Good morning, good afternoon, and good evening to all the investors, service providers, and corporates who have joined the call today. This is Rishi Jhunjhunwala on behalf of IIFL Capital. I welcome you all to the second quarter 2026 ISG Index call. I'd like to thank the team at ISG for their valued work on the industry and for asking us to host this call today. ISG has been hosting these index calls on the IT and business services industry for more than 20 years. ISG influences more than $200 billion of technology spending each year, giving them deep insights into the industry, as well as key changes in enterprise demand. With that, I'd like to hand over the call to Stanton Jones, Distinguished Analyst at ISG. Stanton, over to you. Thanks, Rishi, hi, everyone, and welcome to our 95th consecutive ISG Index call. With m e today is Steve Hall, ISG Chief AI Officer, Kathy Rudy, Partner and Chief Data and Analytics Officer, Namratha Dharshan, Chief Business Leader for ISG India, and Alex Bakker, ISG Distinguished Analyst and Head of Primary Research. For those of you that may be on your first ISG Index call, some quick background. The index measures the overall health and growth of the technology industry, which includes both managed technology services and cloud-based software and infrastructure services. We do this by tracking and analyzing annual contract value, or ACV, as a leading indicator of where revenues are likely to go in the future. Just think of ACV as bookings. We combine that bookings data with our primary research data on enterprise buying trends, along with our on-the-ground experience advising G2000 firms on their technology and sourcing strategies to provide you with the insights that you'll see here today. Before we jump into the data, let's start off with an overall level set on the market. Steve, over to you. Great. First of all, thanks, Stanton, what I say, it has been another for the technology industry, not just because of the data we're about to share. If you look back over the last three months, we've seen the largest technology IPO in history with SpaceX, continued unprecedented investment in frontier AI models, hyperscalers committing hundreds of billions of dollars to expand AI infrastructure. At the same time, many traditional managed service providers continue to struggle with weaker valuations, despite generally solid operating performance. Let's take a look at sort of the top five trends. First of all, the combined market is up significantly, but the growth was really market. Enterprise demand is expanding, but a meaningful portion of today's model providers, including companies like OpenAI and Anthropic, which are purchasing enormous amounts of compute capacity to train and serve their foundation models. That concentration is absolutely accelerating the hyperscaler growth. I think the biggest thing that stood out to me this quarter, though, was the new scope activity within managed services. New scope reached an all-time high of $8.2 billion, which was up 14% over the year and represented more than 75% of all managed services bookings this quarter. That's one of the clear signals in the market today. Enterprises are actively reshaping their sourcing portfolios, consolidating providers, expanding scope, and repositioning work to align with their AI strategies, modernization, and cost optimization strategies. At the same time, though, we need to be careful that some of this activity is just reflecting trading work between providers. Operating models are changing. It could be rather than a lot of new activity coming to market, just the shift as we look at that market. The third observation is that managed services continues to show resilience, but that demand is shifting. If you look at traditional labor-based intensive work, ADM, software-heavy engineering, that remains under pressure, especially because of the advances in AI. While business process transformation, industry-specific services, and larger integration engagements continue to strengthen. Fourth big trend is becoming significantly clear that providers are competing in a much broader ecosystem than they were only a few years ago. GCCs, internal AI teams, software platforms, automation are all competing for work that would previously have flowed almost exclusively through traditional outsourcing providers. Namratha is going to talk about this later in the call. That really makes the market more competitive. It creates new opportunities for providers to differentiate themselves through expertise, speed, and measurable business outcomes. Finally, really the economics of technology services are continuing to evol ve. Clients are asking providers to reduce long-term operating costs while accelerating business outcomes. This is leading to deal deflation and is changing commercial models, extending deal durations, and increasing the amount of provider-funded transformations embedded within those contracts. AI is a big part of that story. Equally important is the evolution of the commercial model itself. Let's dig into the numbers. Stanton, over to you. Great. Thanks, Steve. The combined market, which is the managed services segment combined with the as a service segment, reached $42.3 billion of ACV in the second quarter. That was up 43% year-over-year. That's the first time the combined market has crossed $40 billion in a quarter, and it was the highest growth rate we've ever recorded. On the first half of the year, the combined market ACV was $81.3 billion, up 35% from the first half of last year. That's a big acceleration from the 18% growth that we saw at this point in 2025. As Steve, you just mentioned, the components that make up the combined market saw pretty different quarters. As a service continued to drive most of the growth, it reached $31.5 billion in the quarter. In the first half it was up 53%. That's a huge acceleration from the 28% growth that we saw this time last year. Managed services is a different story. ACV was $10.9 billion in the quarter, up 2.7% year-over-year. For the first half, managed services was also up 2.7%, but that's a deceleration from the 3.9% growth in the first half of 2025. In our view, technology demand is very much still expanding, but most of that growth is happening in infrastructure and software, while managed services remains in its traditional low single-digit growth mode. Let's unpack that growth a little bit, specifically within managed services, and we'll start with ITO. For the first half, ITO was down 6%, as you can see here, and