1 00:00:02,320 --> 00:00:02,840 Hi guys. 2 00:00:02,840 --> 00:00:03,840 Welcome to this session. 3 00:00:03,840 --> 00:00:09,240 So in this session we want to discuss the AI business ecosystem, the different layers in this and the 4 00:00:09,240 --> 00:00:11,280 opportunities across these layers. 5 00:00:11,760 --> 00:00:12,840 So let's have a look at this. 6 00:00:12,840 --> 00:00:17,400 So the artificial intelligence ecosystem is comprising of multiple layers. 7 00:00:17,400 --> 00:00:24,360 So the starting off with semiconductors semiconductors are basically chips like GPUs and TPUs which 8 00:00:24,360 --> 00:00:26,000 powers AI computation. 9 00:00:26,040 --> 00:00:32,480 On top of it there is cloud infrastructure platforms like AWS, Google Cloud, which offers scalable 10 00:00:32,480 --> 00:00:41,680 computing, and then foundation models, which are pre-trained AI models like OpenAI's ChatGPT for enables 11 00:00:41,680 --> 00:00:43,000 custom solutions. 12 00:00:43,320 --> 00:00:49,280 And then you will have orchestration frameworks, which are basically tools which can connect and manage 13 00:00:49,280 --> 00:00:51,320 multiple AI services together. 14 00:00:51,800 --> 00:00:58,840 And then the last is going to be application layer, where AI driven products like chat bots and recommendation 15 00:00:58,840 --> 00:00:59,880 engines work. 16 00:01:00,200 --> 00:01:02,840 So let's have a look at each of these layers one at a time. 17 00:01:03,080 --> 00:01:08,010 So first is semiconductors the basically the AI hardware, which we are talking about. 18 00:01:08,010 --> 00:01:14,330 So what it is, it's basically these are going to be specialized chips like GPUs, TPUs, AI accelerators. 19 00:01:14,610 --> 00:01:20,970 Uh, and why it is needed is because it provides the massive computing power needed for any type of 20 00:01:20,970 --> 00:01:22,410 AI tasks, which we are doing. 21 00:01:23,690 --> 00:01:30,810 The key players in this particular segment are Nvidia, AMD, Intel who are manufacturing these products, 22 00:01:30,810 --> 00:01:37,010 these semiconductors, and the if you look at it the opportunity wise, it's developing chips is capital 23 00:01:37,010 --> 00:01:37,530 incentive. 24 00:01:37,570 --> 00:01:40,810 Businesses can access them through cloud services. 25 00:01:40,810 --> 00:01:44,210 So there are a lot of benefit of this particular product. 26 00:01:44,890 --> 00:01:48,650 On top of it, there is cloud infrastructure, which is the AI power hub. 27 00:01:48,890 --> 00:01:54,690 Now this is basically an on demand computing, storage and AI services which they are providing. 28 00:01:55,690 --> 00:02:01,450 Why it matters is because it provides scalable resources without owning any hardware. 29 00:02:01,450 --> 00:02:06,330 So a lot of small medium businesses can avail it without any issues. 30 00:02:06,730 --> 00:02:13,450 Key players who are providing these services right now are AWS, Google Cloud, Microsoft Azure. 31 00:02:13,770 --> 00:02:17,450 All these are different platforms which you can use for cloud infrastructure. 32 00:02:17,770 --> 00:02:24,210 And then if you look at it, opportunity wise, startups can access enterprise level tech with pay as 33 00:02:24,210 --> 00:02:26,130 you go payment methods. 34 00:02:26,290 --> 00:02:30,970 So easily they can pick up these services and they can use it in their in their business. 35 00:02:31,690 --> 00:02:38,210 Now if you look at on top of this the foundation model, the AI brains, which is basically the pre-trained 36 00:02:38,210 --> 00:02:42,650 AI models like ChatGPT, Cloud Llama, all these are the tools. 37 00:02:42,690 --> 00:02:49,890 Now these ones enables advanced AI capabilities without building models from scratch. 38 00:02:50,290 --> 00:02:57,570 The key players who are who are who are the platforms in this right now are OpenAI, meta and anthropic. 39 00:02:57,730 --> 00:03:03,290 Anthropic which has built cloud and then opportunity wise, if you look at it, businesses can easily 40 00:03:03,330 --> 00:03:09,130 fine tune these models for custom solutions, which they want at a fraction of a development cost. 41 00:03:09,690 --> 00:03:16,540 Now, on top of this, what we get the next is Agentic orchestration, the AI workflow manager. 42 00:03:16,540 --> 00:03:23,060 So these are tools that can manage and coordinate between multiple AI services and APIs. 43 00:03:23,580 --> 00:03:28,340 Now this is something we call as the AI agents, which can work in here. 44 00:03:28,540 --> 00:03:34,060 And it matters because it enables seamless automation across complex AI systems. 45 00:03:34,380 --> 00:03:40,380 Now the key players who are providing this service right now is auto AI. 46 00:03:40,820 --> 00:03:45,700 All these companies and the benefit of it is businesses can streamline processes. 47 00:03:45,700 --> 00:03:50,860 Because of this, they can reduce development time and increasing their efficiency. 48 00:03:50,860 --> 00:03:51,740 Scalability. 49 00:03:52,580 --> 00:03:57,740 And then finally comes the applications layer which is where the AI is in action. 50 00:03:57,900 --> 00:04:03,140 And these are AI powered products basically delivering real world solutions. 51 00:04:03,180 --> 00:04:08,780 It drives business value through automation, personalization and efficiency. 52 00:04:08,820 --> 00:04:13,700 The key key players in this are chatbots, which we use on any platform. 53 00:04:13,700 --> 00:04:17,620 You have seen recommendation engines, AI powered analytics. 54 00:04:17,860 --> 00:04:22,860 All these are the different products of Applications layer where it is being applied. 55 00:04:23,180 --> 00:04:29,220 Now, opportunity wise, if you look at it, entrepreneurs can build market ready apps with minimal 56 00:04:29,220 --> 00:04:31,620 investment using these cloud APIs. 57 00:04:32,220 --> 00:04:37,420 If you look at it, the AI opportunities for businesses, there are plenty of them for small medium 58 00:04:37,420 --> 00:04:41,540 businesses, large organizations as well for small medium businesses. 59 00:04:41,580 --> 00:04:47,260 They can use pre-built models and cloud APIs to build cost effective API apps. 60 00:04:47,900 --> 00:04:53,780 They can automate tasks like customer service marketing operations through the help of these API tools. 61 00:04:54,020 --> 00:05:00,460 They can also focus on application and orchestration layer for quick market entry in cases of large 62 00:05:00,460 --> 00:05:01,420 organizations. 63 00:05:01,540 --> 00:05:06,540 They can develop proprietary AI models and enterprise level workflows. 64 00:05:06,700 --> 00:05:10,940 They can also scale the automation and analytics for deeper insights. 65 00:05:11,100 --> 00:05:17,260 They can invest across all the layers, from infrastructure to applications for industry leadership. 66 00:05:17,700 --> 00:05:24,100 So these are all the developments you can say or the layers which are available in the AI business currently.