitioning from manual, labor-intensive tasks to highly
efficient, automated solutions. Here’s how AI-driven methods differ from traditional
approaches:
Traditional vs. AI-driven approaches in IT Services space
Aspect Traditional Approach AI-Driven Approach
Leveraging low-code/no-code platforms for rapid app
Service Development Extensive manual coding and programming efforts.
development.
Robotic Process Automation (RPA) for task automation and
Process Automation Human-dependent, repetitive manual tasks.
accuracy.
AI chatbots and virtual assistants for personalized
Customer Interaction Basic, scripted customer interactions.
engagement.
Data Analysis Manual, resource-intensive data processing. Real-time analytics and predictive insights powered by AI.
AI-driven product acceleration and automated quality
Product Development Lengthy cycles with manual testing.
assurance.
Source: Centrum Broking

Real-world examples of AI integration:

 Low-Code/No-Code Platforms:
Companies like Appian and Microsoft Power Platform enable businesses to rapidly
develop apps with minimal coding, empowering non-developers and accelerating
digital transformation.
 Robotic Process Automation (RPA):
UiPath provides RPA solutions that automate tasks like data entry, significantly
reducing errors and freeing human talent for more strategic tasks.
 AI-powered customer service:
Microsoft’s AI agents autonomously handle client inquiries and identify sales
opportunities, dramatically improving customer engagement and business efficiency.
 AI-enhanced application development:
Baidu’s no-code app builder, Miaoda demonstrates how AI streamlines app creation
without traditional coding, allowing for quicker deployment and customization.
 Process mining and optimization:
Appian’s process mining capabilities enable businesses to analyze, streamline and
optimize workflows, improving operational efficiency and reducing costs.
By embracing AI, IT companies can deliver agile, efficient and customized solutions, helping
them stay ahead in a rapidly evolving technological landscape.
In Summary

AI adoption is transforming the Indian IT sector—its workforce, business models,

competitive landscape and culture. Companies proactively embracing AI training and
integration are positioned for long-term leadership while slower adopters risk losing
relevance and talent.
We may see widening gaps between leaders and laggards in the coming years based on
their AI readiness.

Centrum Institutional Research 19

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Comparison of AI initiatives by Major Indian

IT firms
All leading Indian IT Service providers have jumped on the AI bandwagon, but with varying
strategies and degrees of aggressiveness. Given below is a comparison of key AI initiatives,
investments and readiness across prominent firms.
Comparison snapshot (AI initiatives)
Company AI Platform & Initiatives Investment & Training Readiness Level
Ignio (AIOps), Cognix (AI frameworks), Leader: Mature platforms, strong AI
TCS 350k employees AI-trained
Agentic AI integration
$2B AI investment, 84% workforce AI- Leader: Comprehensive suite, strategic
Infosys Topaz (Generative AI), 100+ AI agents
trained investments
ai360 initiative, HOLMES automation, $1B investment, 250k employees AI- Challenger: Big commitment, execution
Wipro
Agentic AI trained key
DRYiCE (AIOps), cost-focused AI, Strong Contender: Engineering strength,
HCLTech Significant internal upskilling
engineering AI steady progress
Moderate: Telecom niche focus, unique
Tech Mahindra GAiA (network AIOps), custom LLMs AI lab, $100M fund, moderate training
IP
Mosaic platform, rapid generative AI Fast Follower: Nimble, cost-effective AI
LTIMindtree Reduced hiring due to productivity gains
deployment adoption
AiKno™ (industrial AI), IoT & predictive Specialist Leader: Engineering and
LTTS Engineer-focused AI training
solutions industrial domain
NextLabs cognitive platform (BFSI- Moderate upskilling, strategic fintech
Mphasis Moderate: BFSI specialization
focused AI) partnerships
Domain-specific training and
Coforge AI solutions for insurance, airlines Fast Follower: Domain-specific offerings
partnerships
Dedicated GenAI unit, agile AI Innovator: Agile, quick AI adoption,
Happiest Minds Near-total workforce AI-fluent
deployments niche player
Source: Centrum Broking

The table illustrates that TCS and Infosys are clearly front-runners with comprehensive
programs and demonstrated scale, followed by Wipro and HCL Tech as strong contenders.
The mid-tier players like LTIMindtree and Coforge are punching above their weight by
focusing on agility and domain-specific AI, whereas none of the major firms is truly lagging
in an absolute sense (everyone is at least on the train). The differentiation will sharpen over
the next 1-2 years as real case studies and AI-driven revenue start to tell the story of who’s
capitalizing most on the AI era.

