1 00:00:00,000 --> 00:00:02,720 All right, so Satya Nadella, the CEO of Microsoft, 2 00:00:02,720 --> 00:00:05,240 just made a pretty big statement in London. 3 00:00:05,240 --> 00:00:07,920 He's saying that AI agents are gonna be as common 4 00:00:07,920 --> 00:00:09,360 as apps on our phones. 5 00:00:09,360 --> 00:00:11,240 Wow, that's a... 6 00:00:11,240 --> 00:00:12,080 Can you imagine that? 7 00:00:12,080 --> 00:00:13,360 That is a bold vision. 8 00:00:13,360 --> 00:00:16,280 It's like we're on the cusp of like a whole new era 9 00:00:16,280 --> 00:00:17,560 of how we use technology. 10 00:00:17,560 --> 00:00:18,920 Exactly, and you know, 11 00:00:18,920 --> 00:00:21,720 Microsoft's AI assistant co-pilot, 12 00:00:21,720 --> 00:00:23,480 it's getting a huge upgrade. 13 00:00:23,480 --> 00:00:26,400 We're talking way beyond like setting reminders 14 00:00:26,400 --> 00:00:28,480 or playing your favorite playlist. 15 00:00:28,480 --> 00:00:30,920 It's like having a whole team of specialized assistants 16 00:00:30,920 --> 00:00:32,480 each with their own skill set. 17 00:00:32,480 --> 00:00:33,320 Oh, wow. 18 00:00:33,320 --> 00:00:34,640 You know, Nadella actually called it 19 00:00:34,640 --> 00:00:36,840 a constellation of AI helpers. 20 00:00:36,840 --> 00:00:40,120 A constellation, that's a cool way to think about it. 21 00:00:40,120 --> 00:00:40,960 Yeah. 22 00:00:40,960 --> 00:00:42,080 But how does this actually work? 23 00:00:42,080 --> 00:00:44,720 Like what can these AI agents do for us? 24 00:00:44,720 --> 00:00:46,720 Well, think about all the tasks that you do 25 00:00:46,720 --> 00:00:49,440 on your phone or your computer every day, you know? 26 00:00:49,440 --> 00:00:50,280 Sure. 27 00:00:50,280 --> 00:00:52,360 Scheduling meetings, booking travel, 28 00:00:52,360 --> 00:00:55,000 researching information, responding to emails. 29 00:00:55,000 --> 00:00:55,840 Yeah. 30 00:00:55,840 --> 00:00:56,920 These are all things AI agents 31 00:00:56,920 --> 00:00:59,160 could potentially handle for you. 32 00:00:59,160 --> 00:01:00,960 So it's like having a personal assistant 33 00:01:00,960 --> 00:01:02,640 who's also like a tech whiz. 34 00:01:02,640 --> 00:01:03,480 Exactly. 35 00:01:03,480 --> 00:01:04,320 Clay. 36 00:01:04,320 --> 00:01:06,560 But here's where it gets really interesting. 37 00:01:06,560 --> 00:01:10,440 Microsoft unveiled a new tool called co-pilot studio. 38 00:01:10,440 --> 00:01:12,400 Co-pilot studio, what's the deal with that? 39 00:01:12,400 --> 00:01:15,160 It's like a factory for building your own AI agents. 40 00:01:15,160 --> 00:01:16,000 Oh, wow. 41 00:01:16,000 --> 00:01:17,040 And the best part is you don't need 42 00:01:17,040 --> 00:01:18,680 to be a coding expert to use it. 43 00:01:18,680 --> 00:01:19,520 That's pretty cool. 44 00:01:19,520 --> 00:01:21,800 Nadella compared it to building a simple spreadsheet. 45 00:01:21,800 --> 00:01:24,600 So you're telling me that I could create my own AI agent 46 00:01:24,600 --> 00:01:26,640 even if I don't know anything about coding. 47 00:01:26,640 --> 00:01:30,280 That's the idea that Microsoft wants to empower anyone 48 00:01:30,280 --> 00:01:32,320 to create AI solutions that are tailored 49 00:01:32,320 --> 00:01:33,800 to their specific needs. 50 00:01:33,800 --> 00:01:34,640 Okay. 51 00:01:34,640 --> 00:01:38,480 I am intrigued, but I need a real world example. 52 00:01:38,480 --> 00:01:41,320 How are businesses actually using these AI agents? 53 00:01:41,320 --> 00:01:43,360 Well, there's a really fascinating case study 54 00:01:43,360 --> 00:01:44,960 featuring McKinsey. 55 00:01:44,960 --> 00:01:45,800 McKinsey. 56 00:01:45,800 --> 00:01:47,960 Yeah, the global consulting giant. 57 00:01:47,960 --> 00:01:48,800 Wow. 58 00:01:48,800 --> 00:01:50,560 They've built an AI agent that's revolutionizing 59 00:01:50,560 --> 00:01:52,600 how they onboard new clients. 60 00:01:52,600 --> 00:01:54,320 Onboard, so you mean like all that paperwork 61 00:01:54,320 --> 00:01:56,160 and back and forth at the beginning of a project? 62 00:01:56,160 --> 00:01:57,000 Exactly. 63 00:01:57,000 --> 00:01:59,720 It's often a very time consuming and tedious process. 64 00:01:59,720 --> 00:02:00,560 Yes. 65 00:02:00,560 --> 00:02:03,040 But McKinsey's AI agent is able to automate 66 00:02:03,040 --> 00:02:04,760 a huge chunk of it. 67 00:02:04,760 --> 00:02:07,880 And they're seeing a 90% reduction in lead time. 68 00:02:07,880 --> 00:02:09,640 90%, that's incredible. 69 00:02:09,640 --> 00:02:10,480 It is. 70 00:02:10,480 --> 00:02:11,600 But what does that actually look like? 71 00:02:11,600 --> 00:02:12,920 What's the AI doing? 