Your Wasted AI Compute Could Solve Real Problems (Agents for Humanity)

1 - > SPEAKER_03: Hi everyone and welcome back to our latest 2 - > episode of the Dumb Monkey Podcast. 3 - > It is a really special podcast here today because not only do 4 - > we, of course, have the Enterprise Monkey Dumb Monkey 5 - > Academy founder and entrepreneur extraordinaire Amir Katob, my 6 - > very good friend and host of the show. 7 - > Welcome, Amir. 8 - > SPEAKER_01: Thank you.

9 - > SPEAKER_03: But sitting next to Amir is his partner in the 10 - > fabulously named Agents for Humanity. 11 - > It's the fabulously named Steve Berg. 12 - > Welcome. 13 - > SPEAKER_02: Thank you.

14 - > It's great to be here, Davina. 15 - > Yeah. 16 - > SPEAKER_03: Well, guys, this is really exciting because not only 17 - > have you created Agents for Humanity, and I'm going to ask 18 - > you to tell us all about that, but the reason that you've 19 - > created Agents for Humanity, or the reason we get to talk about 20 - > it in this space at least, is because the UN runs an AI for 21 - > good challenge. 22 - > They hold challenges all over the world.

23 - > There was heats in Australia recently, and you too won with 24 - > Agents for Humanity. 25 - > Congratulations. 26 - > SPEAKER_01: Which tells us the state of innovation in 27 - > Australia. 28 - > If we have won it.

29 - > SPEAKER_03: Yeah, I feel like you're underselling it, and I 30 - > appreciate that because it's a competition and always, you 31 - > know, always come in low and then surprise them. 32 - > So that's okay. 33 - > Yes. 34 - > We'll accept that.

35 - > But it is a fabulous concept. 36 - > I haven't had anyone that I've explained it to that hasn't 37 - > absolutely lit up. 38 - > Um, so I'm gonna give you guys the challenge of responding to 39 - > me, pitching your idea to you, and you tell me if I'm getting 40 - > it right. 41 - > SPEAKER_01: That's a good idea.

42 - > Okay. 43 - > SPEAKER_03: So in the world of compute, big enterprise users 44 - > are now having very large programs or very large buys of 45 - > tokens that they are using, and they're left with a whole lot of 46 - > spare compute power. 47 - > Now, as we know, the cost of compute is really high. 48 - > All of us now, when we talk about the cost of tokens, it's 49 - > got nothing to do with playing battleship on a board game.

50 - > It is all about AI tokens and compute tokens. 51 - > So you guys know that there's a real value around this. 52 - > For enterprise-level users of AI, they can afford to pay the 53 - > big bucks to do this big compute and they're getting value out of 54 - > that. 55 - > But we know that AI has the power to make a huge change in 56 - > the world, to start tackling some of the intractable 57 - > problems, the difficult things, and one of the best ways to do 58 - > that is from the ground up, from people who are embedded in their 59 - > communities that see those problems for what they are and 60 - > would like to find a solution.

61 - > But of course, if you're a small one-person do-gooder, or you're 62 - > a small community organization, you don't have money to go 63 - > throwing on compute power. 64 - > It's hard enough to get the basic resources to make a 65 - > difference in the place where you are. 66 - > So you guys came up with this fabulous idea that went, what if 67 - > we could get enterprises to donate their spare compute, and 68 - > then we can use that compute to create agents and create a 69 - > community that creates agents to then donate to these do-gooding 70 - > people around the world?

71 - > And that's what you've done. 72 - > SPEAKER_00: That's amazing, man. 73 - > SPEAKER_01: I think this is our Geneva pitch. 74 - > Yeah, we don't need to re-ask it anymore.

75 - > SPEAKER_02: Cranch it down to two minutes and we're done. 76 - > Yes. 77 - > SPEAKER_03: Yeah, we can do this. 78 - > Yeah.

79 - > Oh, yes. 80 - > It is a fabulous idea. 81 - > Like it genuinely is. 82 - > It's such a simple idea, and that in itself is its genius.

83 - > Um where did the idea come from? 84 - > And what was the first step in making it happen? 85 - > SPEAKER_01: So it came uh from it came in different stages. 86 - > There was like it almost like evolved into something from 87 - > nothing.