13. Conclusion and key takeaways

AI’s pivotal role in IT Services sector’s growth: AI has transitioned from a buzzword to a
core growth driver for IT Services. For these firms, AI is both a tool to improve their own
operations and a major new service line to sell. Companies that successfully weave AI into
their offerings are likely to see higher productivity, protecting or expanding margins,
besides also gaining access to larger transformation deals as clients invest in AI. Conversely,
failing to embrace AI could render traditional outsourcing services less competitive (as
automated solutions undercut labor-intensive work). The consensus among industry
leaders is clear - AI is not optional, it’s an imperative.
From an investor perspective, the Indian IT sector’s adoption of AI will determine future
winners and losers.
In conclusion, AI is set to redefine the business model of IT Services. The Indian IT sector
has generally shown agility in past technology waves (e.g., cloud, digital) and is doing so
again for AI – with the added advantage of a huge skilled talent pool they are rapidly
upskilling. Institutional investors should look for companies that have put in place a clear AI
strategy, delivered early results (case studies or productivity metrics) and has a culture of
innovation, as those will likely emerge as long-term winners.

Centrum Institutional Research 20

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

While there may be short-term disruptions (e.g., reskilling costs, some legacy service
revenue stagnation), the overall effect of AI should be to expand the pie of services the IT
firms can offer and increase the value they deliver to clients, thus strengthening client
relationships and opening new revenue streams. Gartner’s prediction that 80% of
enterprises will use GenAI by 2026implies a robust pipeline of AI projects for IT service
partners in the next few years. Firms ready to capture that demand will enjoy sustained
growth.
AI is not just hype for these IT companies – it is becoming ingrained in how they operate
and what they sell. The Indian IT firms that combine domain expertise with AI prowess will
effectively move up the value chain, deserve higher valuation multiples and should see
resilient growth even as traditional low-end services face automation. Monitoring each
firm’s AI execution will be key, but the trend is unequivocal: AI will be the engine driving IT
services industry growth in the coming decade, much like outsourcing was in the 2000s and
digital transformation was in the 2010s. Investing in the AI leaders of IT Services could thus
be a rewarding strategy as this technology cycle plays out.
Centrum IT Universe
EPS EPS PE PE PE
Ticker Price(Rs) Target Price(Rs) Rating EPS FY25E
FY26E FY27E FY25E FY26E FY27E
TCS IN 3,611 4,589 ADD 136.2 155.2 176.5 26.5 23.3 20.5
INFO IN 1,686 2,218 BUY 65.0 74.9 86.7 25.9 22.5 19.4
HCLT IN 1,558 2,076 ADD 65.9 73.3 83.1 23.6 21.3 18.8
WPRO IN 285 2,91 REDUCE 12.3 13.7 14.5 23.2 20.8 19.6
TECHM IN 1,493 1,868 ADD 49.8 70.0 84.9 30.0 21.3 17.6
LTIM IN 4,721 7,188 BUY 169.9 212.0 239.5 27.8 22.3 19.7
LTTS IN 4,820 5,050 REDUCE 123.6 163.5 180.4 39.0 29.5 26.7
MPHL IN 2,301 3,071 REDUCE 90.9 103.5 118.1 25.3 22.2 19.5
COFORGE IN 7,708 10,540 BUY 145.2 216.1 277.4 53.1 35.7 27.8
HAPPSTMN IN 695 769 ADD 14.2 22.7 27.5 48.8 30.6 25.3
Source: Company, Centrum Broking

