Рубрика: AI

AI world news.

Indian IT’s AI Turning Point: Embedded Today, Modular Tomorrow

Indian IT services firms have embedded AI capabilities—AI infused into delivery, engineering, operations, testing, and managed services. Infosys’ Topaz platform, TCS’ MFDM (Machine First Delivery Model), Ignio AI operations, Wipro Lab45 are a few examples, among others.

Very few firms have dedicated P&L or org structures around AI/Gen AI enterprise application business, and hence, few have translated these capabilities into visible revenue outcomes, said Gaurav Vasu, CEO & founder of UnearthInsight, a cognitive intelligence platform that provides business & financial intelligence on Indian Startups.

TCS and HCLTech are the only firms recognising AI-led revenue. This is a result of past product and platform business exposure, where independent AI solutions are sold, outside the current services deals construct.

Embedded AI is largely invisible, running in the background to reduce effort, improve cycle time, and boost quality. Clients now expect AI/GenAI clauses in deals that promise faster outcomes, lower costs, and predictable delivery.

Tech service providers are embedding AI in their delivery models to create efficiency, cost, and time-to-market impact for their clients, said Yugal Joshi, partner at global research firm Everest Group. “The productivity promises of 20–30% have become common,” he added.

But here’s the catch: despite the AI push, most real money still comes from traditional levers, large contract wins, multi-year deal renewals, and price-based expansions. Embedded AI keeps Indian IT competitive, but its monetisation potential is yet to unfurl.

Why Cling to Embedded AI?

Globally, Modular AI — reusable, auditable AI building blocks — is gaining momentum. But, Indian IT continues to prioritise embedded AI. Deepak Dastrala, CTO at Intellect Design Arena, explained why.

“The hesitation is less about technology and more about the prevailing business model. Indian IT is optimised for people, projects, and billable hours, not products and IP. Embedded AI fits the traditional services model perfectly.”

Modular AI demands upfront investment, a product mindset, and sometimes saying no to custom client work which is a tough shift for firms built on services scale. Limited access to high-end compute, curated datasets, and research-heavy talent adds friction.

Jayaprakash Nair, head of AI and analytics at Altimetrik, a digital business services company, pointed this out. “Creating foundational models demands sustained GPU capacity and significant capital expenditure. Fine-tuning existing models remains the practical approach.”

This is also why many internal AI accelerators, copilots, and automation engines never get productised. Indian IT has built powerful internal AI tools, but few have been converted into modular products, or subscription offerings that influence pricing or deal structures.

Embedded AI fits the current business model, and that is both its strength and limitation.

The Hidden Limits

Embedded AI boosts efficiency, but it also creates complexity.

Each project ends up with its own prompts, integrations, guardrails, and custom workflows. Over time, this produces technical debt and governance headaches. When regulations shift or base models change, you are forced to patch 10 to 50 different places. Observability and safety become a game of whack-a-mole.

As Dastrala noted,“A module reused across 10 clients reduces new-build revenue. Firms fear standardisation because it cannibalises services revenue.”

Without measurable metrics, embedded AI rarely becomes a deal-winning differentiator. Vasu highlighted, “Embedded AI is still under the hood. Firms need observability dashboards that show real-time impact on cycle time, effort saved, defects avoided, and SLA improvements.”

This lack of visibility means clients continue awarding mega-deals and renewing large contracts based on transformation scale and pricing, not on embedded AI value, because it isn’t packaged or quantified well.

Internal transformation is another challenge. Embedding AI requires rethinking workforce planning, updating delivery assets, and preparing teams for automation-driven role changes, a cultural shift that moves slower than expected.

Starting Embedded a Smart Move?

Despite its limitations, embedded AI is a logical first step. It integrates directly into existing workflows and delivers early, low-risk wins. As Joshi put it,“Starting with embedded AI helps companies deliver faster outcomes and validate impact. It creates confidence before they scale with modular AI.”

A modular foundation of reusable agents, policy checkers, KYC summarisers and governance modules helps solve long-term problems such as technical debt, versioning, security, and cross-client standardisation.

Dastrala called this hybrid approach: “Modular at the core, embedded at the edge.” This lets firms scale proven modules across clients, while customising only the last mile.

Embedded AI should evolve into a client-facing differentiator with clear ROI attribution and deal-linked monetisation, added Vasu.

Modular architectures also allow upgrades without tearing apart legacy systems. Nair noted that embedded solutions don’t always win when scaling or managing complexity. “Modular architectures reduce risk by separating functionality into components that can be upgraded independently,” he said.

The path is clear: embedded for early wins, modular for sustainable scale.

The Future: Modular at the Core

Clients won’t choose between embedded and modular AI, they will demand both. Modular AI provides governance, security, and reusability; embedded AI ensures contextual, workflow-specific integration.

“Once the API buzz fades, clients will ask platform-oriented questions. Providers who remain purely embedded will get trapped in project debt,” warned Dastrala.

Indian IT firms that can show measurable AI impact — reduced cycle time, predictable SLAs, fewer defects — already enjoy higher win ratios in transformation and managed services deals.

“But, clients still treat it [embedded AI] as table stakes because it isn’t standardised or productised,” noted Vasu.

This is where modularity becomes commercially essential. Firms that link AI impact to pricing, renewals, and value-based deals will move away from volume-led revenue and into outcome-led monetisation.

Nair summed up the opportunity: “Reusable modular blocks on top of proven embedded AI delivery models let clients start small, iterate fast, and embed fully once value is proven.”

The firms that productise their embedded intelligence — and combine it with modular building blocks — will shape how enterprises adopt AI at scale.

Conclusion

Indian IT’s AI strategy is at an inflection point. Embedded AI has delivered efficiency and client trust. But, long-term growth will come from productised, modular AI that compounds across clients, shapes deal structures, and influences pricing.