that's the first half decline for this segment since 2022. A big trend we've really started flagging last quarter is around new scope activity, as Steve just mentioned. That was up 9% in the quarter and has now grown for 10 straight quarters. As a reminder, new scope here can mean work that's newly outsourced, but it can also mean existing work moving to a new provider, because that's new scope for that provider. In our view, what we're seeing here is consolidation and share movement within the market, Alex is going to dive deeper into that here in a bit. We can continue to see evidence of that consolidation wave in the strength of bundled deals. As you can see here, deals where ADM and infrastructure are combined continue to be strong. That segment is up 15% in the first half. However, standalone ADM continues to be under pressure. It was down 12% in the first half, and that's the 4th consecutive year-over-year decline. It's pretty clear at this point in our view that LLMs have found a strong product market fit in software development, and we think that's starting to impact this segment of the market, as well as in engineering, which we'll talk about next. The net here related to ITO is that the picture's mixed. Traditional ADM is under pressure, but larger, more integrated deals that combine several areas of IT-related scope are still holding up. As I mentioned, let's take a look at engineering next. For the first half, engineering was down 3%. Award activity was strong, however, with 129 awards in the first half, and t hat was up 18% from the first half of the year. Within the segment, network, mechanical, and manufacturing engineering were all up double digits, as you can see here. However, the softness in this segment, which has been so strong for the past couple of years, was primarily due to weakness in software engineering. It was down 27% in the first half. As I mentioned, LLMs have found a strong product market fit in coding. Anywhere where software development is a major part of the work, AI is starting to deflate some of that effort. The key and really important point here is that effort doesn't disappear, it shifts. If you start to think about software development in the context of the software development life cycle, or SDLC, the implementation phase, and that's where a lot of the coding happens, a lot of that gets automated with LLMs. Given the probabilistic nature of these systems, that means that more work needs to be validated by a human. In the case of the SDLC, more work shifts out to, for example, the testing phase. Given the unpredictable nature of token costs and the new architecture patterns LLMs introduce, it also m eans more time on the front end of the SDLC in phases like planning, analysis, and design. The net here is that with generative AI and the software development work, it doesn't disappear, it shifts. That shift is starting to have an impact on the way software is created, how much it costs, how long it takes, and of course, the skills needed to do that work. Really when specifically looking at the engineering sector, that segment, the software-heavy parts of the market are under pressure, while more physical and product and infrastructure linked engineering work continues to hold up better. Okay, let's take a look at our last service line, which is BPO. For the first half, BPO ACV was $4.8 billion. That was up 47% from the first half of last year. That's a very strong rebound from last year. Part of that growth reflects a pretty easy comparison after a pretty weak first half of 2025, and there were also a number of BPO megadeals in the market that also skewed the results a bit. The strength that we saw, though, in the first half in BPO was really concentrated in back office and industry-specific BPO. Back office, which includes areas like F&A, HR, and facilities, was up almost 100% in the first half, and industry specific was up almost 35%. The weak spot here in BPO continues to be customer experience. It was down 16% for the first half, and that's its lowest level since 2020. That's of course another area where automation and AI-enabled platforms are putting pressure on labor-based work. The net here for BPO is that labor-heavy categories remain under pressure. However, as you can see, especially in areas like the industry-specific work, regulated and complex industry verticals where industry context shapes a lot of the work, continues to perform well, and we think that that's going to continue. Before I close out BPO, I think it's also important, I want to make sure and mention that if you want to dig deeper into what we see happening in BPO, we just published and released our 2026 State of BPO Report, the lin ks to that report will be available here in the slides, as well as on the replay of the call on YouTube. Okay, let's move to our regional update. Kathy, over to you. Thanks, Stanton. As I run through the regional results, you'll see a picture of how uneven the managed service market is right now. The Americas is the main source of pressure, with managed services ACV of $5.2 billion in the second quarter, down 12% year-over-year. For the first half, the Americas was down 5.5%. This is the first time we've seen a decline in the first half for the region since 2019. Contract awards, mega deal ACV, and small deal band were all down for the quarter. This points to a softer demand across multiple parts of the Am ericas market. EMEA, however, was up. Managed services ACV was $4.7 billion in the second quarter, up 21% year-over-year, and the region exceeded $4.5 billion for the second quarter in a row. New scope was a big part of that story. EMEA new scope ACV was up 32%, which again points to consolidation, share movement, larger transition activity, as much as pure new outsourcing. Asia Pacific also improved, with managed services ACV above $1 billion for the first time in 6 quarters. Asia is a smaller, more volatile market. Collectively, for the regions, the market is not moving in one direction. The strength in EMEA and improvements in Asia are offsetting real pressure in the Americas. If we now look at the managed services market from an industry perspective, it does tell a similar story. Demand is patchy rather than accelerating evenly. BFSI improved in the second quarter, with ACV up 17% year-over-year, and for the first half, BFSI was up 2%. This is the best first half result for the sector since 2023. As we've discussed on past calls, BFSI is one of the larger indicators of health in the services market. It had