88 - > So it originated from my father, and he was trying to help girls 89 - > in Aligar, which is a small town in India, uh, trying to help 90 - > girls educate uh and give them employment, employability. 91 - > But he was an engineer, so he didn't know how to do it. 92 - > So he was trying to work out what to do, how to do. 93 - > And so he reached out to me and my sisters like, how can I run 94 - > this program?

95 - > And we didn't know as well. 96 - > But I knew how to use AI. 97 - > My sister is a researcher, so we used our research and AI power 98 - > to actually work out how we can use AI agents to help him. 99 - > And uh, so me and Steve, we have been having these conversations 100 - > for like last one year about different AI concepts.

101 - > So I was sharing this story with him and how I'm helping my 102 - > father and some of the other organizations in a bit 103 - > unorganized way, use AI to help them. 104 - > And Steve looked at it and it's like his eyes lit up. 105 - > He's like, that's an excellent idea. 106 - > What if we could actually evolve into a global platform where we 107 - > could use the same thing but at a much larger scale?

108 - > And I think we continue to talk from there and we thought about 109 - > how what if, where where would that compute come from? 110 - > And it's like, what if people could actually contribute their 111 - > compute to it? 112 - > So it emerged from there and then evolved into the concept 113 - > which we know now known as the agents for humanity. 114 - > SPEAKER_02: I'd say, you know, Amir and I've had these 115 - > discussions far-ranging and sprawling discussions about the 116 - > things we're working on and different ideas, and I think 117 - > there always been this kernel of um how could we use technology 118 - > and in a distributed way to solicit contributions and 119 - > feedback and um the goodwill of people around the world to bring 120 - > it together to help solve problems.

121 - > And so it kind of evolved over time, and I think uh with the um 122 - > particular case that Amir had with his father to make it so um 123 - > such a tangible issue that we could try to solve and bring 124 - > something to bear on a specific problem that's extremely 125 - > helpful, and also then to look at how it can more yeah broadly 126 - > be deployed. 127 - > SPEAKER_01: So and then Steve's mind went crazy. 128 - > He started sending me what WhatsApp messages can we use it 129 - > to solve this problem, this problem in Kenya.

130 - > My my friend is uh you know running a health program. 131 - > Can we help him? 132 - > Can we do that? 133 - > It just went from there.

134 - > So it started with one simple problem, but then we looked at 135 - > the power of AI and how it can be actually solved to use so 136 - > many problems in the world. 137 - > SPEAKER_03: That to me is when you first told me about this, 138 - > Amir, that was the thing that really sang to me because when 139 - > we have conversations about AI, it's easy to get lost in the 140 - > tech sometimes. 141 - > But everything that you guys are describing and the process that 142 - > you've gone through to get to this point is about people.

143 - > It's about how do we use a technology to help to solve 144 - > people problems. 145 - > So tell me about the process of building this up into a working 146 - > model. 147 - > And I know that you've been deploying this model now. 148 - > So can you tell me a little bit about or tell us a little bit 149 - > about how it's being used now and functionally how does it 150 - > work?

151 - > If if it's a small, if it's a small you know, one man in India 152 - > that's sitting there saying, My mission is to um live the dream 153 - > of my wife and educate young girls, give them an education 154 - > that they wouldn't otherwise have. 155 - > How do I do that using Agents for Humanity? 156 - > SPEAKER_01: So very simple. 157 - > You go to agentsforhumanity.

ai, the platform. 158 - > It's going to go live, hopefully by the time this episode is 159 - > released. 160 - > You go there, post a problem, simply type it the problem that 161 - > you have. 162 - > We'll add voice features to it as well, and explain what the 163 - > problem is.

164 - > That's it. 165 - > It takes a couple of minutes to post a problem. 166 - > And once you have done that, there are a lot of contributed 167 - > agents in the platform that people are contributing towards. 168 - > They start to work on the problem.

169 - > So they start by taking that problem and dividing into 170 - > subproblems and saying, okay, what is the actual problem over 171 - > there? 172 - > Setting up research objective, and then new agents come and do 173 - > rigorous research on every sub-problem. 174 - > Once the research is completed, then uh an agent comes that 175 - > creates a proposal around that particular problem. 176 - > So that every problem through sub-problems, multiple proposals 177 - > are created, and then every proposal gets dissected and 178 - > argued and debated by agents who critique it, steel man it, 179 - > verify it.