The winning formula is now clear: start embedded, scale modular, productise everything that delivers repeatable value.

Indian IT must evolve embedded AI from an internal efficiency tool into a visible, governable, monetisable product layer, because in the next wave of global technology spending, clients will expect nothing less.

The post Indian IT’s AI Turning Point: Embedded Today, Modular Tomorrow appeared first on Analytics India Magazine.

Indian IT’s AI Turning Point: Embedded Today, Modular Tomorrow

Indian IT services firms have embedded AI capabilities—AI infused into delivery, engineering, operations, testing, and managed services. Infosys’ Topaz platform, TCS’ MFDM (Machine First Delivery Model), Ignio AI operations, Wipro Lab45 are a few examples, among others.

Very few firms have dedicated P&L or org structures around AI/Gen AI enterprise application business, and hence, few have translated these capabilities into visible revenue outcomes, said Gaurav Vasu, CEO & founder of UnearthInsight, a cognitive intelligence platform that provides business & financial intelligence on Indian Startups.

TCS and HCLTech are the only firms recognising AI-led revenue. This is a result of past product and platform business exposure, where independent AI solutions are sold, outside the current services deals construct.

Embedded AI is largely invisible, running in the background to reduce effort, improve cycle time, and boost quality. Clients now expect AI/GenAI clauses in deals that promise faster outcomes, lower costs, and predictable delivery.

Tech service providers are embedding AI in their delivery models to create efficiency, cost, and time-to-market impact for their clients, said Yugal Joshi, partner at global research firm Everest Group. “The productivity promises of 20–30% have become common,” he added.

But here’s the catch: despite the AI push, most real money still comes from traditional levers, large contract wins, multi-year deal renewals, and price-based expansions. Embedded AI keeps Indian IT competitive, but its monetisation potential is yet to unfurl.

Why Cling to Embedded AI?

Globally, Modular AI — reusable, auditable AI building blocks — is gaining momentum. But, Indian IT continues to prioritise embedded AI. Deepak Dastrala, CTO at Intellect Design Arena, explained why.

“The hesitation is less about technology and more about the prevailing business model. Indian IT is optimised for people, projects, and billable hours, not products and IP. Embedded AI fits the traditional services model perfectly.”

Modular AI demands upfront investment, a product mindset, and sometimes saying no to custom client work which is a tough shift for firms built on services scale. Limited access to high-end compute, curated datasets, and research-heavy talent adds friction.

Jayaprakash Nair, head of AI and analytics at Altimetrik, a digital business services company, pointed this out. “Creating foundational models demands sustained GPU capacity and significant capital expenditure. Fine-tuning existing models remains the practical approach.”

This is also why many internal AI accelerators, copilots, and automation engines never get productised. Indian IT has built powerful internal AI tools, but few have been converted into modular products, or subscription offerings that influence pricing or deal structures.

Embedded AI fits the current business model, and that is both its strength and limitation.

The Hidden Limits

Embedded AI boosts efficiency, but it also creates complexity.

Each project ends up with its own prompts, integrations, guardrails, and custom workflows. Over time, this produces technical debt and governance headaches. When regulations shift or base models change, you are forced to patch 10 to 50 different places. Observability and safety become a game of whack-a-mole.

As Dastrala noted,“A module reused across 10 clients reduces new-build revenue. Firms fear standardisation because it cannibalises services revenue.”

Without measurable metrics, embedded AI rarely becomes a deal-winning differentiator. Vasu highlighted, “Embedded AI is still under the hood. Firms need observability dashboards that show real-time impact on cycle time, effort saved, defects avoided, and SLA improvements.”

This lack of visibility means clients continue awarding mega-deals and renewing large contracts based on transformation scale and pricing, not on embedded AI value, because it isn’t packaged or quantified well.

Internal transformation is another challenge. Embedding AI requires rethinking workforce planning, updating delivery assets, and preparing teams for automation-driven role changes, a cultural shift that moves slower than expected.

Starting Embedded a Smart Move?

Despite its limitations, embedded AI is a logical first step. It integrates directly into existing workflows and delivers early, low-risk wins. As Joshi put it,“Starting with embedded AI helps companies deliver faster outcomes and validate impact. It creates confidence before they scale with modular AI.”

A modular foundation of reusable agents, policy checkers, KYC summarisers and governance modules helps solve long-term problems such as technical debt, versioning, security, and cross-client standardisation.

Dastrala called this hybrid approach: “Modular at the core, embedded at the edge.” This lets firms scale proven modules across clients, while customising only the last mile.

Embedded AI should evolve into a client-facing differentiator with clear ROI attribution and deal-linked monetisation, added Vasu.

Modular architectures also allow upgrades without tearing apart legacy systems. Nair noted that embedded solutions don’t always win when scaling or managing complexity. “Modular architectures reduce risk by separating functionality into components that can be upgraded independently,” he said.

The path is clear: embedded for early wins, modular for sustainable scale.

The Future: Modular at the Core

Clients won’t choose between embedded and modular AI, they will demand both. Modular AI provides governance, security, and reusability; embedded AI ensures contextual, workflow-specific integration.

“Once the API buzz fades, clients will ask platform-oriented questions. Providers who remain purely embedded will get trapped in project debt,” warned Dastrala.

Indian IT firms that can show measurable AI impact — reduced cycle time, predictable SLAs, fewer defects — already enjoy higher win ratios in transformation and managed services deals.

“But, clients still treat it [embedded AI] as table stakes because it isn’t standardised or productised,” noted Vasu.

This is where modularity becomes commercially essential. Firms that link AI impact to pricing, renewals, and value-based deals will move away from volume-led revenue and into outcome-led monetisation.