72 00:02:12,920 --> 00:02:17,640 Well, think of it as like a super efficient assistant 73 00:02:17,640 --> 00:02:21,000 who's gathering all the necessary client information, 74 00:02:21,000 --> 00:02:23,040 performing initial risk assessments, 75 00:02:23,040 --> 00:02:25,680 and even generating preliminary reports. 76 00:02:25,680 --> 00:02:26,520 Wow. 77 00:02:26,520 --> 00:02:28,720 Tasks that used to take weeks 78 00:02:28,720 --> 00:02:30,800 are now being done in a matter of days. 79 00:02:30,800 --> 00:02:32,080 So it's not just saving time, 80 00:02:32,080 --> 00:02:33,880 it's actually changing the whole dynamic 81 00:02:33,880 --> 00:02:34,880 of the client relationship. 82 00:02:34,880 --> 00:02:37,120 Exactly, and that's just one example. 83 00:02:37,120 --> 00:02:42,120 Microsoft is also rolling out 10 new pre-built AI agents 84 00:02:42,160 --> 00:02:45,160 for their Dynamics 365 business software. 85 00:02:45,160 --> 00:02:46,840 So they're giving businesses a head start 86 00:02:46,840 --> 00:02:48,640 with ready made AI solutions. 87 00:02:48,640 --> 00:02:49,480 Very clever. 88 00:02:49,480 --> 00:02:50,560 Yeah, very clever. 89 00:02:50,560 --> 00:02:52,800 And these agents cover a wide range of functions 90 00:02:52,800 --> 00:02:54,640 from sales and marketing to finance 91 00:02:54,640 --> 00:02:55,880 and supply chain management. 92 00:02:55,880 --> 00:02:59,400 Okay, so let's zoom in on one of these pre-built agents. 93 00:02:59,400 --> 00:03:00,240 Okay. 94 00:03:00,240 --> 00:03:02,640 Give me an example that would be useful for, say, 95 00:03:02,640 --> 00:03:03,760 a sales team. 96 00:03:03,760 --> 00:03:06,760 Well, imagine a sales team that's drowning in leads. 97 00:03:06,760 --> 00:03:08,640 They're trying to figure out which ones are most likely 98 00:03:08,640 --> 00:03:10,800 to convert into actual customers. 99 00:03:10,800 --> 00:03:13,160 That's where the sales qualification agent comes in. 100 00:03:13,160 --> 00:03:15,960 Okay, so this AI agent helps them prioritize leads. 101 00:03:15,960 --> 00:03:20,240 Exactly, it automatically researches potential clients, 102 00:03:20,240 --> 00:03:21,400 analyzes their data, 103 00:03:21,400 --> 00:03:23,640 and flags the most promising prospects. 104 00:03:23,640 --> 00:03:24,480 Wow. 105 00:03:24,480 --> 00:03:26,440 Even drafts personalized emails 106 00:03:26,440 --> 00:03:28,040 for sales people to send out. 107 00:03:28,040 --> 00:03:29,880 So it's like having a research assistant 108 00:03:29,880 --> 00:03:32,280 and a marketing guru all rolled into one. 109 00:03:32,280 --> 00:03:33,400 That's a great way to put it. 110 00:03:33,400 --> 00:03:35,640 And the results speak for themselves. 111 00:03:35,640 --> 00:03:38,160 Microsoft claims that their internal sales teams 112 00:03:38,160 --> 00:03:40,480 are closing 20% more deals 113 00:03:40,480 --> 00:03:42,960 thanks to co-pilot and these AI agents. 114 00:03:42,960 --> 00:03:44,960 20% more deals, that's significant. 115 00:03:44,960 --> 00:03:48,200 It is, and it highlights how AI is becoming a game changer 116 00:03:48,200 --> 00:03:50,280 for businesses looking to increase efficiency 117 00:03:50,280 --> 00:03:51,160 and productivity. 118 00:03:51,160 --> 00:03:53,280 Okay, so Microsoft is clearly walking the walk 119 00:03:53,280 --> 00:03:55,720 using these AI tools themselves. 120 00:03:55,720 --> 00:03:57,160 But what about the bigger picture? 121 00:03:57,160 --> 00:03:59,680 How are they envisioning AI transform the world 122 00:03:59,680 --> 00:04:01,680 beyond just big businesses? 123 00:04:01,680 --> 00:04:03,120 Well, Nadella highlighted a partnership 124 00:04:03,120 --> 00:04:06,360 with the UK government to bring AI to the public sector. 125 00:04:06,360 --> 00:04:09,520 He talked about using AI to improve healthcare education 126 00:04:09,520 --> 00:04:11,280 and even energy efficiency. 127 00:04:11,280 --> 00:04:13,320 So AI isn't just for corporations, 128 00:04:13,320 --> 00:04:16,640 it has the potential to benefit society as a whole. 129 00:04:16,640 --> 00:04:19,160 Exactly, and it's not just talk. 130 00:04:19,160 --> 00:04:21,640 Data from LinkedIn shows a massive increase 131 00:04:21,640 --> 00:04:23,360 in AI skills globally. 132 00:04:23,360 --> 00:04:24,200 Really? 133 00:04:24,200 --> 00:04:26,960 People are recognizing the potential of this technology 134 00:04:26,960 --> 00:04:30,080 and investing in learning how to use it. 135 00:04:30,080 --> 00:04:31,120 It's fascinating. 