180 - > And after all of that debate, a council of agents is set. 181 - > So what we mean is that for a particular problem, who would be 182 - > the right stakeholders who should have a say in it? 183 - > Would it be a school teacher, would it be a local uh 184 - > groundworker, would it be a mother of a girl? 185 - > All of those voices are represented through different 186 - > agents who look at a particular proposal and then score that 187 - > proposal in terms of what they think about it.

188 - > So, and then finally, the the proposal that gets 75 more than 189 - > 75% of the score or majority vote, that's the one that makes 190 - > it to synthesis. 191 - > So we might have like 20, 30 proposals for a particular 192 - > problem, but then the ones that survive make it to the final 193 - > solution. 194 - > But doesn't stop over there. 195 - > Once the solution is actually generated, an agent comes in and 196 - > sees what would a person on the ground need to actually 197 - > implement this solution.

198 - > They might need a budget, they might need cash flow, they might 199 - > need a funding application, a curriculum, a meal plan, and it 200 - > lists all of the things that they would need. 201 - > And down the new agents that come to that particular problem, 202 - > they work on creating those artifacts. 203 - > So the person on the ground who posted this problem, in a few 204 - > days' time, receive a full implementation pack that they 205 - > can take, print, use on the ground and implement straight 206 - > away.

207 - > SPEAKER_03: It's just brilliant thinking. 208 - > It it really is. 209 - > It's such a it's such a complete solution to what is so 210 - > demonstrably a real need. 211 - > How does it so I'm gonna I'm gonna get all touchy-feely about 212 - > it then.

213 - > Tell me about what it feels like to realize you've had these 214 - > conversations, you've seen this need, you've you've done all the 215 - > work to get this. 216 - > And I, you know, it's been a long time of working on this, a 217 - > lot of hours and a lot of conversations for you two. 218 - > But how does it feel now to be staring down the barrel of going 219 - > over to the UN in Switzerland to pitching this idea to the world 220 - > and knowing that you're going to make one way or the other, ages 221 - > for humanity is going to be out there?

222 - > SPEAKER_02: What does it mean to you? 223 - > Well, it it it's I think again on the personal level, it's it's 224 - > very satisfying when we know that there's in the case that uh 225 - > we're talking about now with Amir's father and girls that are 226 - > coming out of this program and they have a ability to generate 227 - > income to contribute into their family. 228 - > And that goes throughout the community. 229 - > We have people that are now looking at expanding the program 230 - > or looking at um how can they um become a facilitator of the 231 - > program in their in their community.

232 - > And I think one of the to make one note on what Amir was 233 - > saying, I think that in terms of how this is differentiated from 234 - > a LLM or something like that, where if you put something into 235 - > Chat GPT, it's just it's going to give you an answer because it 236 - > doesn't want to say it doesn't know the answer. 237 - > Um, and the way this is structured is such that um even 238 - > in the output that Amir's talking about, there are open 239 - > questions that come as part of the the deliverable 240 - > implementation pack.

241 - > And so that the humans in the field deploying these solutions 242 - > know let's keep an eye on these certain areas, let's generate on 243 - > the ground intelligence and feed it back into the platform so 244 - > that in a feedback loop, the the solution can continue to evolve 245 - > and get better over time. 246 - > And so I think that's another really innovative aspect of the 247 - > of the program is that it's not just here you go and see you 248 - > later. 249 - > SPEAKER_03: Um, we've delivered you a nice pack, off you go, and 250 - > do your good work.

251 - > Thank you for coming. 252 - > SPEAKER_02: And the idea of of the solution also being um open 253 - > access. 254 - > Yeah. 255 - > So that if you think about there might be Amir's father in India, 256 - > there might be another um educator or um workforce trainer 257 - > in America or South America or uh Northern Australia or all.

258 - > But there's a lot of overlap in some of the the problems are 259 - > different, but there's overlap um that can be drawn from and 260 - > extracted to help create another solution that's also customized. 261 - > SPEAKER_03: And the part of the part of what underpins this is 262 - > the donation of compute power. 263 - > How do you go about approaching people? 264 - > What's that process been like of getting getting the kind of 265 - > compute that you need?

266 - > SPEAKER_01: Yeah. 267 - > So first of all, we do not call it donation, uh, we actually 268 - > call it contribution. 269 - > Well, that's what I was supposed to do. 270 - > And and contribution.