Centrum Institutional Research 21

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Disclaimer
Centrum Broking Limited (“Centrum”) is a full-service, Stock Broking Company and a member of The Stock Exchange, Mumbai (BSE) and National Stock Exchange of
India Ltd. (NSE). Our holding company, Centrum Capital Ltd, is an investment banker and an underwriter of securities. As a group Centrum has Investment Banking,
Advisory and other business relationships with a significant percentage of the companies covered by our Research Group. Our research professionals provide important
inputs into the Group's Investment Banking and other business selection processes.
Recipients of this report should assume that our Group is seeking or may seek or will seek Investment Banking, advisory, project finance or other businesses and may
receive commission, brokerage, fees or other compensation from the company or companies that are the subject of this material/report. Our Company and Group
companies and their officers, directors and employees, including the analysts and others involved in the preparation or issuance of this material and their dependants,
may on the date of this report or from, time to time have "long" or "short" positions in, act as principal in, and buy or sell the securities or derivatives thereof of
companies mentioned herein. Centrum or its affiliates do not own 1% or more in the equity of this company Our sales people, dealers, traders and other professionals
may provide oral or written market commentary or trading strategies to our clients that reflect opinions that are contrary to the opinions expressed herein, and our
proprietary trading and investing businesses may make investment decisions that are inconsistent with the recommendations expressed herein. We may have earlier
issued or may issue in future reports on the companies covered herein with recommendations/ information inconsistent or different those made in this report. In
reviewing this document, you should be aware that any or all of the foregoing, among other things, may give rise to or potential conflicts of interest. We and our Group
may rely on information barriers, such as "Chinese Walls" to control the flow of information contained in one or more areas within us, or other areas, units, groups or
affiliates of Centrum. Centrum or its affiliates do not make a market in the security of the company for which this report or any report was written. Further, Centrum
or its affiliates did not make a market in the subject company’s securities at the time that the research report was published.
This report is for information purposes only and this document/material should not be construed as an offer to sell or the solicitation of an offer to buy, purchase or
subscribe to any securities, and neither this document nor anything contained herein shall form the basis of or be relied upon in connection with any contract or
commitment whatsoever. This document does not solicit any action based on the material contained herein. It is for the general information of the clients of Centrum.
Though disseminated to clients simultaneously, not all clients may receive this report at the same time. Centrum will not treat recipients as clients by virtue of their
receiving this report. It does not constitute a personal recommendation or take into account the particular investment objectives, financial situations, or needs of
individual clients. Similarly, this document does not have regard to the specific investment objectives, financial situation/circumstances and the particular needs of
any specific person who may receive this document. The securities discussed in this report may not be suitable for all investors. The securities described herein may
not be eligible for sale in all jurisdictions or to all categories of investors. The countries in which the companies mentioned in this report are organized may have
restrictions on investments, voting rights or dealings in securities by nationals of other countries. The appropriateness of a particular investment or strategy will
depend on an investor's individual circumstances and objectives. Persons who may receive this document should consider and independently evaluate whether it is
suitable for his/ her/their particular circumstances and, if necessary, seek professional/financial advice. Any such person shall be responsible for conducting
his/her/their own investigation and analysis of the information contained or referred to in this document and of evaluating the merits and risks involved in the securities
forming the subject matter of this document.
The projections and forecasts described in this report were based upon a number of estimates and assumptions and are inherently subject to significant uncertainties
and contingencies. Projections and forecasts are necessarily speculative in nature, and it can be expected that one or more of the estimates on which the projections
and forecasts were based will not materialize or will vary significantly from actual results, and such variances will likely increase over time. All projections and forecasts
described in this report have been prepared solely by the authors of this report independently of the Company. These projections and forecasts were not prepared
with a view toward compliance with published guidelines or generally accepted accounting principles. No independent accountants have expressed an opinion or any
other form of assurance on these projections or forecasts. You should not regard the inclusion of the projections and forecasts described herein as a representation
or warranty by or on behalf of the Company, Centrum, the authors of this report or any other person that these projections or forecasts or their underlying assumptions