Nair summed up the opportunity: “Reusable modular blocks on top of proven embedded AI delivery models let clients start small, iterate fast, and embed fully once value is proven.”

The firms that productise their embedded intelligence — and combine it with modular building blocks — will shape how enterprises adopt AI at scale.

Conclusion

Indian IT’s AI strategy is at an inflection point. Embedded AI has delivered efficiency and client trust. But, long-term growth will come from productised, modular AI that compounds across clients, shapes deal structures, and influences pricing.

The winning formula is now clear: start embedded, scale modular, productise everything that delivers repeatable value.

Indian IT must evolve embedded AI from an internal efficiency tool into a visible, governable, monetisable product layer, because in the next wave of global technology spending, clients will expect nothing less.

The post Indian IT’s AI Turning Point: Embedded Today, Modular Tomorrow appeared first on Analytics India Magazine.

Indian IT’s AI Turning Point: Embedded Today, Modular Tomorrow

Indian IT services firms have embedded AI capabilities—AI infused into delivery, engineering, operations, testing, and managed services. Infosys’ Topaz platform, TCS’ MFDM (Machine First Delivery Model), Ignio AI operations, Wipro Lab45 are a few examples, among others.

Very few firms have dedicated P&L or org structures around AI/Gen AI enterprise application business, and hence, few have translated these capabilities into visible revenue outcomes, said Gaurav Vasu, CEO & founder of UnearthInsight, a cognitive intelligence platform that provides business & financial intelligence on Indian Startups.

TCS and HCLTech are the only firms recognising AI-led revenue. This is a result of past product and platform business exposure, where independent AI solutions are sold, outside the current services deals construct.

Embedded AI is largely invisible, running in the background to reduce effort, improve cycle time, and boost quality. Clients now expect AI/GenAI clauses in deals that promise faster outcomes, lower costs, and predictable delivery.

Tech service providers are embedding AI in their delivery models to create efficiency, cost, and time-to-market impact for their clients, said Yugal Joshi, partner at global research firm Everest Group. “The productivity promises of 20–30% have become common,” he added.

But here’s the catch: despite the AI push, most real money still comes from traditional levers, large contract wins, multi-year deal renewals, and price-based expansions. Embedded AI keeps Indian IT competitive, but its monetisation potential is yet to unfurl.

Why Cling to Embedded AI?

Globally, Modular AI — reusable, auditable AI building blocks — is gaining momentum. But, Indian IT continues to prioritise embedded AI. Deepak Dastrala, CTO at Intellect Design Arena, explained why.

“The hesitation is less about technology and more about the prevailing business model. Indian IT is optimised for people, projects, and billable hours, not products and IP. Embedded AI fits the traditional services model perfectly.”

Modular AI demands upfront investment, a product mindset, and sometimes saying no to custom client work which is a tough shift for firms built on services scale. Limited access to high-end compute, curated datasets, and research-heavy talent adds friction.

Jayaprakash Nair, head of AI and analytics at Altimetrik, a digital business services company, pointed this out. “Creating foundational models demands sustained GPU capacity and significant capital expenditure. Fine-tuning existing models remains the practical approach.”

This is also why many internal AI accelerators, copilots, and automation engines never get productised. Indian IT has built powerful internal AI tools, but few have been converted into modular products, or subscription offerings that influence pricing or deal structures.

Embedded AI fits the current business model, and that is both its strength and limitation.

The Hidden Limits

Embedded AI boosts efficiency, but it also creates complexity.

Each project ends up with its own prompts, integrations, guardrails, and custom workflows. Over time, this produces technical debt and governance headaches. When regulations shift or base models change, you are forced to patch 10 to 50 different places. Observability and safety become a game of whack-a-mole.

As Dastrala noted,“A module reused across 10 clients reduces new-build revenue. Firms fear standardisation because it cannibalises services revenue.”

Without measurable metrics, embedded AI rarely becomes a deal-winning differentiator. Vasu highlighted, “Embedded AI is still under the hood. Firms need observability dashboards that show real-time impact on cycle time, effort saved, defects avoided, and SLA improvements.”

This lack of visibility means clients continue awarding mega-deals and renewing large contracts based on transformation scale and pricing, not on embedded AI value, because it isn’t packaged or quantified well.

Internal transformation is another challenge. Embedding AI requires rethinking workforce planning, updating delivery assets, and preparing teams for automation-driven role changes, a cultural shift that moves slower than expected.

Starting Embedded a Smart Move?

Despite its limitations, embedded AI is a logical first step. It integrates directly into existing workflows and delivers early, low-risk wins. As Joshi put it,“Starting with embedded AI helps companies deliver faster outcomes and validate impact. It creates confidence before they scale with modular AI.”

A modular foundation of reusable agents, policy checkers, KYC summarisers and governance modules helps solve long-term problems such as technical debt, versioning, security, and cross-client standardisation.

Dastrala called this hybrid approach: “Modular at the core, embedded at the edge.” This lets firms scale proven modules across clients, while customising only the last mile.

Embedded AI should evolve into a client-facing differentiator with clear ROI attribution and deal-linked monetisation, added Vasu.

Modular architectures also allow upgrades without tearing apart legacy systems. Nair noted that embedded solutions don’t always win when scaling or managing complexity. “Modular architectures reduce risk by separating functionality into components that can be upgraded independently,” he said.

The path is clear: embedded for early wins, modular for sustainable scale.

The Future: Modular at the Core

Clients won’t choose between embedded and modular AI, they will demand both. Modular AI provides governance, security, and reusability; embedded AI ensures contextual, workflow-specific integration.

“Once the API buzz fades, clients will ask platform-oriented questions. Providers who remain purely embedded will get trapped in project debt,” warned Dastrala.