136 00:04:31,120 --> 00:04:33,360 It sounds like we're on the cusp of a major shift 137 00:04:33,360 --> 00:04:36,160 with AI becoming like a fundamental part of our lives, 138 00:04:36,160 --> 00:04:37,480 both professionally and personally. 139 00:04:37,480 --> 00:04:38,840 Yeah. 140 00:04:38,840 --> 00:04:40,560 But there must be challenges too, right? 141 00:04:40,560 --> 00:04:43,960 Building and implementing AI on this scale can't be easy. 142 00:04:43,960 --> 00:04:45,000 You're absolutely right. 143 00:04:45,000 --> 00:04:46,880 Nadella acknowledged those challenges, 144 00:04:46,880 --> 00:04:50,040 talking about the need for robust data management systems 145 00:04:50,040 --> 00:04:51,880 and better developer tools to support 146 00:04:51,880 --> 00:04:53,320 the development of AI agents. 147 00:04:53,320 --> 00:04:54,160 Yeah. 148 00:04:54,160 --> 00:04:56,560 He gave us a glimpse into the massive infrastructure 149 00:04:56,560 --> 00:04:59,400 Microsoft is building, from Azure cloud computing 150 00:04:59,400 --> 00:05:02,440 to advancements in their developer platforms. 151 00:05:02,440 --> 00:05:03,920 So it's like they're laying the groundwork 152 00:05:03,920 --> 00:05:06,720 for a whole new technological ecosystem 153 00:05:06,720 --> 00:05:08,280 with AI agents at the center. 154 00:05:08,280 --> 00:05:10,920 Precisely, but it also raises some questions. 155 00:05:10,920 --> 00:05:13,520 How do we ensure that these powerful AI tools 156 00:05:13,520 --> 00:05:16,160 are used ethically and responsibly? 157 00:05:16,160 --> 00:05:17,000 Right. 158 00:05:17,000 --> 00:05:21,520 To prevent bias and discrimination in AI algorithms. 159 00:05:21,520 --> 00:05:22,800 Those are big questions. 160 00:05:22,800 --> 00:05:23,120 Yeah. 161 00:05:23,120 --> 00:05:25,000 And honestly, a whole other deep dive. 162 00:05:25,000 --> 00:05:25,520 Right. 163 00:05:25,520 --> 00:05:29,040 But for now, it seems clear that understanding AI agents 164 00:05:29,040 --> 00:05:32,240 and their implications is going to be crucial in the years 165 00:05:32,240 --> 00:05:32,880 to come. 166 00:05:32,880 --> 00:05:33,520 Absolutely. 167 00:05:33,520 --> 00:05:37,160 As Nadella emphasized, this isn't science fiction anymore. 168 00:05:37,160 --> 00:05:39,360 These AI tools are being used right now 169 00:05:39,360 --> 00:05:41,880 today to solve real world problems and boost 170 00:05:41,880 --> 00:05:44,000 productivity across various industries. 171 00:05:44,000 --> 00:05:44,440 Yeah. 172 00:05:44,440 --> 00:05:48,840 The key takeaway is the AI revolution is happening. 173 00:05:48,840 --> 00:05:49,200 It is. 174 00:05:49,200 --> 00:05:50,720 And it's time to get on board. 175 00:05:50,720 --> 00:05:51,320 I like that. 176 00:05:51,320 --> 00:05:51,760 Yeah. 177 00:05:51,760 --> 00:05:52,920 It's time to get on board. 178 00:05:52,920 --> 00:05:53,480 Yeah. 179 00:05:53,480 --> 00:05:53,760 OK. 180 00:05:53,760 --> 00:05:55,280 So we've covered a lot of ground here. 181 00:05:55,280 --> 00:05:56,000 We have. 182 00:05:56,000 --> 00:05:58,480 We've talked about the potential of AI agents 183 00:05:58,480 --> 00:06:02,640 to revolutionize how we work, how we live. 184 00:06:02,640 --> 00:06:05,560 But as with any new technology, there are also risks 185 00:06:05,560 --> 00:06:07,320 and challenges that need to be addressed. 186 00:06:07,320 --> 00:06:07,600 Right. 187 00:06:07,600 --> 00:06:09,120 We've just scratched the surface here. 188 00:06:09,120 --> 00:06:09,720 We have. 189 00:06:09,720 --> 00:06:11,600 But I think it's clear that AI agents 190 00:06:11,600 --> 00:06:13,960 are going to be a major force in the years to come. 191 00:06:13,960 --> 00:06:15,000 It seems that way. 192 00:06:15,000 --> 00:06:17,440 And it's going to be fascinating to see how this technology 193 00:06:17,440 --> 00:06:18,760 continues to evolve. 194 00:06:18,760 --> 00:06:21,280 It is going to be very interesting to see what happens next. 195 00:06:21,280 --> 00:06:21,960 Absolutely. 196 00:06:21,960 --> 00:06:22,440 All right. 197 00:06:22,440 --> 00:06:25,120 So in the next part of our deep dive, 198 00:06:25,120 --> 00:06:29,200 we're going to take a closer look at the issue of trust in AI. 199 00:06:29,200 --> 00:06:30,480 OK. 200 00:06:30,480 --> 00:06:35,040 How can we ensure that these powerful AI systems are 201 00:06:35,040 --> 00:06:36,840 developed and used responsibly? 202 00:06:36,840 --> 00:06:37,320 Right. 203 00:06:37,320 --> 00:06:40,640 We'll explore the ethical considerations, 204 00:06:40,640 --> 00:06:43,320 the potential biases, and the safeguards 205 00:06:43,320 --> 00:06:45,680 that are being put in place to mitigate those risks. 