Leadership in AI innovation | Satya Nadella & Nandan Nilekani | Microsoft AI Tour

Summary

Satya Nadella and Nandan Nilekani discuss the rapid advancements and implications of AI technology at the Microsoft AI Tour. They explore how Moore's law, which has historically driven the tech industry's pace, is complemented by breakthroughs in AI, making innovations even faster. In particular, they stress the importance of applying AI technology to real-world challenges, mentioning India's unique position to lead in AI use cases due to its existing digital infrastructure, political support, and tech adoption culture. They emphasize the need for an equilibrium between innovation and safeguards and consider AI as a tool that amplifies human potential rather than replacing it.

Highlights

• AI advancements are outpacing Moore's Law, accelerating the pace of tech innovation 🚀.
• India is well-positioned to lead AI applications globally due to its established digital frameworks 🇮🇳.
• Responsible AI requires balancing innovation with safeguards to ensure ethical development ⚖️.
• AI and humans can collaborate to enhance productivity, akin to a jazz session 🎷.
• Convergence of AI technology with business processes presents a huge opportunity for innovation and efficiency 💼.

Key Takeaways

• Moore's Law and AI together are accelerating tech developments faster than before 🚀.
• India could become the global leader in AI applications due to its robust digital infrastructure 🏗️.
• AI is a tool for enhancing human capabilities, not replacing them 🤖❤️.
• Change management is crucial in adapting to new AI technologies in businesses ⚙️.
• Enterprise AI will require focusing on entitlements, memory, and tools use for efficiency 🔧.

Overview

In this engaging discussion, Satya Nadella and Nandan Nilekani dive into the rapid evolution of AI and how it has surpassed the pace set by Moore's Law. They highlight the unique position of India in the AI landscape, thanks to its infrastructure, leadership, and societal readiness to adopt technology. The speakers point out that while innovation is crucial, it should be balanced with responsible practices to prevent possible pitfalls.

A major theme of the conversation is the synergy between AI and human capabilities. Instead of viewing AI as a replacement, both leaders advocate for it being a tool that amplifies human potential and facilitates new ways of working. This mindset nurtures an environment of collaboration where AI and humans work together, enhancing process efficiencies across various sectors.

They also discuss the changing nature of enterprise AI, emphasizing the importance of managing entitlements, scalability, and security. Companies are urged to embrace new AI workflows and technological integration to achieve better business outcomes. As India places itself at the forefront of AI applications, this discussion marks a pivotal moment in understanding how to leverage AI for societal good and business efficiency.

Chapters

• 00:00 - 02:00: Introduction and Moore's Law The chapter begins with a conversation between the speaker and Satya, who appear to be at an event or conference. The speaker expresses their happiness to be back in Bangalore and reflects on how their interaction has become an annual tradition over the past few years. They discuss the rapid rate of change experienced over the last two years, alluding to technological and societal shifts akin to the principles of Moore's Law, which predicts the doubling of transistors on a microchip approximately every two years, leading to exponential growth in computing power. The mood is positive, and the setting suggests a gathering of thought leaders or technologists.
• 02:01 - 05:00: AI Revolution and Indian Context The chapter discusses the rapid pace of technological advancements and how it impacts individuals and industries, particularly in the context of the AI revolution in India. It highlights the concept of Moore's Law as a driving force behind these advancements, noting its significance as an empirical observation over the past several decades. The conversation explores the challenges and anticipations associated with constant change and innovation in the tech landscape.
• 05:01 - 09:00: Change Management in the Age of AI The concept of Moore's Law is explored, not as a formal law, but as an empirical observation that has been consistently validated over time. The chapter recalls an annual tradition where Microsoft executives were shown a chart depicting Moore's Law. The message was clear: the exponential improvement in memory should be leveraged by filling it with software. This anecdote underscores the ongoing challenge and opportunity in the field of technology to utilize ever-increasing hardware capabilities effectively through strategic software development.
• 09:01 - 18:00: Enterprise AI vs Consumer AI The chapter explores the transformative impact of AI technology on the pace of computational advancements. It contrasts the previous era defined by Moore's Law, where computing power doubled approximately every 18 months, with the current accelerated growth driven by AI innovations. Key advancements such as Deep Neural Networks (DNNs), Graphical Processing Units (GPUs), and specialized AI accelerators contribute to this rapid evolution, reducing the doubling time to six months. The chapter highlights the significant acceleration in technological development as a result of these advancements in AI.
• 18:01 - 23:00: Call to Action for CEOs The chapter 'Call to Action for CEOs' discusses the accelerated pace of change in the business environment and the technological advancements impacting organizations. The speaker reflects on the rapid evolution over six months and hints at future shifts, highlighting a reluctance to predict or label the change but instead focus on adapting and riding the wave of innovation. The chapter suggests that the coming years will bring significant transformations, urging CEOs to prepare for the continuous evolution in their industries.
• 23:01 - 28:00: AI Fact-Checking and Knowledge Work The chapter titled 'AI Fact-Checking and Knowledge Work' emphasizes the importance of utilizing technological advancements like Moors law, new Moors law, LLMs, and models. The discussion reflects on Bill's statement about filling technology with software, which suggests that the key lies in how people employ these technologies rather than just admiring them. In essence, the chapter advocates for using abundant technological commodities, such as AI and software tools, to effectively enhance productivity and knowledge work.
• 28:01 - 30:00: Accuracy in AI Applications The chapter explores the topic of accuracy in AI applications. The discussion draws a parallel to the advent of spreadsheets, suggesting that just as people quickly adapted to and utilized spreadsheets for managing data, a similar adaptation process will occur with AI. The conversation also hints at a more localized application, asking how this change can be contextualized for different regions, such as India.
• 30:01 - 31:00: Conclusion The chapter discusses the potential impact of AI on the Indian economy and society. It suggests that India could become the use case capital of AI globally due to its extensive experience in building population-scale digital infrastructure efficiently. Additionally, the chapter highlights the advantage of having a tech-savvy political leadership in India.