Indian IT firms that can show measurable AI impact — reduced cycle time, predictable SLAs, fewer defects — already enjoy higher win ratios in transformation and managed services deals.

“But, clients still treat it [embedded AI] as table stakes because it isn’t standardised or productised,” noted Vasu.

This is where modularity becomes commercially essential. Firms that link AI impact to pricing, renewals, and value-based deals will move away from volume-led revenue and into outcome-led monetisation.

Nair summed up the opportunity: “Reusable modular blocks on top of proven embedded AI delivery models let clients start small, iterate fast, and embed fully once value is proven.”

The firms that productise their embedded intelligence — and combine it with modular building blocks — will shape how enterprises adopt AI at scale.

Conclusion

Indian IT’s AI strategy is at an inflection point. Embedded AI has delivered efficiency and client trust. But, long-term growth will come from productised, modular AI that compounds across clients, shapes deal structures, and influences pricing.

The winning formula is now clear: start embedded, scale modular, productise everything that delivers repeatable value.

Indian IT must evolve embedded AI from an internal efficiency tool into a visible, governable, monetisable product layer, because in the next wave of global technology spending, clients will expect nothing less.

The post Indian IT’s AI Turning Point: Embedded Today, Modular Tomorrow appeared first on Analytics India Magazine.

Disney Plans $1 Bn Investment in OpenAI, to Bring Over 200 Characters to Sora

Even as the entertainment industry is reshaping, with Netflix announcing plans to acquire Warner Bros., The Walt Disney Company on Thursday announced it has entered into a three-year licensing agreement with OpenAI. This will allow Sora, OpenAI’s generative video platform, to create short, user-prompted social videos featuring more than 200 characters from Disney, Marvel, Pixar and Star Wars.

The deal also includes a $1 billion equity investment in OpenAI, along with warrants to purchase additional equity. The transaction remains subject to definitive agreements, corporate and board approvals, and other closing conditions. Sora and ChatGPT Images are expected to begin generating Disney-licensed content in early 2026.

Under the agreement, Sora will generate short videos based on user prompts, using a library of animated, masked and creature characters, as well as costumes, props, vehicles and iconic environments.

The licence does not include any talent likenesses or voices. ChatGPT Images will also be able to generate images using the same intellectual property.

Disney+ will stream a curated selection of fan-inspired Sora videos, while both companies collaborate to build new subscriber experiences using OpenAI’s models. Disney will adopt OpenAI APIs across products, tools and internal workflows, including ChatGPT for employees.

Among the characters fans will be able to use are Mickey Mouse, Minnie Mouse, Lilo, Stitch, Ariel, Belle, Beast, Cinderella, Baymax, Simba and Mufasa, as well as characters from Encanto, Frozen, Inside Out, Moana, Monsters Inc., Toy Story, Up and Zootopia.

The licence will also include animated or illustrated versions of Marvel and Lucasfilm characters such as Black Panther, Captain America, Deadpool, Groot, Iron Man, Loki, Thor, Thanos, Darth Vader, Han Solo, Luke Skywalker, Leia, the Mandalorian, Stormtroopers and Yoda.

Both companies said the partnership includes a framework for responsible AI use. They said it focuses on user safety, protection of creator rights, age-appropriate policies, and systems to prevent illegal or harmful content.

“Technological innovation has continually shaped the evolution of entertainment, bringing with it new ways to create and share great stories with the world,” said Robert Iger, CEO of The Walt Disney Company.

“Bringing together Disney’s iconic stories and characters with OpenAI’s groundbreaking technology puts imagination and creativity directly into the hands of Disney fans in ways we’ve never seen before, giving them richer and more personal ways to connect with the Disney characters and stories they love.”

Sam Altman, co-founder and CEO of OpenAI, said, “Disney is the global gold standard for storytelling, and we’re excited to partner to allow Sora and ChatGPT Images to expand the way people create and experience great content.”

The post Disney Plans $1 Bn Investment in OpenAI, to Bring Over 200 Characters to Sora appeared first on Analytics India Magazine.

Meet Satya Nadella, The Developer Hiding in Plain Sight

Microsoft CEO Satya Nadella stepped onto the Bengaluru stage not like a global tech leader carrying the weight of one of a $3.5 trillion company in the AI race, but like an engineer who simply couldn’t wait to show off a side project that he’d been hacking for hours.

Like some young developer out there exploring what wonders AI can do, Nadella has quietly been doing the same. His day doesn’t start with board meetings, but with AI assistant Copilot on his phone.

He wakes up, sets off agents, and then heads to work. He keeps coming back to it, only to see if the tasks been completed or what improvements the agents have made. “It’s just fun to be able to play with GitHub and just be constantly modifying,” Nadella said.

While the world was shutting off for Thanksgiving Day, Nadella started building what he calls an LLM Council, a multi-agent reasoning system. It has been inspired by Andrej Karpathy’s reasoning system that lets AI models debate inside an app.

Was fun to be at a dev event in Bengaluru and demo an app I built recently for deep research with multiple models and decision frameworks…think of it as "chain of debate"… Next stop, Copilot!! pic.twitter.com/oiinGN05BA

— Satya Nadella (@satyanadella) December 11, 2025

Creating one like a builder, he says, as a platform company, it’s always exciting to see the builders bring their creative and entrepreneurial energy. He even showed the app to Indian billionaire Gautam Adani, a day after meeting Prime Minister Narendra Modi.

Nadella shared that he works inside a Windows 365 environment using GitHub Codespaces, where his setup automatically generates several draft branches each morning. He reviews and deletes most of them by the end of the day, but usually keeps a piece or two that proves useful. Over time, those small daily experiments eventually grew into the decision-making framework he is now building.