271 - > Yes. 272 - > Because uh uh what we are asking people to do is we're asking 273 - > people to participate, it's participation. 274 - > So getting their agents to participate in a problem, and 275 - > the way it's unique and different is like, of course, 276 - > you know, how suppose on Reddit or Facebook or Wikipedia, people 277 - > would go and contribute their time to for uh to solve a 278 - > problem. 279 - > Here we are saying you we uh want you to contribute your AI 280 - > or agents to actually uh solve this problem.

281 - > Uh, but in terms of being asking people to do that, so far the 282 - > conversation that we have had, it's been such a positive uh 283 - > reception. 284 - > Like for me, because I've been working on this for a very long 285 - > time, it just felt like sometimes I feel like maybe I'm 286 - > just too obsessed with things, or I feel a bit shy about 287 - > talking about things because of course I personally felt it was 288 - > a great idea, uh, but I didn't feel know that I would get this 289 - > good validation.

290 - > SPEAKER_03: But and you have that very personal connection to 291 - > this idea because of the because of your family's connection to 292 - > that. 293 - > SPEAKER_01: That's right. 294 - > And I keep going back to India a lot. 295 - > And every time I go back to India, I still have got a home, 296 - > and like 500 meters from home, I meet the people that I grew up 297 - > with.

298 - > Some of them are still struggling. 299 - > And although I am doing all of these fancy things over here and 300 - > helping organizations become profitable and so on. 301 - > I when I go back to those communities, I feel like, am I 302 - > really doing something tangible for them? 303 - > Like, what's the I always felt like, what's the purpose of me 304 - > doing all of this if I cannot make a difference in anyone's 305 - > life?

306 - > You can give them, help them with some money, some programs 307 - > that you help with, but the state remains over there because 308 - > there's so many people around around out out on the ground 309 - > that are still struggling, still trying to help others as well, 310 - > but they don't do not have any resources. 311 - > So for me, it just feels like it's coming full circle. 312 - > Uh, because I've otherwise I I had a feeling of guilt in me as 313 - > well. 314 - > Like I'm doing all of these amazing things, and I'm trying 315 - > to help the people out there, but I was not able to do enough.

316 - > I feel with agents for humanity, we have this one chance for 317 - > humanity where we can actually empower the people who are 318 - > trying to do something and make a difference in the lives of 319 - > millions. 320 - > So the the feedback that we have got so far, Steve, you can talk 321 - > a bit more about is it's been really fantastic. 322 - > Like people are saying, how can we contribute? 323 - > or we have got a problem we would like to post.

324 - > How can we be involved? 325 - > Yes, yeah. 326 - > SPEAKER_02: I I think what we might might have kind of come 327 - > across here is an interest from both sides of the platform with 328 - > the um individuals or enterprises contributing into 329 - > the platform to help power the solutions and the organizations 330 - > on the other side. 331 - > It feels like there's maybe this bridge that's being created 332 - > where, like Amir saying, people want to help.

333 - > And it's easy to look at negative things happening in the 334 - > world. 335 - > But um I think at the core, everyone is good and optimistic 336 - > and wants to contribute. 337 - > And so this might be one way that they can do that on a small 338 - > scale. 339 - > And on the other side, you can imagine the millions of 340 - > organizations that are out there on the ground working on a cause 341 - > and feeling like they're you know yelling into a void and 342 - > there's no one on the other side that's really recognizing.

343 - > SPEAKER_03: Well, just staring at a wall, going, I don't know. 344 - > I don't know. 345 - > SPEAKER_02: And how do I do this? 346 - > And my budget is is is tight.

347 - > I don't have connections into others around the world that I 348 - > know are doing something similar, but I just don't have 349 - > those those connections. 350 - > Um, and I don't have the budget to pay for a high-priced 351 - > consultant to come in and help me. 352 - > Um, so there's just that hopefully that feeling we can 353 - > help create of support and goodwill and hope. 354 - > SPEAKER_01: And it's so funny, it's not even about like smaller 355 - > organizations.

356 - > We have had conversations with some of the really big 357 - > not-for-profits, but not only not-for-profits, like social for 358 - > good organizations, local councils, and everyone that we 359 - > had conversation with, they said, Oh, we have got a problem. 360 - > Here's the problem. 361 - > We we want to put this problem because it's like problems are 362 - > on the top of their mind. 363 - > They know that they're not have time or resources to be able to 364 - > solve those problems, and they would really like them to be 365 - > solved.