Leadership in AI innovation | Satya Nadella & Nandan Nilekani | Microsoft AI Tour Transcription

• 00:00 - 00:30 [Music] it's great to be here with you Satya thank you so much we did this a couple of years back that is right and this has become our annual thing and I'm really glad to be back in Bangalore with you and uh it's wonderful you know one thing I was thinking you know the last two years there's been an exponential rate of change and you know we're all every
• 00:30 - 01:00 day we get up and something new is happening how do you deal with that and how do you deal with what do you think is coming now yeah I mean that's a I think the this entire industry in fact backstage we were talking about this um the last whatever 35 years that I've been in this industry um the thing that has powered us is this Moore's law right i mean it's just been the most unbelievable um empirical observation right we call
• 01:00 - 01:30 it a law but it's really uh an empirical observation that is held true for a long period of time um and I remember distinctly like we used to have these um um you know he builds to get the top Microsoft folks every year and he would do only one thing he would just put a chart he would just show us Moors law uh he would say here's what's happening with memory uh go fill it with software that was sort of the instruction right that was basically what we've been doing
• 01:30 - 02:00 all these decades and then comes this AI revolution and it fundamentally accelerates because we were all bemooning the fact that oh my god Moors law is maybe ending or it's not like the same as before uh and then here is one algorithmic breakthrough DNN's here is sort of the graph GPUs and AI accelerators now and you put those things together what was an 18month doubling has become a six-month doubling wow right so you can think think about
• 02:00 - 02:30 it right which is we were already on a fast pace and now we've done it six month and now you have another thing this inference time comput's law this well I I think I'll avoid u you know the you know naming anything except I just want to ride the wave um and so that is sort of the change but the question you asked is the interesting one right which is I think a little bit of where what's going to be the phase shift going forward is I don't think we're going be sitting here next year the year after
• 02:30 - 03:00 admiring either quite frankly uh the Moors law or the new Moors law or even the LLMs or the models it will be what we're doing with all that abundance because the interesting thing is that's kind of go I go back to that bill's statement right which is fill it with software and so it goes back to what do people in this room and everywhere else do with it right if there's going to be an abundant commodity you don't sit there and pray to the abundant commodity you use it it's kind of like when Excel
• 03:00 - 03:30 came out we didn't say "Oh my god here is an Excel spreadsheet let's sort of put a Morty of it and play." We just said "Let's create spreadsheets." That's right and I think that that's what's going to happen uh with and that's I think how we as humans will understand to manage the change sure and so one of the things that brings for me Nandan is you know you you and I just talked briefly backstage even you take all this contextualize it for me in India like
• 03:30 - 04:00 how how does this apply to the Indian economy and Indian society broadly uh to totally I think India will be the use case capital of AI in the world uh I think we have a number of things working for us one is uh I think we have 15 years of experience in building population scale digital infrastructure which we know how to make it work cheap you know high volume billions of transactions all that stuff so we know that game we have a political leadership which is very tech-savvy i know you met
• 04:00 - 04:30 from Prime Minister Modi yesterday and they understand that we have to strike the right balance between AI innovation and safeguards because in some parts of the world they're saying safeguards first without worrying about the innovation so I think we know the right balance between responsible AI and u and and and and innovation and we have a population which has learned to accept technology i mean you know when you think about it UPI was launched about pretty unbelievable seven years back now 400 million
• 04:30 - 05:00 users 16 billion transactions a month i mean unbelievable this can happen right so I think AI is at that spot we have to make it work and we already seeing early uh you know signs for example if you look at Aadhaar authentication it's all AI based because we had to do livveness detection for biometrics so people don't spoof it it's all AI based if you look at our tax systems back end is all AI you know figuring out who's doing fraud and all that's why tax revenues are going up so I think India is now ready
• 05:00 - 05:30 for that and we are going to see many many applications coming so I think this is the place for you to see this stuff working at population scale no 100% i mean we that's why we are very very excited about you know our investments here we see the I mean like the combination you said it's a pretty unique place where you have this virtual cycle between the entrepreneurial energy the government sort of yojenas the India stack and then the population