On the model front, he said he relies heavily on Codex-Max because of its speed. Nadella said he wanted to go beyond building a simple deep-research tool, which pushed him toward the idea of the LLM Council. He even joked that all this work was really his attempt to get a job on the Copilot team.

What is LLM Council?

The LLM Council brings multiple models together. “We have all of these models available to you — GPT, Claude Opus, Gemini, Grok,” Nadella said. Users can pick one model to serve as the chair, with the rest acting as council members. The group then works like a selection committee, debating and weighing the query before producing an answer.

He compared the process to a chain of thought, but called it a ‘chain of debate’ instead.

For the demo, Nadella showed the council debated an all-time Indian Test cricket team, with different models offering contrasting judgments on openers, bowlers and captaincy. He explained that the system can spot problems such as era bias in cricket analysis, and that each model helps correct and refine the reasoning.

He then introduced DxO, a framework adapted from earlier healthcare research known as MAI-DxO. MAI-DxO stands for Microsoft AI Diagnostic Orchestrator, a model-agnostic multi-agent system originally built for medical diagnostics. “DxO is a thing that we implemented in healthcare first,” Nadella said. In this setup, Claude Opus leads the research, GPT-5.1 reviews for gaps and bias, and Gemini contributes domain and data analysis.

Nadella gave an example of a decision-making framework called Ensemble. In this setup, all the available AI models are used at once, but their identities are hidden. Instead of knowing whether a response came from GPT, Gemini, Claude, or Grok, the system anonymises them by assigning neutral labels like Alpha, Beta, Gamma, etc.

By removing model identities, the system prevents bias toward any specific model. All the responses are then combined, or synthesised, into a final single answer, so that the evaluation stays neutral.

Meta Cognition

Across all three frameworks, Nadella emphasised what he called a new form of user-level metacognition. “To me, this is the next generation of metacognition,” he said.

Although the agents produce research, critique and synthesis, he noted that “the metacognition is still us,” and these systems function as tools that strengthen human oversight.

His interface showed what he calls a “chain of debate,” allowing users to watch the models challenge each other and improve the answers before giving a final recommendation.

While Nadella used cricket as a demonstration, he said the same structures apply to complex decisions in supply chains, healthcare, finance and enterprise planning. “These types of chains of debates with multiple agents participating are going to be a lot of what we are all going to build,” he said.

Nadella also spoke about Agent365, Microsoft’s runtime layer for securing agentic systems. When he attempted to deploy his own agent into Microsoft’s tenant, the system blocked it.

He explained that Agent365 requires admin approval even for internally built agents. He described the runtime as essential for governance, observability and compliance in a world where enterprises deploy many autonomous agents.

He concluded by mentioning the momentum he is seeing across industries. Companies such as Cognizant, Persistent Systems, Swiggy and others are already building multi-agent orchestration frameworks. These examples, Nadella said, signal how rapidly agentic architectures are spreading through enterprise workflows.

The post Meet Satya Nadella, The Developer Hiding in Plain Sight appeared first on Analytics India Magazine.

Indian IT’s AI Turning Point: Embedded Today, Modular Tomorrow

Indian IT services firms have embedded AI capabilities—AI infused into delivery, engineering, operations, testing, and managed services. Infosys’ Topaz platform, TCS’ MFDM (Machine First Delivery Model), Ignio AI operations, Wipro Lab45 are a few examples, among others.

Very few firms have dedicated P&L or org structures around AI/Gen AI enterprise application business, and hence, few have translated these capabilities into visible revenue outcomes, said Gaurav Vasu, CEO & founder of UnearthInsight, a cognitive intelligence platform that provides business & financial intelligence on Indian Startups.

TCS and HCLTech are the only firms recognising AI-led revenue. This is a result of past product and platform business exposure, where independent AI solutions are sold, outside the current services deals construct.

Embedded AI is largely invisible, running in the background to reduce effort, improve cycle time, and boost quality. Clients now expect AI/GenAI clauses in deals that promise faster outcomes, lower costs, and predictable delivery.

Tech service providers are embedding AI in their delivery models to create efficiency, cost, and time-to-market impact for their clients, said Yugal Joshi, partner at global research firm Everest Group. “The productivity promises of 20–30% have become common,” he added.

But here’s the catch: despite the AI push, most real money still comes from traditional levers, large contract wins, multi-year deal renewals, and price-based expansions. Embedded AI keeps Indian IT competitive, but its monetisation potential is yet to unfurl.

Why Cling to Embedded AI?

Globally, Modular AI — reusable, auditable AI building blocks — is gaining momentum. But, Indian IT continues to prioritise embedded AI. Deepak Dastrala, CTO at Intellect Design Arena, explained why.

“The hesitation is less about technology and more about the prevailing business model. Indian IT is optimised for people, projects, and billable hours, not products and IP. Embedded AI fits the traditional services model perfectly.”

Modular AI demands upfront investment, a product mindset, and sometimes saying no to custom client work which is a tough shift for firms built on services scale. Limited access to high-end compute, curated datasets, and research-heavy talent adds friction.

Jayaprakash Nair, head of AI and analytics at Altimetrik, a digital business services company, pointed this out. “Creating foundational models demands sustained GPU capacity and significant capital expenditure. Fine-tuning existing models remains the practical approach.”

This is also why many internal AI accelerators, copilots, and automation engines never get productised. Indian IT has built powerful internal AI tools, but few have been converted into modular products, or subscription offerings that influence pricing or deal structures.

Embedded AI fits the current business model, and that is both its strength and limitation.

The Hidden Limits

Embedded AI boosts efficiency, but it also creates complexity.