366 - > So that's there's a huge need for that. 367 - > And on the other hand, there are people who are willing to 368 - > contribute. 369 - > The AI exists, so why not use this for this purpose? 370 - > SPEAKER_03: Yeah, absolutely.

371 - > And have you two thought about now? 372 - > I know you're going over to to Switzerland. 373 - > I think the final is on the 9th of July, am I right? 374 - > Yes, yeah.

375 - > So you two are pitching. 376 - > SPEAKER_02: Yeah, you've got to just it's only one person that 377 - > can pitch in this round, and so Amir is on the hot seat for two 378 - > minutes. 379 - > SPEAKER_03: Two minutes. 380 - > SPEAKER_01: So it's two minutes pitch and three minutes question 381 - > and answers.

382 - > SPEAKER_03: Fantastic. 383 - > Have you have you considered what it means if you win? 384 - > SPEAKER_01: Uh you want to answer that? 385 - > SPEAKER_02: Uh we I I don't know.

386 - > It's almost um, we're not sure what that means. 387 - > Um and a lot of it's just the opportunity to be there and I 388 - > think help expand the the awareness in the world about 389 - > what we're doing and um hopefully have some positive 390 - > feedback, win or lose. 391 - > Um but I haven't thought too far ahead, to be honest. 392 - > SPEAKER_01: You know, I'll I'll go a bit philosophical over 393 - > here.

394 - > So earlier I I used to care a lot about winning, and that used 395 - > to give me motivation. 396 - > And when I was going to Perth, you know, Sarah asked me like, 397 - > How are you feeling? 398 - > You know, I'm praying for you, you should win. 399 - > And now I just feel like I do not need that motivation to for 400 - > for the win.

401 - > For me, if we can do our best and put our best front foot over 402 - > there, it really doesn't matter if we win or not. 403 - > Like, I think where we we have reached, it's it's almost like 404 - > really good. 405 - > And with these competitions, also about timings, who else is 406 - > there as well. 407 - > You know, if there's someone who's doing something really 408 - > well, we would like them to have that platform or acknowledged.

409 - > For us, if we are able to reach as many lives as we can and uh 410 - > across the globe, because only going to work if there are 411 - > people who are willing to, of course, put their problems, 412 - > there are people who are willing to bring their agents and 413 - > actually use that knowledge. 414 - > So it's like we've made this fantastic thing, it's for the 415 - > world to use. 416 - > And we, as two or three people or a small team over here, 417 - > cannot do anything with this.

418 - > It's now in everyone's hand, it's now in humans' hand to 419 - > actually use uh this amazing thing for the help of humanity. 420 - > So, my intention sort of going into Geneva is that if we can 421 - > actually connect with as many people as we can, find the 422 - > people who are willing to support some people who are 423 - > willing to listen to this podcast and can then help just 424 - > spread out the word, that would be the biggest win. 425 - > SPEAKER_03: Yeah, and that would be that would be a really big 426 - > win.

427 - > SPEAKER_02: I I I may also say that I think we'd be remiss if 428 - > we didn't um um acknowledge and thank Innovate Australia as 429 - > well, which was the um organization in Perth that um 430 - > was responsible for putting on the Australia competition. 431 - > And um Peter and the group there that's running that uh running 432 - > the organization, it was incredibly supportive of us 433 - > after we won and we've done sessions with them to help um 434 - > hone the pitch because it is a very it's a very difficult thing 435 - > to get down to this two-minute window.

436 - > SPEAKER_01: I'm still struggling with my friends. 437 - > SPEAKER_03: Yeah, and so you get incredible support though, don't 438 - > you? 439 - > Through that's one of the part of the magic of these of these 440 - > programs. 441 - > And there's it there's innovate organizations all around on it, 442 - > particularly Europe and the Western world into America, but 443 - > they do uh for reasonably small teams, they do an incredible job 444 - > of really going, what do you guys a version of what you guys 445 - > are doing with Agents for Humanity?

446 - > What do you need? 447 - > How can we support you to get there? 448 - > How do we give you the tools that you need? 449 - > Because no one comes in with all the answers on how to do this.

450 - > Yeah. 451 - > Having that support's invaluable, isn't it? 452 - > SPEAKER_01: And and it's such a it's such a darling group, 453 - > actually. 454 - > Like it's this uh this guy called Peter is an absolute gun 455 - > who's been uh cheering Innovate Australia and almost like 456 - > putting all of the strings together.