declined in the past two quarters. The second quarter improvement is a positive sign, albeit a modest first half gain. The energy sector is a positive story. ACV was up 21% in the quarter and 11% for the first half, reaching the highest first half ACV we've recorded for the sector. Manufacturing moved in the other direction. ACV was down 11% in the quarter and down 8% for the first half. The pressure is tied to weakness in ITO and engineering, which fits with the software and discretionary pressure we just discussed. Industry data reinforces the regional outcomes. Service demand is still present, it is uneven. BFSI and energy are improving while manufacturing remains under pressure. With growth uneven by region, industry, and service line, new scope and bundled work are areas of activity. This takes us to the next question. When clients bring work to providers, what kind of deals are they signing? Now we're going to turn to Alex, and he's going to talk in more detail on how the economics of deals are changing. Thanks, Kathy. Earlier this year, we did a study that examined how organizations expected to change their provider landscapes. For the second year in a row, they indicated that they plan to consolidate scope with their largest providers and expand their use of smaller providers for niche and innovation use cases. They're going to reduce those providers that are kind of in the mid-spend band in their organizations. This results in what we call a dumbbell-shape market, where the providers that have been squeezed out of the middle are pushed to either differentiate in more narrow areas or be consolidated out into larger players where there's more cost advantage. The data on this chart specifically looks at their motivations for both consolidation of work with large providers, namely to drive cost savings, and their motivation for expanding the use of the small providers, which is really to focus on speed, agility, and specialization. Major evidence in the market from the blowout new scope number this quarter suggests that really there is deal scope changes and consolidation moving to big providers, as often new scope indicates a change of provider in addition to the work moving around. Finally, the durations we're seeing in the market are also a good indicator of consolidation, because as Stanton mentioned earlier, durations that are long and extending longer indicate that deals are often being financially engineered to support savings over a longer period of time. Now let's look at the deal shapes that we see emerging in the market. We've really started to see two emerging deal shapes. First one, total cost of ownership, and the second one is what we're calling internal rate of return deals. Following the long durations and the consolidation with providers, we see TCO deals really being built around a consolidated scope, long durations, and provider commitment to savings over the terms. The long-term objective here is to reduce the run rate for those scopes of work, those deals are often structured to pull forward savings to help clients redeploy their budgets early on. The internal rate of return deals are very focused right now on the software development life cycle, and they're motivated by accelerating outcomes and faster time to value. For example, taking a six-month deal and doing it in three months, often with the help of AI. We're seeing evidence that providers that are selling these are moving at a higher velocity, they're selling lots of smaller sequential projects or deals that bring that value more quickly. Between these two deal shapes, we're seeing how enterprises are contracting to balance both their need to save money and accelerate their outcomes. These deal shapes and the changes to the provider portfolios are being driven by an underlying enterprise strategic mandate for IT. Again, earlier this year, we examined how enterprises were prioritizing their core technology objectives, we looked at how they were performing, what impact IT was expected to have on achieving those outcomes for 2026. The three items that we've highlighted here that really stand out are technology modernization, cost reduction, and productivity improvement. These items are both important, they tend to be underperforming, and they are the three areas where IT is expected to have the biggest impact. This creates pressure on providers as it's driving those outcomes that create the deal shapes and the provider portfolio motivations that we've seen so far. While the TCO deals drive cost reductions, they also drive in-year redistribution of IT spend since the savings were pulled forward. These pulled forward savings create the new project and investment opportunities for clients to drive productivity, creating the demand for those short-term IRR deals. What we've reliably seen is that organizations are not running out of technology backlog. We think the two models that we are seeing in the market, augmented by AI, have the potential to create their own new demand as organizations continue to pursue their modernization efforts. Let me hand it over to Namratha to discuss how competitive pressures from GCCs are also impacting the managed services market. Thanks, Alex. In the last couple of years, we have seen exponential growth of GCCs. As you can see here, we have significant uptick in headcount growth amongst GCCs while the growth in the services industry continues to remain flat. The new wave of GCCs is no longer focused on keeping the lights on, but they're working towards a broader charter of becoming a critical delivery center for the enterprise, which cuts across multiple services, whether it's ITO or BPO. For example, one of the areas is BPO. For business process functions such as finance and accounting, supply chain, procurement, front office customer service activities, enterprises have traditionally leveraged BPO providers to manage them. However, in our State of BPO Report that we recently released, we noted that the three highlighted business functions are increasingly being managed at GCCs now. That overlaps with BPO services work. This is likely to expand because according to our research, 90% of the firms with recent or planned GCCs expect to increase the scope of work handled by them in the coming year. In specific areas, though, GCCs continue to rely on service providers like setting up the facility, talent acquisition, and other project-based work like COE setup, solution implementations, and learning and development. That doesn't give us the whole demand