206 00:06:45,680 --> 00:06:46,720 That's really important. 207 00:06:46,720 --> 00:06:47,120 It is. 208 00:06:47,120 --> 00:06:50,040 So stay tuned for part two of our deep dyes 209 00:06:50,040 --> 00:06:51,520 into the world of AI agents. 210 00:06:51,520 --> 00:06:52,160 Great. 211 00:06:52,160 --> 00:06:52,440 Yeah. 212 00:06:52,440 --> 00:06:53,440 It's fascinating, isn't it? 213 00:06:53,440 --> 00:06:56,080 We've gone from AI being this thing that's 214 00:06:56,080 --> 00:06:58,920 working behind the scenes to now it's front and center. 215 00:06:58,920 --> 00:06:59,080 Yeah. 216 00:06:59,080 --> 00:07:01,640 It's almost like AI is graduating from tool to teammate. 217 00:07:01,640 --> 00:07:02,760 That is a great analogy. 218 00:07:02,760 --> 00:07:05,960 And just like with any good teammate, trust is essential. 219 00:07:05,960 --> 00:07:06,840 That makes sense. 220 00:07:06,840 --> 00:07:08,680 You need to be able to rely on your AI teammates 221 00:07:08,680 --> 00:07:10,920 to know they're giving you accurate information 222 00:07:10,920 --> 00:07:12,160 and making the right decisions. 223 00:07:12,160 --> 00:07:12,920 Exactly. 224 00:07:12,920 --> 00:07:14,760 And that's why Nadella really emphasized 225 00:07:14,760 --> 00:07:18,280 this idea of trustworthy AI. 226 00:07:18,280 --> 00:07:20,720 It's not enough to just build these powerful systems. 227 00:07:20,720 --> 00:07:22,520 We also need to make sure they're being developed 228 00:07:22,520 --> 00:07:24,080 and used responsibly. 229 00:07:24,080 --> 00:07:26,560 So how does Microsoft plan to build this trust? 230 00:07:26,560 --> 00:07:28,880 Like what are they doing to address those concerns 231 00:07:28,880 --> 00:07:30,720 about ethics and responsibility? 232 00:07:30,720 --> 00:07:33,800 They're really focusing on three key areas. 233 00:07:33,800 --> 00:07:34,440 OK. 234 00:07:34,440 --> 00:07:36,520 Security, privacy, and AI safety. 235 00:07:36,520 --> 00:07:37,120 OK. 236 00:07:37,120 --> 00:07:39,080 So in terms of security, they're 237 00:07:39,080 --> 00:07:41,600 implementing really rigorous testing 238 00:07:41,600 --> 00:07:44,320 to prevent things like prompt injections. 239 00:07:44,320 --> 00:07:45,040 Prompt injections? 240 00:07:45,040 --> 00:07:45,600 What are those? 241 00:07:45,600 --> 00:07:48,920 Imagine someone trying to trick an AI into doing something 242 00:07:48,920 --> 00:07:49,680 it shouldn't. 243 00:07:49,680 --> 00:07:50,240 Oh, OK. 244 00:07:50,240 --> 00:07:52,240 Like finding a back door into the system. 245 00:07:52,240 --> 00:07:52,920 OK. 246 00:07:52,920 --> 00:07:54,720 So Microsoft is building safeguards 247 00:07:54,720 --> 00:07:57,440 to catch these injections before they can cause any harm. 248 00:07:57,440 --> 00:07:58,240 That's reassuring. 249 00:07:58,240 --> 00:07:59,320 So what about privacy? 250 00:07:59,320 --> 00:08:03,120 How are they protecting our data in this world of AI agents? 251 00:08:03,120 --> 00:08:04,320 Privacy is paramount. 252 00:08:04,320 --> 00:08:07,400 Microsoft is using something called confidential computing 253 00:08:07,400 --> 00:08:08,960 to keep data secure. 254 00:08:08,960 --> 00:08:09,480 OK. 255 00:08:09,480 --> 00:08:11,520 Think of it like your data being processed 256 00:08:11,520 --> 00:08:13,240 inside a locked box. 257 00:08:13,240 --> 00:08:16,400 Even the AI can't see the raw information, 258 00:08:16,400 --> 00:08:18,360 just the results of its analysis. 259 00:08:18,360 --> 00:08:21,160 So even the AI can't snoop on my data. 260 00:08:21,160 --> 00:08:21,960 Exactly. 261 00:08:21,960 --> 00:08:23,520 It's like having an accountant who 262 00:08:23,520 --> 00:08:25,680 can manage your finances without ever seeing your bank 263 00:08:25,680 --> 00:08:27,040 statements. 264 00:08:27,040 --> 00:08:27,320 OK. 265 00:08:27,320 --> 00:08:29,120 That makes me feel a little bit better 266 00:08:29,120 --> 00:08:32,440 about sharing sensitive information with an AI. 267 00:08:32,440 --> 00:08:34,360 But what about this AI safety thing? 268 00:08:34,360 --> 00:08:34,760 Right. 269 00:08:34,760 --> 00:08:35,680 What does that involve? 270 00:08:35,680 --> 00:08:38,320 AI safety is all about making sure 271 00:08:38,320 --> 00:08:40,800 that AI systems behave as intended 272 00:08:40,800 --> 00:08:43,920 and they don't produce any harmful or biased outputs. 273 00:08:43,920 --> 00:08:44,520 OK. 274 00:08:44,520 --> 00:08:46,440 One of the biggest challenges right now with AI 275 00:08:46,440 --> 00:08:48,280 is this thing called hallucination. 