• 05:30 - 06:00 scale so you add those four things into a virtuous cycle like the UPI or Aadhaar or now what's going to happen in education and in health and what have you uh I don't think there is a place uh where the atcale benefit so in an interesting way You know in the past people talked about this convergence right between developing countries and developed countries in an interesting way there's abundance of uh tokens can probably reduce that convergence oh I I think in many leaprog I think in many of
• 06:00 - 06:30 these areas you're going to see leap frog because the advantage of no legacy is you can leaprog so I think that's going to happen and but it'll have to be at uh you know the have to be efficient I think inference costs have to be super frugal because if you're going to have a billion people doing you know all kinds of queries or agentic stuff it has to work at at that scale so I think there's a lot but I see it happening but you know you you meet and so do I but we meet a lot of CEOs they're all excited
• 06:30 - 07:00 but little you know what to do next what is your advice to global CEOs yeah I think the the fundamental challenge um is that first question you asked which is change management right so if I if you think about um when the PCs first came out right in fact I met the CEO um of General Ali which is in Italian yeah milan so he started you know an insurance company and he was telling me about how in the
• 07:00 - 07:30 early 80s when he started um he'd have to go ask for permission to send a fax so I said "Oh my it must have been some compliance issue." He said "No it was not a compliance issue it was like expensive to send a fax." Uh and then PCs came out right and then you could literally say "Okay let me put an attachment in an email and send it." And it changed the entire process of how his agent brokers and them communicated same thing take forecasting right how did
• 07:30 - 08:00 forecasting happen in a multinational company like ours uh preC right i mean faxes went around somebody then did an inter office memo and then eventually 3 months later you had a forecast whereas then suddenly there was a new workflow which was email attachments and an Excel spreadsheet so I think that process change right when I think about what's happening even with copilot and agents inside a copilot as the UI layer it's a new workflow um and so the work process
• 08:00 - 08:30 has to change so good old in in the mid '90s we used to talk about business process re-engineering it's back again where you got to go look at how does any process right uh happen today and you got to re-engineer so that means it's moving the cheese around a little bit yeah and so that change management is the hard thing but then let me give you an example even just at Microsoft when I look at my marketing efficiency of dollars spent double digit customer service double digit uh internal IT ops
• 08:30 - 09:00 double digit right so you take any one of these so I'm putting that right into my budget so to as a CEO what I do is I literally say okay 10 10 points or 100 basis points of um operating leverage next year and then go take that for the next five years you compound it right so that means if that's going to be what the efficient frontier of any company is and once that is in the water every CEO
• 09:00 - 09:30 has to wake up and say you know because after all thanks to capital markets what they expect of CEOs is miracles every 90 days and So uh so if you don't produce that miracle well you're one of those who actually delivered every 90 but I think you know we we see it because we are the probably one of the world's largest users of co-pilot on GitHub you can see the impact already but it's not just about technology it's about process improvement it's about training people it's about changing mindsets you know the whole human issue that have to deal with it uh but I think this is something
• 09:30 - 10:00 which we'll we'll take off but one thing which people are I think you know I remember in the early days of the last two year there was this huge hype about AI will rule the world and all that but that seems to have died down and I think people are realizing that responsible AI can be done with innovations what's your take on all that I I think that yeah multiple things there right there's two sets of if I sort of had to characterize what's the big debates today one debate is of course hey are the scaling laws still working right because and I'm a
• 10:00 - 10:30 believer that the scaling laws are working it's just that you know with scale even things become harder so pre-training uh as you scale becomes harder because there's data challenge which you have to overcome just even cluster size I mean it's a massive distributed I mean this workload which is a synchronous data parallel workload is a unlike anything we have seen before too so even the systems problems are unique so there's pre-training itself then there is post-raining or you could say inference time compute uh which essentially is
• 10:30 - 11:00 like there is if you take even pre-training there's a massive sample piece to it we now have a better way to do that with all this chain of thought in a monologue and autograding it and what have you so therefore I think we'll continue to see some good algorithmic and systems innovation so I feel good about that then that same capability that's being used to produce capabilities in the model is what is needed for safety as well okay so when you think about guardrails reasoning