Each project ends up with its own prompts, integrations, guardrails, and custom workflows. Over time, this produces technical debt and governance headaches. When regulations shift or base models change, you are forced to patch 10 to 50 different places. Observability and safety become a game of whack-a-mole.

As Dastrala noted,“A module reused across 10 clients reduces new-build revenue. Firms fear standardisation because it cannibalises services revenue.”

Without measurable metrics, embedded AI rarely becomes a deal-winning differentiator. Vasu highlighted, “Embedded AI is still under the hood. Firms need observability dashboards that show real-time impact on cycle time, effort saved, defects avoided, and SLA improvements.”

This lack of visibility means clients continue awarding mega-deals and renewing large contracts based on transformation scale and pricing, not on embedded AI value, because it isn’t packaged or quantified well.

Internal transformation is another challenge. Embedding AI requires rethinking workforce planning, updating delivery assets, and preparing teams for automation-driven role changes, a cultural shift that moves slower than expected.

Starting Embedded a Smart Move?

Despite its limitations, embedded AI is a logical first step. It integrates directly into existing workflows and delivers early, low-risk wins. As Joshi put it,“Starting with embedded AI helps companies deliver faster outcomes and validate impact. It creates confidence before they scale with modular AI.”

A modular foundation of reusable agents, policy checkers, KYC summarisers and governance modules helps solve long-term problems such as technical debt, versioning, security, and cross-client standardisation.

Dastrala called this hybrid approach: “Modular at the core, embedded at the edge.” This lets firms scale proven modules across clients, while customising only the last mile.

Embedded AI should evolve into a client-facing differentiator with clear ROI attribution and deal-linked monetisation, added Vasu.

Modular architectures also allow upgrades without tearing apart legacy systems. Nair noted that embedded solutions don’t always win when scaling or managing complexity. “Modular architectures reduce risk by separating functionality into components that can be upgraded independently,” he said.

The path is clear: embedded for early wins, modular for sustainable scale.

The Future: Modular at the Core

Clients won’t choose between embedded and modular AI, they will demand both. Modular AI provides governance, security, and reusability; embedded AI ensures contextual, workflow-specific integration.

“Once the API buzz fades, clients will ask platform-oriented questions. Providers who remain purely embedded will get trapped in project debt,” warned Dastrala.

Indian IT firms that can show measurable AI impact — reduced cycle time, predictable SLAs, fewer defects — already enjoy higher win ratios in transformation and managed services deals.

“But, clients still treat it [embedded AI] as table stakes because it isn’t standardised or productised,” noted Vasu.

This is where modularity becomes commercially essential. Firms that link AI impact to pricing, renewals, and value-based deals will move away from volume-led revenue and into outcome-led monetisation.

Nair summed up the opportunity: “Reusable modular blocks on top of proven embedded AI delivery models let clients start small, iterate fast, and embed fully once value is proven.”

The firms that productise their embedded intelligence — and combine it with modular building blocks — will shape how enterprises adopt AI at scale.

Conclusion

Indian IT’s AI strategy is at an inflection point. Embedded AI has delivered efficiency and client trust. But, long-term growth will come from productised, modular AI that compounds across clients, shapes deal structures, and influences pricing.

The winning formula is now clear: start embedded, scale modular, productise everything that delivers repeatable value.

Indian IT must evolve embedded AI from an internal efficiency tool into a visible, governable, monetisable product layer, because in the next wave of global technology spending, clients will expect nothing less.

The post Indian IT’s AI Turning Point: Embedded Today, Modular Tomorrow appeared first on Analytics India Magazine.

NeoSapien raises $2M to Boost AI Wearable Ecosystem

NeoSapien, the AI-native wearable startup, has closed a $2 million seed round led by Merak Ventures. Individual investors, including Awais Ahmed (founder, Pixxel), Anupam Mittal (founder, Shaadi.com), Sameer Mehta (co-founder, boAt), Namita Thapar (Emcure Pharmaceuticals) and Aprameya Radhakrishna (CEO, Koo), participated in the round.

The startup was founded in 2024 by brothers Dhananjay Yadav and Aryan Yadav.

It said the fresh capital will be used to accelerate product development, boost market visibility and expand its team across functions. NeoSapien’s flagship product, Neo 1, acts as an AI assistant that captures everyday conversations and actions in real time.

Unlike basic voice tools that handle only one task at a time, Neo 1 works in the background, remembers past interactions, understands context as it builds, and generates insights on its own.

The startup highlighted the product’s multilingual capability and said it supports more than 100 languages, from Kannada to Mandarin. It combines contextual intelligence with custom hardware to process interactions in real time.

The founders aim to build an operating system for AI assistants across various form factors such as glasses, pendants, watches, and rings, with NeoOS as the core software. Its NeoCore SDK will allow businesses and developers to create applications on top of this intelligence layer.

“We’re building technology that fades into the background so you can stay present in your life,” Dhananjay Yadav said in a statement. He further observed that the mental energy people spend trying to remember past conversations, retain long-term context, and integrate disparate data is considerable. And hence their product Neo 1 handles that seamlessly, freeing up human intelligence for what actually matters, Dhananjay noted.

Lead investor Merak Ventures believes that NeoSapien is by far the top contender in this segment. “Seldom has India had a chance to participate in a global, tectonic shift, but with PAIAs (Personal AI Assistant), it has a real shot at doing so. NeoSapien is by far the top contender to take India there,” Sheetal Bahl, founding partner at Merak Ventures, said.
He acknowledged that the founders have achieved something incredible by bringing a consumer product from India to market in less than a year with a paltry sum of money.

“We believe NeoOS will power hundreds of millions of upcoming AI consumer wearables sold in India and across the world every year,” he said.

NeoSapien is positioning itself as a next-generation AI company focused on wearable intelligence, blending adaptive technology with human intuition.