will be achieved. For these reasons, you should only consider the projections and forecasts described in this report after carefully evaluating all of the information in
this report, including the assumptions underlying such projections and forecasts.
The price and value of the investments referred to in this document/material and the income from them may go down as well as up, and investors may realize losses
on any investments. Past performance is not a guide for future performance. Future returns are not guaranteed and a loss of original capital may occur. Actual results
may differ materially from those set forth in projections. Forward-looking statements are not predictions and may be subject to change without notice. Centrum does
not provide tax advice to its clients, and all investors are strongly advised to consult regarding any potential investment. Centrum and its affiliates accept no liabilities
for any loss or damage of any kind arising out of the use of this report. Foreign currencies denominated securities are subject to fluctuations in exchange rates that
could have an adverse effect on the value or price of or income derived from the investment. In addition, investors in securities such as ADRs, the value of which are
influenced by foreign currencies effectively assume currency risk. Certain transactions including those involving futures, options, and other derivatives as well as non-
investment-grade securities give rise to substantial risk and are not suitable for all investors. Please ensure that you have read and understood the current risk
disclosure documents before entering into any derivative transactions.
This report/document has been prepared by Centrum, based upon information available to the public and sources, believed to be reliable. No representation or
warranty, express or implied is made that it is accurate or complete. Centrum has reviewed the report and, in so far as it includes current or historical information, it
is believed to be reliable, although its accuracy and completeness cannot be guaranteed. The opinions expressed in this document/material are subject to change
without notice and have no obligation to tell you when opinions or information in this report change.
This report or recommendations or information contained herein do/does not constitute or purport to constitute investment advice in publicly accessible media and
should not be reproduced, transmitted or published by the recipient. The report is for the use and consumption of the recipient only. This publication may not be
distributed to the public used by the public media without the express written consent of Centrum. This report or any portion hereof may not be printed, sold or
distributed without the written consent of Centrum.
Registration granted by SEBI and certification from NISM in no way guarantee performance of the intermediary or provide any assurance of returns to investors.
“Investment in securities market are subject to market risks. Read all the related documents carefully before investing.”.
The distribution of this document in other jurisdictions may be restricted by law, and persons into whose possession this document comes should inform themselves
about, and observe, any such restrictions. Neither Centrum nor its directors, employees, agents or representatives shall be liable for any damages whether direct or
indirect, incidental, special or consequential including lost revenue or lost profits that may arise from or in connection with the use of the information.
This document does not constitute an offer or invitation to subscribe for or purchase or deal in any securities and neither this document nor anything contained herein
shall form the basis of any contract or commitment whatsoever. This document is strictly confidential and is being furnished to you solely for your information, may
not be distributed to the press or other media and may not be reproduced or redistributed to any other person. The distribution of this report in other jurisdictions
may be restricted by law and persons into whose possession this report comes should inform themselves about, and observe any such restrictions. By accepting this
report, you agree to be bound by the fore going limitations. No representation is made that this report is accurate or complete.
The opinions and projections expressed herein are entirely those of the author and are given as part of the normal research activity of Centrum Broking and are given
as of this date and are subject to change without notice. Any opinion estimate or projection herein constitutes a view as of the date of this report and there can be no
assurance that future results or events will be consistent with any such opinions, estimate or projection.
This document has not been prepared by or in conjunction with or on behalf of or at the instigation of, or by arrangement with the company or any of its directors or
any other person. Information in this document must not be relied upon as having been authorized or approved by the company or its directors or any other person.
Any opinions and projections contained herein are entirely those of the authors. None of the company or its directors or any other person accepts any liability
whatsoever for any loss arising from any use of this document or its contents or otherwise arising in connection therewith.
Centrum and its affiliates have not managed or co-managed a public offering for the subject company in the preceding twelve months. Centrum and affiliates have
not received compensation from the companies mentioned in the report during the period preceding twelve months from the date of this report for service in respect
of public offerings, corporate finance, debt restructuring, investment banking or other advisory services in a merger/acquisition or some other sort of specific
transaction.