• 11:00 - 11:30 when I think about reasoning when you can inter you know when you can do the sampling look at the chain of thought autograde it that's the best way to do alignment in okay so you think some of these hallucination stuff will come hallucination like for example one of the services I'm most excited about even in Azure is this grounding service right so you can kind of use AI to check AI uh how grounded it is right so I think one of the frontiers right now will be a lot of how good are we at our eval for performance eval for
• 11:30 - 12:00 groundedness eval for safety uh and I think that the state-of-the-art is moving quite frankly like one of the other things I locked like if you remember every platform shift has had both the core system but also a core app server that we built whether it is for the web or the uh mobile and now we have a full-blown app server uh that we're building in our case called Foundry you know on everything how do I do fine-tuning how do I do evaluation how do I do safeguards and that way I think
• 12:00 - 12:30 so that application developers don't have to recreate it and you have made this famous statement that SAS is gone or dead no no i I that's what I mean that's kind of what the social media says i said I didn't say that um but what I said is you know like all things the the application architecture is changing right so it's kind of like saying did applications change with relational databases absolutely i mean if you remember I you know when I started even it was in the beginning of
• 12:30 - 13:00 relational database and all yeah and so but the bottom line is before that everybody built a vertical database inside the app right so basically they said oh here's a b tree let me build it in u and that was sort of the state-of-the-art and then suddenly said oh well we can have the separation between a database and have SQL and then I can have the app logic so right now with agents what's going to happen is a lot of the business logic will get separated so in some sense there's going to be databases which you can do crowd operations on um and then you will sort
• 13:00 - 13:30 of even manage eventual consistency between multiple of these SAS backends but a lot of the business logic will move uh to a new tier uh which then will be a multi- aent tier that needs to be orchestrated so it's going to be not I'm not going to one SAS application to another SAS applications I'm going to go to an agent that will orchestrate across multiple SAS applications so it's going to be a very big change that's why I think of this copilot as the UI for AI
• 13:30 - 14:00 like for example just to give you a real world feel i just go to copilot every day to get to my dynamic CRM okay i never go to dynamic CRM because all I go is at sales give me what's happening with Infosys and Microsoft and how are we working doing well and it'll come back with by the way not just data inside the CRM system but in all the communications a little and you and I may have had I get back so that ability to aggregate for a user query information across the front office
• 14:00 - 14:30 system or a productivity system a back office system multiple of them that's an agentic behavior that I think is going to be fairly disruptive in how users use things no I agree with you i think agents will really now drive enterprise consumption i think uh because it's very easy to tell a CEO that we can create an agent he can be a digital worker or he can amplify human potential i think it's it's easy also to communicate and I think that's going to be a big thing and I think this year will be a big it's kind of like humans and the swarm of
• 14:30 - 15:00 agents yeah yeah you know uh that I think is the next frontier as you point rightly pointed out and that's where I think a lot of the productivity uh will come from right which is the operating leverage for any function uh will come where the agency just like I I say you know and by the way the creation of agents today people mystify it and they will always be there will be the high end of you know what's the uh the specific agent about but just like I can build a spreadsheet I will build
• 15:00 - 15:30 thousands hundreds of agents that are working to help me streamline my own workflow and how do you see you know uh we are obviously have seen the consumer AI stuff but how do you differentiate enterprise AI from consumer AI I think the biggest one is obviously entitlements the if I sort of think about today what's the next big thing we talked about agents in order for agents to work memory and tools use right so basically you have a very sophisticated model let's say which is more capable
• 15:30 - 16:00 you then have to be able to give it a lot of context and the context is in memory uh but also what tools to use for what task and so if you can postrain essentially a model or inside of the model's invocation um there is tools use built in and memory is provided that's where and that by the way requires entitlements right so that means one of the other things that we're working very very hard on in
• 16:00 - 16:30 data governance and entitlements because agents can't work without scoping just like humans and security will be a big deal here you don't want the rogue agents roaming around that's right so you have entitlements memory and tools use and that I think is going to really change I think enterprise infrastructure enterprise application tier and enterprise usage sure yeah and I think one of the advantages of Azure is that because you're led from the front on