With its proprietary “Second Brain” OS and privacy-first approach, the startup aims to make AI-native wearables a global standard, transforming how people think, work, and live in the age of context-aware AI.

The post NeoSapien raises $2M to Boost AI Wearable Ecosystem appeared first on Analytics India Magazine.

10 Notable Executives Who Left Apple in 2025

Despite a wave of rumours, Johny Srouji, Apple’s senior vice president of hardware technologies, has made it clear he has no intention of leaving the company.

The speculation triggered genuine concern online, given Srouji’s key role in shaping Apple’s silicon strategy and spearheading the ARM-based M-series revolution that reshaped the Mac lineup and set a new course for Apple’s hardware trajectory.

His reassurance, however, doesn’t change the broader picture.

Apple has seen a notable amount of executive turnover in 2025. Some long-standing leaders have retired, others have been recruited by competitors, and several transitions were deliberately initiated as Apple restructured to strengthen its position in AI.

Alan Dye: Departs Apple After 19 Years, Joins Meta as Chief Design Officer

Recently, Bloomberg reported that Alan Dye, who has served as the head of Apple’s user interface (UI) design team since 2015, has now been ‘poached’ by Meta.

Citing sources familiar with the matter, the report added that Apple is replacing Dye with Stephen Lemay, a longtime UI designer at the company.

Meta is reportedly creating a new design studio and appointing Dye to lead the design of hardware, software and AI integration for its UI.

Dye was instrumental in defining the look and the feel of Apple’s most recent software products. As per his LinkedIn profile, he has been a part of Apple since 2006.

John Giannandrea: Stepped Down After 7 Years Leading AI Strategy, To Retire in 2026

Apple announced that John Giannandrea, senior vice president for machine learning and AI strategy, stepped down from his role in December and moved into an advisory position ahead of his planned retirement in Spring 2026.

Apple said Giannandrea had been responsible for building out Apple’s modern AI organisation since joining the company in 2018, overseeing foundation models, search and knowledge, core ML research and AI infrastructure.

The remaining parts of his organisation were redistributed to Sabih Khan and Eddy Cue, while Apple hired Amar Subramanya—previously a senior AI leader at Microsoft and Google—to assume the role of vice president of AI and report directly to Craig Federighi, Apple’s senior vice president of software engineering.

Jeff Williams: Retires After 26 Years, Steps Down as COO

In July, Apple announced that Jeff Williams will step down as chief operating officer as part of a long-planned transition, with Sabih Khan taking over the role. Williams will continue reporting to CEO Tim Cook until the former’s retirement, slated later in 2025, while still overseeing Apple’s design team, Apple Watch and Health initiatives.

Williams spent 27 years at the company, shaping its global supply chain, leading supplier responsibility programmes, and playing key roles in launching the iPod, iPhone and Apple Watch. His transition marks one of the most consequential leadership handovers at Apple in recent years.

Robby Walker: Leaves AI/Search Team After Long Tenure, Exits in Late 2025

Bloomberg reported that Robby Walker, one of Apple’s senior-most AI executives, left the company in late 2025. Walker had been a direct reportee to John Giannandrea and previously ran Siri, before oversight of the assistant was reassigned to Craig Federighi earlier in the year.

After that shift, he became the senior director responsible for Apple’s Answers, Knowledge and Information (AKI) team and helped build a new AI-powered search system intended to compete with Perplexity and ChatGPT.

The report added that Walker had remained an influential voice inside Apple’s AI organisation despite his responsibilities being gradually reduced, making his exit another meaningful loss for the company’s already strained AI efforts.

Kate Adams: Retiring After 7 Years as General Counsel

Apple confirmed that Kate Adams, who had served as general counsel since 2017, will retire, with her transition extending into late 2026.

Apple said Adams had been a central figure in shaping the company’s legal strategy, consistently advocating for user privacy and defending Apple’s ability to innovate across contentious regulatory environments.

During her eight-year tenure, she oversaw some of Apple’s most complex legal and policy challenges and later took on interim oversight of the Government Affairs organisation as part of a broader leadership reshuffle.

Apple stated that her responsibilities would ultimately transfer to Jennifer Newstead, who will assume the combined role of SVP, general counsel and government affairs in 2026.

Lisa Jackson: Retiring After 12 Years Leading Environment and Policy Efforts

Apple also announced that Lisa Jackson, vice president for environment, policy and social initiatives, will step down in late January 2026 after more than a decade of leading the company’s sustainability and government-relations work.

Apple noted that Jackson had been instrumental in reducing Apple’s global emissions by over 60% since 2015 and had been a key figure in Apple’s engagements with governments worldwide on issues ranging from privacy to accessibility.

Jackson’s portfolio—spanning climate strategy, environmental programs and social initiatives—became a model for integrating sustainability into Apple’s core operations.

With her departure, the environment and social initiatives groups were reorganised under COO Sabih Khan, completing one of Apple’s most significant policy-leadership transitions of the year.

Ruoming Pang: Departs AI Group, Moves to Meta’s Superintelligence Lab

Ruoming Pang, the manager in charge of Apple’s foundational models team, joined Meta in July after spending four years at the company.

As per a Bloomberg report, Meta offered him a package worth tens of millions of dollars per year.

Pang’s LinkedIn profile shows that he began his role as a ‘distinguished software engineer’ at Apple in 2021, after completing his 15-year tenure at Google.

At Google, he worked on Google Brain’s speech recognition research and the development of leading deep learning frameworks used by TPUs, the company’s in-house hardware

Tom Gunter: Senior LLM Researcher Departs After 8 Years, Joins Meta’s Superintelligence Push

In July, Tom Gunter, one of Apple’s senior-most LLM researchers, left the company after approximately eight years. He joined Meta as an AI research scientist.