picture. As importance of AI continues to grow, GCCs will take central role in shaping enterprise AI and transformation strategy, which will add more pressure on the services market. When we talk about muted services growth, while some work is landing with pure-play BPO providers, we see work is also getting distributed across a broader provider ecosystem. As you can see here, BPO providers face competition from every side of the delivery ecosystem. Consulting and audit firms bring advisory depth and industry context. Vertical specialists bring industry-specific expertise, which is a growing area. IT providers and AI-first BPO firms bring host of AI capabilities, domain expertise, and delivery accelerators. Specialist data and AI firms bring analytics and AI capability, often with a much more targeted scope. Given that today enterprises can choose between external delivery, developing internal capacity, and bring their own AI-enabled internal work, the implications for service providers is very straightforward. Strong demonstration of deep domain expertise, governance, flexible commercial models, and most importantly, focus on outcomes. Despite the headwinds, our leaderboard results reflect that providers are also gaining momentum in multiple areas. On that note, moving to the leaderboard recognitions. As a reminder, providers are listed in alphabetical order, and positioning is based on annual contract value signed over the past 12 months. The companies new to the list are denoted with an asterisk, and a reminder that regional leaderboards can be accessed on the ISG website. In the largest group, the leaderboard was mostly stable, with top companies maintaining their positions. This quarter, we would highlight Infosys. During the quarter, Infosys worked with BP on AI agent initiatives across trading, supply chain, sustainability, and core operations. It also expanded its relationship with DNB Bank in Norway to help modernize fragmented legacy systems into a more unified cloud-native platform for fincrime technology. In the Building 15 group, EPAM is working with Hugo Boss on the XP membership program using digital transformation and blockchain technology to support customer loyalty experience. In Breakthrough 15 group, Coforge is working with VHC Health as a transformation partner across infrastructure, cloud, digital workspace Cybersecurity services. TTEC won a seven-year extension with Volkswagen Group U.K. for customer care, case management, connected vehicle support, and AI-enabled omni-channel contact center operations. In the Booming 15 group, we continue to see turnover. Zensar and GFT Technologies joined the leaderboard, along with BNP Paribas Real Estate here in FM and corporate real estate. GFT signed a long-term strategic transformation contract with Deutsche Bank, where it will support a hybrid cloud program with Google Cloud. On that note, congratulations to all the providers featured in the Managed Services leaderboard. Let's turn to how AI is impacting the sector. Steve, over to you. Great. Thanks a lot, Namratha. I think you and Alex did a great job setting up the story on how AI is really impacting the market right now. This is the second quarter now that we've done our ISG AI Index. This is a quarterly assessment of really how AI is changing the technology market, as you just heard. We've anchored the index to the launch of ChatGPT in December 2022 and track AI's impact across the same three markets we cover in the broader index: infrastructure as a service, software as a service, and managed services. It then brings the three complementary perspectives. We look at AI through an economic lens, which is revenue, profitability, market performance, and several forward-looking indicators. Then we've added an AI confidence factor, which measures how management is thinking about evolving as AI moves from innovation to enterprise adoption. Finally, we combine those perspectives into an AI market assessment that gives us a consistent view of where we think AI is creating value today and how that value is changing over time. With that, let's take a look at the data. The overall Index now stands at 189. This is up nearly 90% since the AI inflection point in December, up 4.6% quarter-over-quarter, and up 25% year-over-year. You're seeing this constant acceleration with it. If you go look at the individual segments, you see that infrastructure continues to be the biggest beneficiary of AI. The IaaS Index is now up 182% since the AI inflection point. The hyperscalers continue to invest in extraordinary levels, and demand for AI infrastructure remains exceptionally strong. A meaningful portion of that demand, though, as I mentioned at the beginning, is still being driven by a relatively small number of frontier model providers. Companies like OpenAI and Anthropic are investing enormous amounts in compute capacity, as we discussed, and that's really accelerating the hyperscaler growth today. One of the things we'll be watching is how quickly that demand broadens as enterprise adoption continues to scale. Software has moved into a different phase. Revenue remains strong with CRPO continues to grow, AI is now becoming part of almost every enterprise application. The discussion has really shifted from capability to commercialization. Investors are really asking different questions now, not whether AI works, but whether software companies can translate AI into sustainable revenue growth without fundamentally changing the economics. Managed service is where the market's becoming much more interesting. Our Index is down 4.7% since the AI inflection point, at first glance, you might conclude that AI is hurting services, but I don't think that's what the data says. If you look at it, revenue continues to grow. It's up 7% since the asce nsion date. Profitability was up in the first quarter but declined this quarter as organizations continued to invest in AI platforms. Revenue per employee continued to improve, which was up 8% this quarter, which shows that the productivity is coming through. Yet the valuations with stock prices for the sector is down almost 50% and continues to come under significant pressure. In many ways, the market is looking forward. Investors aren't reacting to today's operating performance. They're really trying to understand what AI means for tomorrow's business model. This quarter, we're also introducing another enhancement to the ISG AI Index, and this is our new AI confidence factor. This factor assesses management's overall confidence