• 16:30 - 17:00 consumption of AI on Azure the learnings you have of what an AI workload looks like and how to design a cloud for that I think is unparalleled yeah and then two two two points there right most people talk about AI as if it sits in isolation interestingly enough if you look at inside even chat GPT chat GPT happens to be like the biggest users of Azure uh search right after all I mean vector search is a big part so it happens to be one of the biggest users or the biggest users of Cosmos DB because after all the user state still it's a by the way it's like agentic
• 17:00 - 17:30 stuff is stateful right so therefore you need databases uh so any AI application in fact needs all the other sort of services we built it doesn't sit on its own and to your point you know as a hyperscaler in general and in we understand the workload that's why we're building the app server and we're coming at it from two ends nondan one is this co-pilot as the UI for AI plus agents as extensions and we're coming at it from infrastructure and an app server from
• 17:30 - 18:00 the Azure side so those are the two bookends for us but quite frankly we're building them out as platforms so that everyone in this room can start building on top of it because that to me Microsoft has always been two things a platform company and a partner company you your roots are in that and I think you also create an ecosystem where everybody benefits which is different so but last question what is your call to action for all the CEOs here i think there are two calls right one is I would say the conversation you and I were having I think there's more of a meta
• 18:00 - 18:30 call I think from an India perspective which is any country that gets ahead in building out the new performance curve for me is tokens per dollar per watt right so tokens tokens so if you sort of really say where is the infrastructure getting built in such a way that the ability to generate tokens per dollar spent per watt used is the most effic efficient that I think is a countrywide there's private sector public sector
• 18:30 - 19:00 policy there's so many elements to it so that would be one call to action right whenever like our data centers again we just can't build data centers our data centers have to be permitted then they have to have access to renewable energy that renewable energy should be on a grid so we require a lot of upstream investments in any country then on the other side is you know there's a fantastic book uh by a guy called Jeffrey Ding which is I think it's called technology and great powers and competition but fundamentally he makes
• 19:00 - 19:30 the argument that it's not about even building the leading technology it's diffusing the leading technology and using it intensely that leads to great success that's right so it's kind of like the other side of it for all of us quite frankly including me is not just to talk about AI but is to use AI and diffuse it and diffuse it broadly inside the organization um and have it make a material difference in your customer service in your product development in your internal operations so I would say those
• 19:30 - 20:00 would be the two great and I think you taught us something new with software guys never thought of energy but now we have to think about watts so we'll do that thank you very much Satya thank you thank you so and Satya what a fascinating conversation but I want to bring the the leaders in the room in we've got a few minutes for open Q&A i'm Sarup Sahu the CEO for Accenture in India um considering it's a CEO connection session uh just checking your thoughts would we have a day when we have an AI CEO isn't it already [Laughter]
• 20:00 - 20:30 here no i I mean I I think that the if you sort of if you had gone back and said "Hey we're going to create um a billion people who can type um and we'll call it knowledge work right that's what we did with sort of what we did with the PCs and phones and essentially putting uh keyboards in front of people and touchcreens in front
• 20:30 - 21:00 of people to create information so to some degree I think about it whether it's a CEO or a frontline worker what I think we will have is these tokens that then work with our tokens it's kind of a bit of a I describe it as the new juggle bandi right think about a jazz session where AI and humans are able to create real leverage i'll tell you the one of the most fascinating even in my own personal behavior my fear of
• 21:00 - 21:30 stuff the unknown has just gone like I mean I'll give you like I was telling somebody like Punita and I were talking about God how much of high school math do you use i said "You know what i recently relearned my you know chain rule of probability because I said I better might as well know that that's the one thing to know if you really want to know the transformers." Uh and the fact is that expertise is getting commoditized knowledge is more accessible but that doesn't mean that
• 21:30 - 22:00 the system of knowledge creation is going to be something that we don't have agency in that I think is the key realization sort of you know it's not like I'm outsourcing my knowledge I am using these tools to create more knowledge and more knowledge turns that subtle difference is whether it's for a CEO or a frontline worker is going to be the big change and I think you're raising a larger question how do we all remain relevant right so I think it's by having two ends of the spectrum on one side
• 22:00 - 22:30 empathy collaboration intuition judgment which you can't replace on the other deal with first principles if you have these two ends covered the middle stuff will be in AI but the two ends you have to deal with yourself and by the way the one other nice metaphor some you know the CEO of Figma uses which I like a lot is um you know it reduces the floor and raises the ceiling simultaneously uh so if you take anything and you say "Wow both those