276 00:08:48,280 --> 00:08:50,200 Hallucination, like seeing things that aren't there. 277 00:08:50,200 --> 00:08:50,700 Kind of. 278 00:08:50,700 --> 00:08:54,360 It's when an AI just makes things up, presents information 279 00:08:54,360 --> 00:08:55,720 that's not true. 280 00:08:55,720 --> 00:08:58,600 You remember that time when your AI assistant gave you 281 00:08:58,600 --> 00:08:59,440 a recipe? 282 00:08:59,440 --> 00:08:59,720 Yeah. 283 00:08:59,720 --> 00:09:01,960 That included a totally made up ingredient. 284 00:09:01,960 --> 00:09:02,200 Yeah. 285 00:09:02,200 --> 00:09:03,800 We all had a good laugh about that one. 286 00:09:03,800 --> 00:09:04,120 Right. 287 00:09:04,120 --> 00:09:05,520 Well, in that case, it was harmless. 288 00:09:05,520 --> 00:09:07,880 But in other situations, AI hallucinations 289 00:09:07,880 --> 00:09:10,120 could have really serious consequences. 290 00:09:10,120 --> 00:09:10,720 Right. 291 00:09:10,720 --> 00:09:13,960 So Microsoft is developing ways to detect and mitigate 292 00:09:13,960 --> 00:09:15,120 these hallucinations. 293 00:09:15,120 --> 00:09:15,800 Oh, wow. 294 00:09:15,800 --> 00:09:18,240 It's like having the AI double check its own work. 295 00:09:18,240 --> 00:09:20,720 So they're fighting AI fire with fire? 296 00:09:20,720 --> 00:09:21,600 In a way, yes. 297 00:09:21,600 --> 00:09:24,520 It's about creating a layered approach to safety, 298 00:09:24,520 --> 00:09:27,640 using multiple techniques to ensure these AI systems are 299 00:09:27,640 --> 00:09:30,160 reliable and trustworthy. 300 00:09:30,160 --> 00:09:31,640 So it sounds like Microsoft is really 301 00:09:31,640 --> 00:09:33,320 putting a lot of thought and effort 302 00:09:33,320 --> 00:09:35,640 into building responsible AI. 303 00:09:35,640 --> 00:09:36,160 They are. 304 00:09:36,160 --> 00:09:37,960 And they recognize that it's not just 305 00:09:37,960 --> 00:09:39,240 about the technology itself. 306 00:09:39,240 --> 00:09:39,480 Right. 307 00:09:39,480 --> 00:09:42,600 It's also about human oversight and collaboration. 308 00:09:42,600 --> 00:09:43,000 Right. 309 00:09:43,000 --> 00:09:45,840 AI should be augmenting our capabilities, 310 00:09:45,840 --> 00:09:47,400 not replacing us entirely. 311 00:09:47,400 --> 00:09:48,080 Exactly. 312 00:09:48,080 --> 00:09:50,840 It's about creating this partnership between humans 313 00:09:50,840 --> 00:09:52,640 and AI, where each can contribute 314 00:09:52,640 --> 00:09:54,880 their own unique strengths. 315 00:09:54,880 --> 00:09:57,520 So it's not about AI taking over our jobs, 316 00:09:57,520 --> 00:10:00,880 but rather transforming how we work, 317 00:10:00,880 --> 00:10:02,680 making us more efficient and productive. 318 00:10:02,680 --> 00:10:03,320 That's the goal. 319 00:10:03,320 --> 00:10:05,680 And this transformation is already happening. 320 00:10:05,680 --> 00:10:08,240 In addition to those really impressive internal results 321 00:10:08,240 --> 00:10:11,160 that Microsoft shared, they also showcase some really cool 322 00:10:11,160 --> 00:10:13,560 real-world use cases from their partners. 323 00:10:13,560 --> 00:10:15,120 I'd love to hear about those. 324 00:10:15,120 --> 00:10:17,880 How are other companies actually using these AI tools 325 00:10:17,880 --> 00:10:19,000 in practice? 326 00:10:19,000 --> 00:10:20,400 One example that really stood out 327 00:10:20,400 --> 00:10:22,200 was the British Heart Foundation. 328 00:10:22,200 --> 00:10:24,880 They're using Azure Speech Services 329 00:10:24,880 --> 00:10:28,280 to create these incredibly realistic simulations 330 00:10:28,280 --> 00:10:29,800 of emergency calls. 331 00:10:29,800 --> 00:10:30,120 Really? 332 00:10:30,120 --> 00:10:31,120 What's the purpose of that? 333 00:10:31,120 --> 00:10:33,600 They're using these simulations to train people 334 00:10:33,600 --> 00:10:37,000 how to respond effectively in high-pressure situations. 335 00:10:37,000 --> 00:10:37,480 Oh, wow. 336 00:10:37,480 --> 00:10:41,560 Like, imagine talking someone through CPR over the phone. 337 00:10:41,560 --> 00:10:43,360 It's a safe and controlled environment 338 00:10:43,360 --> 00:10:46,520 to practice those really critical skills. 339 00:10:46,520 --> 00:10:48,200 So it's like using AI to save lives. 340 00:10:48,200 --> 00:10:48,840 Exactly. 341 00:10:48,840 --> 00:10:51,200 And another great example comes from HSBC, 342 00:10:51,200 --> 00:10:53,040 the global banking giant. 