on how AI is evolving as it moves from experimentation to enterprise adoption. To do that, we've analyzed AI-related commentary from quarterly earning calls from across 25 of the largest infrastructure, software, and service providers. We're not simply measuring where executives are opt imistic or pessimistic about AI. We're looking at how that conversation itself is changing, where companies are becoming more confident and where concerns are emerging, and how the discussion shifts as the AI becomes a larger part of their business. One of the things we wanted to understand was not simply how much companies were talking about AI, but how their thinking about AI was changing as the market matured. If you start with the chart at the top, the trend really stands out. Since the launch of ChatGPT, AI has moved from being an emerging topic to a permanent management agenda item. AI mentions across the provider universe have increased more than fivefold, yet management confidence has remained remarkably resilient throughout that period. Despite one of the fastest technology shifts we've ever seen, confidence remained consistently high. We did see some moderation earlier this year, particularly in managed services, but confidence recovered in the second quarter as the conversation became more focused on execution and business outcomes. If you look at the bottom chart, though, this is really interesting. A year ago, most of the discussion centered on governance, risk, pricing, and improving return on investment. In fact, governance and risk-related commentary increased by 400 basis points over the past year, rising from 7% to the 11% of all AI-managed commentary. Quite honestly, this is exactly what we'd expect in a maturing market. Discussion has really shifted from what can AI do, to how do we deploy it responsibly and create measurable business value. That leads us to how do we look at value? When we put the AI Index together with the confidence factor, you can see every part of the technology industry moving through a different stage of the AI value chain. Infrastructure is generating extraordinary returns on investments. The economics are already visible in revenue growth, capital investment, and market performance. Software's moved into the commercialization stage. AI capabilities are now mainstream, but the conversation has shifted towards monetization, pricing, and long-term economics. The market's no longer rewarding AI features alone. It wants proof that providers can convert AI innovation into sustainable revenue and profitability. Managed services, though, is facing a different challenge altogether. AI is no longer being judged on whether it can automate work. The question is whether providers can reinvent their delivery models while maintaining attractive economics. That's exactly why we see the large disconnect between improving operating metrics and weaker market valuations. What's interesting is that the confidence factor mirrors that journey. Confidence remains high across all three sectors, but the conversation is becoming much more commercial. Management teams are spending less time talking about the AI opportunity and much more time talking about the execution, return on investment, and business outcomes. To me, that's really the most important insight from this quarter's AI Index. The market has moved beyond asking whether AI creates value. The question now is really much more practical. Who can prove that value, and who can capture it? With that, now let's move to the summary and our forecast. As we wrap up, let me leave you with a few observations for the quarter. Starting with managed services, the market continues to demonstrate resilience. First half ACV was up 2.7%, so the demand is still there, but the shape of that demand continues to evolve. The biggest signal for me still remains the new scope. We saw another record quarter, which tells us clients are still driving sourcing decisions, resetting portfolios, and moving work to support AI modernization and cost optimization. The market's very active, but much of that activity reflects portfolio realignment rather than broad-based expansion. The picture underneath that, though, remains mixed. BPO continued to recover while ITO recorded its first half decline since 2022, and engineering moderated after an exceptionally strong 2025. That's consistent with what we've discussed throughout the call. AI is reshaping where enterprises invest and increasingly how they structure that investment. Regionally, we continue to see different stories. EMEA had a very strong first half, while the Americas remained more cautious. That tells us demand is still healthy, but enterprises are becoming increasingly selective about where and how they spend. On the as-a-service side, the momentum continues. First half ACV increased 53%, while both IaaS and SaaS reached record levels. AI infrastructure demand remained the primary growth engine, supporting hyperscale backlog and capital investment. Software also continued to evolve. We're seeing the market move towards more agentic architectures and consumption-based commercial models, that technology is maturing and the business models are evolving right alongside it. Looking at the broader market, the outlook really hasn't changed. Interest rates and geopolitics will continue to create periods of uncertainty, but the technology demand remains very strong. For managed services, we expect growth to remain modest but positive. New scope continues to tell us that work is being structured, even if the overall market growth remains constrained. At the same time, AI-led productivity improvements, together with GCCs, software platforms, and new delivery models, will continue to put pressure on traditional labor-based services. Providers that adapt their commercial models and demonstrate measurable business outcomes will be best positioned to benefit. That leads us to our forecast. We're maintaining our managed services growth forecast at 2.1% for the full year. We think the market is still resilient. It's becoming competitive and more selective, but there's still a lot of activity there. We're increasing our as-a-service forecast 500 basis points to 30%, reflecting the continued strength we're seeing in AI infrastructure and enterprise software demand. There's one thought I'd leave you, though, with this. The market really isn't standing still. It's just being reallocated. Organizations that understand where value is moving and adapt their business models accordingly will be