• 22:30 - 23:00 things are happening." Uh it's a pretty empowering way to think about it thanks a lot happy that I have my job still so Rajiv if you want to go next and then I'll come to you sir nandan Satya for a extremely exhilarating conversation india is going to be the AI use case capital of the world is a line I would like to plagiarize very often if I may have the permission please um my question is the following the last decade and Punit showed his digital life and his digital cler life but the last
• 23:00 - 23:30 decade has been dictated by let's say social media as a technology lots of positives but the one thing that came with social media was hash fact check because with so many people contributing you don't know what's factual versus what's fictitious AI assuming that's going to be the next generation and the next decade reality # hallucination becomes a uh counter side to that the more we contribute the more all
• 23:30 - 24:00 of us engage with AI and the more that the AI learns and agent AI learns from all of these inputs how do we deal with that kind of a reality yeah I mean I think we briefly chatted about it I mean I think that the there are two sides to it right one is um can you use AI itself essentially uh to do a lot more of that factchecking for you right I mean there are I would say two practical things if you sort of think about even our everyday habit now any AI response we
• 24:00 - 24:30 are learning to be inspectors of it right because it's a new skill like it's kind of like the editor skill uh is the new skill not just the input skills that we have right even like take GitHub copilot uh it's sort of fascinating what happens is when you do a continuation and it just comes you fundamentally look at that code and now a lot of the tools are about being able to explain this explain it to me in multiple ways um and so even when you have an agentic suite
• 24:30 - 25:00 you are really going to use lot more tools that you're going to build in order to be able to check the output so that's sort of one side of it the other side of it is can you build like you know the the real issue is when it starts doing sophisticated things that we don't even know about um it creates new science let's say uh how do you check that um and that I think is one of the frontiers of research which is how do you really responsibly scale this such that our ability to make sure that
• 25:00 - 25:30 the unintended consequences of this technology in the real world are things that we can deal with right I mean that that's the social media example which is you don't spring something in the world and then it breaks and then you deal with it you will have to responsibly learn how to scale this as a society and I think that we as tech companies need to be grounded that that's how the social permission even will work uh and so therefore we'll have we can't run ahead of our own ability to figure out how to responsibly scale it thank you go
• 25:30 - 26:00 ahead sir and then Mohit will come to you uh thank you very much Nandan two words you use system of knowledge and nandan you use first principle thinking i think these are two things which is required for AI to get a granularity of input knowledge frames which can fed into the AI to get a target knowledge frame which is the output to be there so in a banking and financial sector the issue is 90% is not right it's 99% right and that's a difference of 90 to 95%
• 26:00 - 26:30 right where the granularity of the knowledge which is available AI is able to deliver primary AI is not able to deliver that beyond 90 92% kind of a accuracy which is not applicable for actual application usage but as a secondary application it's possible but what is the steps you are taking in making 90% to 97% or 98% yeah so I think that first of all you kind of have to parse it out right so given the state of any frontier model at a point in time
• 26:30 - 27:00 you don't necessarily need to use it for something that has to be 100% accurate right i mean like you can even use good old machine learning for something that actually can be solved you know using that right so therefore you don't need to use general purpose intelligence uh that has an error bound uh so that's why I think of it as more an assist for what is knowledge work right which is knowledge work does have error bounds it has human error bounds so can we improve and reduce in fact the human error
• 27:00 - 27:30 bounds i was you know seeing I think ICICI Lombard was sort of showing me an example of their um uh insurance processing the claims processing i believe in India there's a pretty unique challenge where there's no standard even in how the claims come in and today whatever it takes you know um multiple hours or maybe mult you know tens of minutes uh and how they can reduce it so just even reducing the human drudgery in inspection that's where I would use the
• 27:30 - 28:00 general knowledge and then wherever you have something that is just more can be on rails I would even use traditional means and so that's one of the interesting things is the practical ways of using this technology are all now getting a lot more distributed I would say thank you Satya Nandan what a fascinating conversation thank you for pumping us up and folks thank you for being here lots more customer stories coming lots more interesting conversations so please do stay but thank you Satya thank you thanks guys thank you thank you so good always thank you thank you
• 28:00 - 28:30 [Music]