Gunter was a key member of Apple’s Foundation Models team and is regarded internally as difficult to replace due to his specialised expertise in large-scale LLM systems.

His departure came amid escalating talent pressure inside Apple’s AI group, as Meta had been offering dramatically higher compensation packages to recruit researchers for its Superintelligence Labs unit.

A Bloomberg report notes that his exit intensified concerns about morale and retention within Apple’s core AI organisation, which has been strained by uncertainty around the company’s shifting Siri and generative-AI strategy.

Ke Yang: Leaves Apple Weeks After Promotion, Joins Meta’s Superintelligence Labs

Ke Yang, the executive recently appointed to lead Apple’s AI-driven web search initiative, left the company in October to join Meta. As per his LinkedIn profile, he spent six years at Apple as the senior director of machine learning.

Reports state that Yang had been appointed only weeks earlier to lead the AKI division—the team responsible for building ChatGPT-like web search features for Siri and for the broader AI reboot Apple planned for early 2026.

Before taking over the AKI group, Yang had led the search-focused components of the team and became a direct report to John Giannandrea following the departure of AKI senior director Robby Walker, both of whom are no longer with Apple.

Jian Zhang: Exits Robotics AI Team to Join Meta’s Robotics Studio

Jian Zhang, Apple’s lead researcher for AI-driven robotics, left the company in September to join Meta’s Robotics Studio. He spent 10 years at Apple as the head of robotics research in AI and machine learning.

According to Bloomberg, Zhang led a small team of academics within Apple’s AI and Machine Learning division, focusing on automation technologies and the role of AI in Apple’s early-stage robotics concepts. These include devices Bloomberg previously described as part of Apple’s long-term hardware pipeline.

At Meta, Zhang moved to a robotics organisation developing hardware and software for next-generation AI-powered devices, reinforcing Meta’s aggressive push to hire Apple’s top AI researchers at the time.

Note: The list above reflects some of the most notable shifts, but they do not represent an exhaustive account of every leadership change.

The post 10 Notable Executives Who Left Apple in 2025 appeared first on Analytics India Magazine.

Microsoft Names Cognizant, Infosys, TCS, Wipro as “Frontier Firms” for Copilot Deployment

Microsoft will expand partnerships with Cognizant, Infosys, Tata Consultancy Services, and Wipro, positioning the four Indian IT majors as “frontier firms” in the global adoption of agentic AI, its chairman and CEO Satya Nadella announced during his AI tour in Bengaluru.

Each company will deploy more than 50,000 Microsoft Copilot licenses, collectively exceeding 200,000 seats, the company said.

The announcement comes a day after Microsoft unveiled plans to invest $17.5 billion in cloud and AI infrastructure, skills and operations in India over the next four years.
According to the company, these partnerships will enable enterprises to enhance productivity, efficiency and accessibility, while accelerating AI-powered innovation across industries.

Microsoft said embedding AI into core operations will position these Indian IT majors as ‘frontier firms’ that not only adopt AI early but also redesign workflows around human-agent collaboration to drive measurable impact across delivery, sales, finance, HR, and customer engagement.

Puneet Chandok, president, Microsoft India & South Asia, said the four companies “are moving beyond experimentation to full-scale deployment, embedding Microsoft Copilot into the fabric of everyday work.”

Microsoft has also significantly expanded its partnership with Cognizant, calling the company its “client zero” for Copilot.

The collaboration enables Cognizant to refine Copilot and agentic solutions for large-scale enterprise use, going beyond productivity improvements to fundamentally change how organisations access data, make decisions, and scale innovation.

Cognizant CEO Ravi Kumar S, in a statement, said the technology sector is witnessing “a historic, largest infrastructure investment, with companies investing hundreds of billions annually into AI infrastructure.”

He added, “As an AI builder company, our mission is to bridge the gap between these investments and extract business value, ensuring our associates and clients benefit from Generative AI.”

Infosys, which has one of Microsoft’s largest Copilot deployments across the world, is integrating Microsoft’s intelligence layer into Infosys Topaz Fabric and Infosys Cobalt to operationalise multi-agent workflows.

CEO and MD Salil Parekh said deploying Copilot at scale and embedding AI into the Topaz operating model is enabling Infosys to “shift from traditional workflows to a human-plus-agent powered AI-first enterprise.”

TCS is working with Microsoft to transform sales, HR and finance processes by democratising tools such as the 365 Copilot and GitHub Copilot across its global workforce.

The company said all employees now have a personalised AI coach.

Microsoft also recently hosted a global hackathon to deep dive into AI agent development, with participation from more than 281,000 employees.

“TCS has equipped tens of thousands of its professionals with Microsoft AI solutions. Microsoft Cloud, data, and AI technologies are integral to our business transformation,” TCS CEO and MD K Krithivasan said.

Wipro has entered into a three-year partnership with Microsoft and launched a Microsoft Innovation Hub at its Partner Labs in Bengaluru.

With more than 50,000 Copilot licenses deployed and over 25,000 employees already upskilled in Microsoft Cloud and GitHub technologies, Wipro is embedding agentic AI across workflows in sectors ranging from financial services and retail to manufacturing and healthcare, Microsoft noted.

“Wipro Intelligence… is helping us deliver game-changing outcomes that are reshaping how enterprises work and compete in the AI era,” Wipro CEO and MD Srini Pallia said.

“Our partnership with Microsoft amplifies this vision, accelerating the adoption of agentic AI and unlocking value for our clients and us.”

The post Microsoft Names Cognizant, Infosys, TCS, Wipro as “Frontier Firms” for Copilot Deployment appeared first on Analytics India Magazine.