the ones that lead the next phase of the AI economy. With that, thank you for joining us today. We look forward to speaking to you next quarter. That brings us, then, to the call. We'll now open with a Q&A. Please type your questions at the bottom of the screen, then Rishi, over to you to lead us through the Q&A. Fantastic. Thank you so much, Steve. Maybe a few questions right on the deck as well. It was quite surprising to see the kind of growth we saw in as-a-service market. A two-part question to that, right? Firstly, do you expect some of this strength to flow down for managed services providers as well, if not this year, then next year? Also, in your best estimate, what do you think is the percentage of revenues that these companies have already seen from a deflation perspective? Steve, do you want to take that? Yeah, I'll take that one, Rishi. First of all, I do think we'll see a flow down of the growth from the hyperscalers. We're already seeing that in many ways. Again, as we stated, the growth right now is really driven by just massive spend by the model providers. What we see going on now is really rethinking data centers, rethinking their AI strategy, and a lot of front-end work to really drive better tokenomics. When you think about that work, we're seeing service providers create platforms, harnesses, new ways to route prompts to the right engine, new solutions coming through. I think the more you see on the SaaS side, you'll see that flow down more through the service provider side as well. Quite frankly, you still have all the management issues and operating issues of managing these large cloud environments, which will continue to expand as we go forward. I guess the last thought on that, you still have the challenges or the opportunities with sovereign clouds, the whole tech sovereignty issue across other parts of the world. I think that will continue to drive investments as well as we see the hyperscalers grow. Yeah, I'll add on to that, Rishi. I think it's also just important to note, I think we said this on the last call, in some ways, this kind of feels like cloud maybe 15 years ago. There is a completely new architecture pattern emerging in the market, as Steve just talked about. We weren't using the word agents and agent harnesses two years ago. If you think about what's happening in I definitely see this happening in BFSI. We were talking earlier in the call around capital markets. There's some big underlying architectural changes happening right now that frankly happened during cloud, which was, "Where do we run this workload? Do we run it on-prem? Do we run it in the public cloud? Do we run it in both? How do we keep track of it? How much does it cost? How do we secure it?" That same pattern is happening right now. I still think we're very early into that. That kind of pattern, even though the technology, for example, the frontier models, is progressing really fast. The whole system architecture and governance around that is way behind. Maybe not behind, that's probably it hasn't been really developed yet. That to me means systems integration and type work that as providers get their sea legs around how this all comes together and start to develop those patterns specific to an industry, that is going to represent a lot of new opportunity. But as we talked about, there's also that AI, the AI itself, those agentic patterns also compressing the work itself. There's multiple things changing at once right now. Great. Another follow-up on that. You talked about decision making in terms of what to spend, where to spend. Probably link question would be GCC growth. Is there a question with the enterprise whether they need to keep it within their own GCC versus try and outsource? We have seen tremendous growth in GCC for the past few years. Some questions or debates have started to come up suggesting growth there also can potentially slow down. You discussed some of it in the presentation, but what are your thoughts around that? Namratha, you want to take that? Yeah, I think, Rishi, at this stage, for the last couple of years, like I mentioned, we've seen an exponential growth. That seems to be, compared to all the previous waves, I don't think we have seen this kind of an unprecedented growth, which seems to be continuing, because practically every week, we are adding at least two to three GCCs here. The focus, I think, is a large part of it is more on the how do we control our IP, how do we control our data, and that focus I think is continuing and very strongly continuing because we spoke about even the hiring. One of the key things is that the skill sets that they are hiring is also focused on product engineering, AI, and data specialist skills. That is going to be the focus is they really have the controllership is something they just want to be very close to the GCCs, and that is what is driving the growth of the GCCs. More than the IT services engagements are really on either setting up the GCCs or just to kind of help them with some of the staff up projects or some of them on to learning and development. That trend seems to be growing strongly, and we hear that even from the provider community, as well as even from the GCCs from the standpoint of how they're taking it forward. Rishi, I don't know if you saw this or not, but Namratha just mentioned two or three announcements every day, it seems like. If you saw yesterday, Nestlé and Genpact announced new GCC in Hyderabad. We actually just finished a number of primary research interviews with GCC leaders, both at the GCC and corporate leaders. I would say to me, the biggest change that's happened over the past year is a sense that not for all companies, but for companies that are pretty mature on this path, less about thinking of it as a GCC. In fact, many of them don't call it a GCC inside of their own company. It's just company name India, company name Poland, company name Philippines, wherever it is. It's an extension of HQ, work moves where the talent is and where the price scales. That to me is a big change from where we were a couple of years ago when it was like, "Well, we've got to have a GCC." I'm not saying companies aren't still doing that. Of course, there's still a lot of FOMO out in the market, but there definitely has been a change over the past year towards now, this movement towards moving the work to where the talent is and where the price will scale. Whether you want to call it a GCC or not, it doesn't really matter. That's all levels of types of work being moved, especially