343 00:10:53,040 --> 00:10:53,640 Oh, OK. 344 00:10:53,640 --> 00:10:55,640 They've actually developed AI agents 345 00:10:55,640 --> 00:10:58,120 to streamline their credit approval process. 346 00:10:58,120 --> 00:11:00,160 Oh, that sounds like it would save a ton of time. 347 00:11:00,160 --> 00:11:02,360 I mean, manually reviewing all those loan applications 348 00:11:02,360 --> 00:11:03,240 must be a nightmare. 349 00:11:03,240 --> 00:11:03,800 It is. 350 00:11:03,800 --> 00:11:07,640 And these AI agents can analyze financial data, assess risk 351 00:11:07,640 --> 00:11:09,520 factors, even recommend approval decisions. 352 00:11:09,520 --> 00:11:10,280 Oh, wow. 353 00:11:10,280 --> 00:11:12,240 This frees up human loan officers 354 00:11:12,240 --> 00:11:15,880 to focus on more complex cases or provide more personalized 355 00:11:15,880 --> 00:11:17,200 customer service. 356 00:11:17,200 --> 00:11:19,680 It's amazing to think about how AI is changing 357 00:11:19,680 --> 00:11:20,960 the way banks operate. 358 00:11:20,960 --> 00:11:22,680 Any other cool examples? 359 00:11:22,680 --> 00:11:23,240 Absolutely. 360 00:11:23,240 --> 00:11:24,200 There's Mondra. 361 00:11:24,200 --> 00:11:25,560 It's a fascinating company. 362 00:11:25,560 --> 00:11:28,720 They're building a digital twin for food safety and supply 363 00:11:28,720 --> 00:11:29,800 chain management. 364 00:11:29,800 --> 00:11:30,320 Wow. 365 00:11:30,320 --> 00:11:33,320 They're using AI to track food from the farm 366 00:11:33,320 --> 00:11:36,400 all the way to your table, improving both sustainability 367 00:11:36,400 --> 00:11:37,560 and safety along the way. 368 00:11:37,560 --> 00:11:38,080 That's incredible. 369 00:11:38,080 --> 00:11:41,600 Imagine the impact that could have on reducing food waste 370 00:11:41,600 --> 00:11:43,840 and making sure that the food we eat is actually safe. 371 00:11:43,840 --> 00:11:44,680 It's really impressive. 372 00:11:44,680 --> 00:11:46,040 And then there's Wave. 373 00:11:46,040 --> 00:11:46,880 Wave. 374 00:11:46,880 --> 00:11:50,800 They're a startup taking this really bold AI first approach 375 00:11:50,800 --> 00:11:53,760 to advanced driver assistance systems. 376 00:11:53,760 --> 00:11:56,640 They're using AI to develop self-driving technology that 377 00:11:56,640 --> 00:11:59,480 can learn and adapt to real world driving conditions. 378 00:11:59,480 --> 00:12:01,960 So they're teaching cars to drive themselves. 379 00:12:01,960 --> 00:12:04,200 That sounds like something straight out of a sci-fi movie. 380 00:12:04,200 --> 00:12:05,000 It really does. 381 00:12:05,000 --> 00:12:07,800 And all of these examples really highlight 382 00:12:07,800 --> 00:12:11,800 just how diverse the applications for AI agents are. 383 00:12:11,800 --> 00:12:14,760 From health care to finance to transportation, 384 00:12:14,760 --> 00:12:16,960 the possibilities seem endless. 385 00:12:16,960 --> 00:12:19,520 It's mind blowing to think about how much AI is already 386 00:12:19,520 --> 00:12:20,920 changing the world. 387 00:12:20,920 --> 00:12:22,560 But where does it go from here? 388 00:12:22,560 --> 00:12:24,880 What's the next frontier in AI development? 389 00:12:24,880 --> 00:12:26,480 That is the big question, isn't it? 390 00:12:26,480 --> 00:12:29,360 And the Della hinted at the going importance of something 391 00:12:29,360 --> 00:12:32,240 called edge computing in the future of AI. 392 00:12:32,240 --> 00:12:33,680 Edge computing refreshed my memory. 393 00:12:33,680 --> 00:12:35,080 What's that all about? 394 00:12:35,080 --> 00:12:38,440 So in simple terms, it's about bringing that AI processing 395 00:12:38,440 --> 00:12:41,320 power closer to the source of the data. 396 00:12:41,320 --> 00:12:43,800 So instead of relying solely on the cloud, 397 00:12:43,800 --> 00:12:48,240 edge computing allows AI to run directly on devices. 398 00:12:48,240 --> 00:12:48,880 Oh, wow. 399 00:12:48,880 --> 00:12:52,480 Like phones, laptops, even specialized hardware. 400 00:12:52,480 --> 00:12:55,840 So it's like having a mini AI brain right there in your pocket. 401 00:12:55,840 --> 00:12:56,440 You got it. 402 00:12:56,440 --> 00:13:01,080 And this shift to edge computing is being driven by a few factors. 403 00:13:01,080 --> 00:13:02,840 We need faster processing speeds. 404 00:13:02,840 --> 00:13:05,320 We need increased beta privacy. 405 00:13:05,320 --> 00:13:08,120 And also, we need the ability for AI to work, 406 00:13:08,120 --> 00:13:10,160 even when there's no internet connection. 407 00:13:10,160 --> 00:13:10,480 Right. 408 00:13:10,480 --> 00:13:11,360 That makes sense. 