Centrum Institutional Research 22

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

As per the declarations given by Piyush Pandey, research analyst and and/or any of their family members do not serve as an officer, director or any way connected to
the company/companies mentioned in this report. Further, as declared by them, they are not received any compensation from the above companies in the preceding
twelve months. They do not hold any shares by them or through their relatives or in case if holds the shares then will not to do any transactions in the said scrip for
30 days from the date of release such report. Our entire research professionals are our employees and are paid a salary. They do not have any other material conflict
of interest of the research analyst or member of which the research analyst knows of has reason to know at the time of publication of the research report or at the
time of the public appearance.
While we would endeavour to update the information herein on a reasonable basis, Centrum, its associated companies, their directors and employees are under no
obligation to update or keep the information current. Also, there may be regulatory, compliance or other reasons that may prevent Centrum from doing so.
Non-rated securities indicate that rating on a particular security has been suspended temporarily and such suspension is in compliance with applicable regulations
and/or Centrum policies, in circumstances where Centrum is acting in an advisory capacity to this company, or any certain other circumstances.
This report is not directed to, or intended for distribution to or use by, any person or entity who is a citizen or resident of or located in any locality, state, country or
other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation or which would subject Centrum Broking Limited or
its group companies to any registration or licensing requirement within such jurisdiction. Specifically, this document does not constitute an offer to or solicitation to
any U.S. person for the purchase or sale of any financial instrument or as an official confirmation of any transaction to any U.S. person unless otherwise stated, this
message should not be construed as official confirmation of any transaction. No part of this document may be distributed in Canada or used by private customers in
United Kingdom. The information contained herein is not intended for publication or distribution or circulation in any manner whatsoever and any unauthorized
reading, dissemination, distribution or copying of this communication is prohibited unless otherwise expressly authorized. Please ensure that you have read “Risk
Disclosure Document for Capital Market and Derivatives Segments” as prescribed by Securities and Exchange Board of India before investing in Indian Securities Market.

Ratings definitions
Our ratings denote the following 12-month forecast returns:
Buy – The stock is expected to return above 15%.
Add – The stock is expected to return 5-15%.
Reduce – The stock is expected to deliver -5-+5% returns.
Sell – The stock is expected to deliver <-5% returns.

Centrum Institutional Research 23

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.
 IT Services 9 March 2025

Disclosure of Interest Statement

1 Business activities of Centrum Broking Centrum Broking Limited (hereinafter referred to as “CBL”) is a registered member of NSE (Cash, F&O and Currency Derivatives
Limited (CBL) Segments), MCX-SX (Currency Derivatives Segment) and BSE (Cash segment), Depository Participant of CDSL and a SEBI registered
Portfolio Manager.
2 Details of Disciplinary History of CBL CBL has not been debarred/ suspended by SEBI or any other regulatory authority from accessing /dealing in securities market.

3 Registration status of CBL: CBL is registered with SEBI as a Research Analyst (SEBI Registration No. INH000001469)

Member (NSE and BSE). Member MSEI (Inactive)

Single SEBI Regn. No.: INZ000205331

Depository Participant (DP)

CDSL DP ID: 120 – 12200
Single SEBI Regn. No.: IN-DP-537-2020

PORTFOLIO MANAGER

SEBI REGN NO.: INP000004383

Research Analyst
SEBI Registration No. INH000001469

Mutual Fund Distributor

AMFI REGN No. ARN- 147569

Website: www.centrumbroking.com
Investor Grievance Email ID: investor.grievances@centrum.co.in

Compliance Officer Details:

Ajay S Bendkhale
(022) 4215 9000/9023; Email ID: compliance@centrum.co.in

Centrum Broking Ltd. (CIN :U67120MH1994PLC078125)

Registered and Corporate Office:
Level -9, Centrum House, C.S.T. Road,
Vidyanagari Marg, Kalina,
Santacruz (East) Mumbai – 400098
Tel.: - +91 22 4215 9000

Centrum Institutional Research 24

Powered by TCPDF (www.tcpdf.org)

Downloaded by in-iimcalcutta from 115.240.20.86 at 2025-08-16 09:25:40 UTC. EMIS. Unauthorized Distribution Prohibited.