in areas like data and analytics, product engineering, et cetera. That to me is the biggest shift that's happened over the past year. Understood. The one thing that has been debated a lot over the past few months in this part of the world is some of the mega M&A that has been announced by Indian IT companies, which is unique because historically, they've not been known to do such large M&As. There's a question mark whether is that coming because growth is becoming a challenge? The debate also is whether should they be investing their ca pital to scale, or should they be looking at investing it in new gen technologies, including AI? What are your thoughts on some of these? I know you had written about Persistent, Nagarro as well. Would love to understand what are your views from that side of the world. Sure. Alex, do you want to take that? Sure. I think this is an extension of the same trend that is pushing clients to consolidate work with fewer providers, is that being one of the fewer providers requires being able to consolidate more work. The acquisitions, I think, are largely to add capacity outside of your existing core, so that you're eligible to be able to roll up larger scopes of work on your client's behalf. Clients are very clear that they would like providers not to be building out undifferentiated capacity at kind of a mid-scale. They are looking for specialization and the ability to get the whole thing to drive cost leverage across a larger scope of work. I'd say the M&A is largely in service of that. Got it. The reason I found that interesting is because in the presentation, you discussed about how the large deals are being distributed among the larger players for cost transformation, but a lot of the smaller projects are going to smaller vendors who are far more nimble, right? Is scale really that important going forward given the way in which the work is getting distributed now? I think it is, at least at the real cost transformation, because it's not just a total cost of ownership reduction over the term. It is re-engineering and modernizing the work and the systems that are doing the work. The reason that you need to scale it up is because you don't want to have to try to manage the responsibility of one system with one provider and running and editing that system with another provider, and then the services of using that system with another provider. They're trying to consolidate it to reduce that kind of authority versus responsibility mismatch that happens when you have too many providers in your ecosystem to give a provider the ability to run and change and modernize that system over the term. Got it. I think Rishi, I'm sorry, go ahead. Go ahead, Rishi. No, sorry. If you're completing the answer, go ahead. Otherwise, I'll wrap up. Yeah, I was just going to say, I think obviously the Persistent, Nagarro, Capgemini, WNS we talked about last time, the big ones are super interesting. You're right. It doesn't happen very often. In fact, that's why I led off when I wrote about that. That doesn't happen every day, you see a billion-dollar acquisition in services. I 100% agree with Alex on that. I think also looking at some of the smaller ones, for example, Cognizant and Astreya or EXL and iMerit, a lot of that focus or what Accenture did recently, a lot of that all focused on what are the two areas that are growing fastest in the IT budget? It's AI and cyber. That's where that acquisition activity is happening in order to capture where that spending is happening. You mentioned nimble, getting access to faster ability to execute faster for more of those IRR deals. I think the large deal activity will be here and there, I think it's the smaller cyber and AI work we're gonna just continue to see a lot of roll-up happening in that space. That's where the dollars are. Got it. Fantastic. Yeah. You'll continue to see that, Rishi, in engineering as well, which was a lot of the Persistent strategy as they look through that. Engineering was down, I think we said 3% this quarter, but that's off really record highs. Engineering spend has just really accelerated if you look at the last three-year pattern. That's going to continue, even with AI as a big part of the market. Again, getting capabilities and specialties, capacity, as Alex mentioned, are all important to that transition. Understood. Fantastic. The market will bear it, too, because there are more deals of larger size in the market now than there were 10 years ago by a pretty large factor. Got it. Thank you. Thank you for the opportunity. Let's open the floor for other people to ask questions. Fantastic. Thank you very much, Rishi. Great to see you. Let's hit some of the questions. Kathy, I'm going to come to you first. We've got a number of questions coming in around the impact, just rolling these up. Basically, AI impact on pricing, specifically in managed services, not software or infrastructure. Sure. If you think about the impact of AI on managed services pricing. What we're really seeing happening is deal acceleration in the out years being faster than ever. If you think about it, we always had deals three to five years. You would expect a price reduction in each one of those years. What we're seeing now is larger price reductions year on year than we've ever seen. We did a recent study. I'll just use one example. In procurement, we would normally see a reduction of about 25%. We're now seeing it closer to 40%. We're looking at that and saying, "Oh, can we attribute all of this to AI?" Maybe, but there could be other variables at play. We know that clients are buying based on the fact that they think that they're going to get really bigger reductions because of AI. Service providers are delivering on that. It remains to be seen how and when those reductions will be made. How they will be delivered. AI is definitely a big play there. Yep. Okay, we've got a super interesting question. This is a red-hot topic right now. Steve, I'll come to you first, and I know we probably all have a point of view on this one, but I'll come to you first. Basically, sharing of AI savings, what do we see happening there? How is that working? Yeah, that's the amalgam of the question. Yeah. This is sort of the question du jour for quite a while now on who gets the economic value or who shares in the economic value created by AI. I think, in general, what we've seen on the managed services side is there has been a tremendous amount of conversation about moving to outcome-based models or output-based models. A lot