409 00:13:11,360 --> 00:13:14,240 So how does edge computing actually fit into this whole idea 410 00:13:14,240 --> 00:13:15,480 of AI agents? 411 00:13:15,480 --> 00:13:18,160 Well, imagine a world where your AI agents can operate 412 00:13:18,160 --> 00:13:20,280 independently on your devices. 413 00:13:20,280 --> 00:13:22,680 Learning from your personal data and your preferences 414 00:13:22,680 --> 00:13:24,640 without having to constantly connect to the cloud. 415 00:13:24,640 --> 00:13:26,920 That sounds both incredibly convenient 416 00:13:26,920 --> 00:13:28,160 and a little bit creepy. 417 00:13:28,160 --> 00:13:31,440 It definitely raises some questions about data ownership 418 00:13:31,440 --> 00:13:32,400 and control. 419 00:13:32,400 --> 00:13:35,680 But it also opens up some really exciting possibilities 420 00:13:35,680 --> 00:13:38,680 for personalized AI experiences. 421 00:13:38,680 --> 00:13:39,480 Like what? 422 00:13:39,480 --> 00:13:41,920 Well, think of an AI assistant that knows you so well. 423 00:13:41,920 --> 00:13:44,800 It can actually anticipate your needs 424 00:13:44,800 --> 00:13:48,080 and seamlessly integrate with your daily routines. 425 00:13:48,080 --> 00:13:49,320 That's pretty wild. 426 00:13:49,320 --> 00:13:51,640 But I can also see some potential downsides. 427 00:13:51,640 --> 00:13:54,040 What if AI gets too personalized? 428 00:13:54,040 --> 00:13:55,640 Could it create these echo chambers 429 00:13:55,640 --> 00:13:57,920 or reinforce existing biases? 430 00:13:57,920 --> 00:13:59,720 You're raising some really valid concerns. 431 00:13:59,720 --> 00:14:01,760 And these are exactly the types of questions 432 00:14:01,760 --> 00:14:04,600 we need to be thinking about as this technology keeps 433 00:14:04,600 --> 00:14:05,520 advancing. 434 00:14:05,520 --> 00:14:08,120 Developing ethical guidelines and safeguards 435 00:14:08,120 --> 00:14:09,960 is going to be crucial to make sure 436 00:14:09,960 --> 00:14:13,760 that AI is used responsibly and for the benefit of everyone. 437 00:14:13,760 --> 00:14:14,880 It's a lot to consider. 438 00:14:14,880 --> 00:14:16,480 But one thing's for sure, the future of AI 439 00:14:16,480 --> 00:14:18,000 is going to be anything but boring. 440 00:14:18,000 --> 00:14:18,880 Absolutely. 441 00:14:18,880 --> 00:14:21,360 And Nadella actually left us with a pretty thought-provoking 442 00:14:21,360 --> 00:14:22,560 question to ponder. 443 00:14:22,560 --> 00:14:23,640 Oh, was that? 444 00:14:23,640 --> 00:14:25,560 He asked, what human capabilities 445 00:14:25,560 --> 00:14:28,720 will we uniquely value in an AI-powered world? 446 00:14:28,720 --> 00:14:30,680 Yeah, that is a great question. 447 00:14:30,680 --> 00:14:32,520 It really makes you think about what 448 00:14:32,520 --> 00:14:36,120 it means to be human in a world where AI is becoming so capable. 449 00:14:36,120 --> 00:14:36,880 It does. 450 00:14:36,880 --> 00:14:40,320 It gets to the heart of why Microsoft is betting so heavily 451 00:14:40,320 --> 00:14:41,880 on this technology. 452 00:14:41,880 --> 00:14:44,680 Nadella really believes that AI agents represent 453 00:14:44,680 --> 00:14:48,200 a fundamental shift in how we interact with technology. 454 00:14:48,200 --> 00:14:51,600 We're moving from a world of apps to a world of agents. 455 00:14:51,600 --> 00:14:54,640 So we've talked about the whats of AI agents. 456 00:14:54,640 --> 00:14:58,680 They can automate tasks, analyze data, even interact 457 00:14:58,680 --> 00:15:00,320 with people on our behalf. 458 00:15:00,320 --> 00:15:03,120 But what about the why? 459 00:15:03,120 --> 00:15:04,760 Help me understand this shift. 460 00:15:04,760 --> 00:15:07,880 Why is Microsoft so convinced that agents are the future? 461 00:15:07,880 --> 00:15:09,160 OK, well, think about it this way. 462 00:15:09,160 --> 00:15:11,120 Apps are like tools. 463 00:15:11,120 --> 00:15:13,320 Each one is designed for a specific purpose. 464 00:15:13,320 --> 00:15:14,800 You need to know which tool to use 465 00:15:14,800 --> 00:15:16,960 and how to use it to get the result that you want. 466 00:15:16,960 --> 00:15:19,680 So choosing a hammer to drive a nail or a wrench 467 00:15:19,680 --> 00:15:20,560 to tighten a bolt. 468 00:15:20,560 --> 00:15:21,280 Exactly. 469 00:15:21,280 --> 00:15:24,640 But AI agents are more like skilled craftspeople. 470 00:15:24,640 --> 00:15:26,200 You tell them what you want to achieve, 471 00:15:26,200 --> 00:15:28,200 and they figure out the best way to do it using 472 00:15:28,200 --> 00:15:29,600 the right tools and techniques. 