of organizations are really working to understand, both in the BPO space, industry-specific apps, IT, et cetera, what outcomes really mean and how do they share. What we're seeing in deals is productivity, though, really aligned with what we call autonomy levels. We've got what we call autonomy-level pricing that really aligns the level of autonomy within an organization. Whether it's regulatory, whether it's cultural, whatever those changes say, how much do they want people involved in those decisions versus agents being aligned with that. When you look at a lot of provider pricing right now, you're going to see outcome-based tied to some sort of an autonomy level. It's almost what that last part is saying is how do you step down with AI in a very managed perspective. I think the one thing that's really important for the market right now is it's not an agent replacing a human, which is sort of the simple math that we do sometimes. It's one of the reasons we really don't like digital labor or agent labor or other things. It's offset with everything with platforms, harnesses, training, inference costs, tokenomics, all of the other aspects that have to be added up to create this sort of total cost of ownership when you look at deals. That complicates the message, but in many ways, it also can make it easier to say, "Okay, I want to operate at this level of autonomy for this business process, and this is the output or the outcome that I want, and here's how I'm going to achieve that." Those types of deals are becoming more and more pushed throughout the market. Okay. Alex, I've got a question for you. It's basically around AI use cases, or what's getting in the way of AI use cases getting into production. Yeah. Thanks, Stan. I think there's two main factors here that we've been seeing a lot and talking about a lot. The first is kind of one I'm calling the validation bottleneck, which is really when organizations deploy AI to try to solve a process, automate a process, there's a need to validate it. We've called this the human in the loop part, and to make sure that the AI is producing an acceptable and accurate result. I think as many people, many of you probably have experienced, that takes more work than you thought it would. What happens is you try to automate use cases, especially where there's talent constraints or where expertise is a bottleneck in your organization. The validation work piles on top to the existing experts and slows work down even further. It can be very difficult to manage where automation creates more backlog for validation in QA, versus where automation can bypass a kind of human bottleneck to create more capacity. I think that is partly informing how organizations are selecting use cases now and thinking about where the capacity you're creating is not going to pile on top of an existing scarce resource. The other half of it, though, and I think we've seen more discussion in the news recently on this, is that the token model is also kind of problematic for getting things into production. The thing I'd point out about it a lot is that there's no time component to a token, but there is a time component to savings and the kind of ROI metrics. Every measure of financial return is predicated on the amount of time it takes for the return to happen. You don't really know how much time a token saves until you actually deploy the whole use case and run it and empirically measure what it delivers back to you. That can be a bigger hurdle for organizations because you try the pilots, you get things into production, only then do you find out if it's really worth it. Not every use case out there actually benefits much from acceleration. The example I give about this is, it may be possible to cook dinner by 8:00 A.M., but no one benefits from that. You have to kind of build your use cases, continuing that metaphor, to get your meal prep done by 8:00 A.M., but still have to cook the dinner at dinner time. I think there are problems in the token model that make it difficult to predict the time value that you're going to get back. You know what, Alex, there's one element that I think you could add to all of that. You can cook your dinner by 8:00 A.M., but you need to have all the ingredients. I think a lot of what is causing the deceleration is the access to data and clean data. It really is tanking a ton of use cases when all of a sudden you get into it and you find out you don't have the ingredients, you don't have the data, or it's not the right ingredients. I think that's another thing that we should be thinking about. Absolutely. Nice analogy. I like that. Okay. I think we have time for one more question. Steve, I'm going to ask you to close it out. We've got a question around kind of we talked about at the beginning, the difference, we're tracking bookings here, not revenue, but obviously we're tracking revenue as well and sort of the potential disconnect there. Revenue staying relatively resilient, but bookings have been declining. Is that a timing issue or is that a slowdown issue? Yeah. We've looked at this quite a bit, Stanton, as you and I have talked, and others. I think the bookings have moderated a little bit as we talked through, but I wouldn't describe that really as the market weakening or anything. I think one of the key findings, again, going back to what we reported out, is really the new scope and the new scope increasing. Market was down a little bit, in ITO, up in BPO. We saw some good signs there, but that new services piece being up so high, 14%, was really, really strong. That just tells us that there's still a lot of activity there. I think you'll see the bookings catch up with revenue. I think on the profitability side, we still are in this investment phase where we've got to build platforms, harnesses, other things, as the business models change. I think we'll continue to see bookings continue to evolve. Right now, it's just a bit of timing on the revenue side. Okay. Thanks, Steve. Okay, we are just about at time, so we're going to go ahead and close out the call. Rishi, as always, a huge thanks to you and your team for hosting the call today. As a reminder, you can access a copy of the slides that we just showed you, and that will include the regional leaderboards, which you didn't get a chance to see today, as well as I mentioned, a link to the State of BPO Report. That'll be available as a link in the report as well. Thanks again for spending an hour with us, and we will see you on the Q3 index call on October 8th. Thanks.