473 00:15:29,600 --> 00:15:30,880 So instead of me having to learn how 474 00:15:30,880 --> 00:15:32,600 to use all these different apps, 475 00:15:32,600 --> 00:15:34,760 the AI agent will just do the work for me. 476 00:15:34,760 --> 00:15:35,520 That's the idea. 477 00:15:35,520 --> 00:15:37,160 It's like having a personal assistant who 478 00:15:37,160 --> 00:15:39,200 knows how to use all the tools in your workshop 479 00:15:39,200 --> 00:15:41,560 and can just handle tasks for you automatically. 480 00:15:41,560 --> 00:15:43,320 OK, I'm starting to see the appeal. 481 00:15:43,320 --> 00:15:46,320 It's like having this team of invisible helpers 482 00:15:46,320 --> 00:15:48,840 that are working behind the scenes to make my life easier. 483 00:15:48,840 --> 00:15:52,400 Precisely, and this shift from apps to agents 484 00:15:52,400 --> 00:15:56,760 has the potential to really transform not just how we work, 485 00:15:56,760 --> 00:15:58,400 but also how we live. 486 00:15:58,400 --> 00:15:59,200 Give me an example. 487 00:15:59,200 --> 00:16:02,280 How could AI agents actually impact our everyday lives 488 00:16:02,280 --> 00:16:03,120 outside of work? 489 00:16:03,120 --> 00:16:05,880 OK, well, imagine your AI agent managing your calendar, 490 00:16:05,880 --> 00:16:08,040 booking appointments, making travel arrangements, 491 00:16:08,040 --> 00:16:10,560 even controlling all your smart home devices, 492 00:16:10,560 --> 00:16:12,960 all based on your preferences and your habits. 493 00:16:12,960 --> 00:16:15,240 So it's like having a personal concierge service 494 00:16:15,240 --> 00:16:16,720 available 24-7. 495 00:16:16,720 --> 00:16:17,840 That's a great way to put it. 496 00:16:17,840 --> 00:16:21,040 And as AI agents become more sophisticated, 497 00:16:21,040 --> 00:16:23,440 they could even handle much more complex tasks, 498 00:16:23,440 --> 00:16:25,760 like financial planning, health care management, 499 00:16:25,760 --> 00:16:27,560 or even personalized education. 500 00:16:27,560 --> 00:16:29,800 That's a pretty amazing vision of the future. 501 00:16:29,800 --> 00:16:33,040 But it also feels like a big leap from where we are today. 502 00:16:33,040 --> 00:16:34,840 Like, how do we get from here to there? 503 00:16:34,840 --> 00:16:35,720 It is a big leap. 504 00:16:35,720 --> 00:16:38,760 And there are certainly challenges and risks to consider. 505 00:16:38,760 --> 00:16:41,360 But Nadella's keynote was a very clear signal 506 00:16:41,360 --> 00:16:44,840 that this is the direction technology is heading. 507 00:16:44,840 --> 00:16:46,720 The AI revolution is happening now, 508 00:16:46,720 --> 00:16:48,920 and it's time to start thinking about how we can prepare 509 00:16:48,920 --> 00:16:50,240 for this future. 510 00:16:50,240 --> 00:16:52,200 So what can we actually do? 511 00:16:52,200 --> 00:16:55,520 Should we all go out and start learning how to code AI agents? 512 00:16:55,520 --> 00:16:58,720 Not necessarily just like, you don't need to be a mechanic 513 00:16:58,720 --> 00:16:59,680 to drive a car. 514 00:16:59,680 --> 00:17:01,440 You don't need to be an AI expert 515 00:17:01,440 --> 00:17:03,440 to benefit from these AI agents. 516 00:17:03,440 --> 00:17:04,000 OK. 517 00:17:04,000 --> 00:17:08,080 The key is to stay informed, be open to new possibilities, 518 00:17:08,080 --> 00:17:11,320 and really start thinking about how AI could enhance 519 00:17:11,320 --> 00:17:12,920 your work and your life. 520 00:17:12,920 --> 00:17:14,240 That's great advice. 521 00:17:14,240 --> 00:17:15,680 Any final thoughts for our listeners 522 00:17:15,680 --> 00:17:18,480 as they kind of ponder this brave new world of AI agents? 523 00:17:18,480 --> 00:17:21,920 You know, Nadella's closing question really stuck with me. 524 00:17:21,920 --> 00:17:25,280 He asked, what human capabilities will we uniquely 525 00:17:25,280 --> 00:17:27,720 value in an AI-powered world? 526 00:17:27,720 --> 00:17:31,640 It's a powerful reminder that as we embrace all these possibilities 527 00:17:31,640 --> 00:17:35,200 that AI offers, we shouldn't lose sight of what makes us human. 528 00:17:35,200 --> 00:17:36,720 That's a perfect note to end on. 529 00:17:36,720 --> 00:17:38,600 So to our listeners, as we continue 530 00:17:38,600 --> 00:17:41,560 to explore the depths of this AI revolution, 531 00:17:41,560 --> 00:17:43,640 remember to stay curious, ask questions, 532 00:17:43,640 --> 00:17:44,840 and challenge assumptions. 533 00:17:44,840 --> 00:17:47,600 Until next time, keep diving.