457 - > Such such a nice, lovely, down-to-earth guys, and uh the 458 - > panel of the judges, because what they also did was they 459 - > invited all of the participants to attend the pre-pitching 460 - > sessions where they provided the feedback and also like insights 461 - > and so on, and they were so good in volunteering their time to 462 - > teach and educate as well. 463 - > Even they're supporting us after this. 464 - > So I think the the support group that we have found over there is 465 - > absolutely amazing.

466 - > We've got two Phils over there. 467 - > Yeah. 468 - > So uh, and like uh the the level of help that these guys are 469 - > providing. 470 - > Is they do the pitch feedback session and then Phil would put 471 - > together all of the notes, like put it into AI and create like a 472 - > full feedback pack for us that we can take.

473 - > And now they are flying us to Geneva. 474 - > They're fund funding the the trip as well. 475 - > So it's I I think uh it's good to have that support system. 476 - > Yeah.

477 - > SPEAKER_02: And part of it too is that that the experience has 478 - > been so good. 479 - > Hopefully we do well. 480 - > Um, I would like to win, Amir, I have to admit. 481 - > Um, so don't don't mess it out.

482 - > Don't mess it up. 483 - > Uh um, but uh yeah, we're also hoping that maybe we can help 484 - > spread the word about their initiative. 485 - > It's it's it'll be the third year next year. 486 - > Um, I knew a group know a group significantly from the first 487 - > year.

488 - > So if we can foster the culture of innovation and AI for good 489 - > concept is um something we can hope to do that. 490 - > SPEAKER_03: Yeah, because it's an amazing name, isn't it? 491 - > I mean, AI for good is is a brilliant name. 492 - > Yeah, but it's only brilliant if it means something.

493 - > SPEAKER_04: Right. 494 - > SPEAKER_03: And it only means something if these sort of 495 - > projects actually come to fruition, get on the ground and 496 - > then grow and become real things. 497 - > So it's it's not to say AI for good, but what we want to see is 498 - > AI for good. 499 - > SPEAKER_00: You know what we are joking over there?

500 - > We're saying there's AI for good, there should be an AI for 501 - > bad competition. 502 - > Wouldn't want to see AI for bad competition. 503 - > Yeah. 504 - > SPEAKER_03: Are we not seeing that in the open AI and drop 505 - > competing?

506 - > SPEAKER_01: Who can make autonomous weapons? 507 - > SPEAKER_03: But there'd be a nice voting system with all of 508 - > it. 509 - > Maybe we can get the world involved in a voting system for 510 - > AI for bad. 511 - > Yeah.

512 - > SPEAKER_01: Yeah. 513 - > SPEAKER_03: Maybe not held by the UN, just saying. 514 - > SPEAKER_01: Right. 515 - > Plenty of AI for bad.

516 - > Plenty of AI for bad. 517 - > Maybe held by US. 518 - > So sorry. 519 - > SPEAKER_03: That's okay.

520 - > He's come over to the good sign. 521 - > Um I'm hearing in all of this that there's so much of you two 522 - > in this. 523 - > Tell me about the relationship of working together and what you 524 - > two bring to each other. 525 - > Because again, none of these things got off the ground 526 - > without great relationships, without a whole lot of buy-in.

527 - > Um, and you kind of need to find that magic when you're working 528 - > with people on these sort of projects. 529 - > And it seems that you two have. 530 - > What's the magic? 531 - > Romance and coffee.

532 - > SPEAKER_02: Well, it it it's that's where it started. 533 - > It was funny because Amir and I were introduced by a mutual 534 - > friend, um uh Adrian, who's uh a legend himself. 535 - > Yes. 536 - > And um he suggested for quite a while before we actually did 537 - > meet, and we finally, yeah, a year, 18 months ago, sat down 538 - > over coffee.

539 - > And um, I think we both had, like Amir was saying, we had 540 - > connections and ideas, and uh obviously we're not from 541 - > anywhere near one another where we grew up. 542 - > Our backgrounds are so different. 543 - > Um, in the US, I grew up in Pittsburgh, uh, believe it or 544 - > not, and lived in Colorado for a long time. 545 - > And so yeah, on the surface, nothing in common, but Adrian, I 546 - > think, sensed something of like we would do well to get 547 - > together, and we just have had um a really fun relationship 548 - > over the over the past um period of time.

549 - > SPEAKER_01: And for the first six months, we just had 550 - > meetings, chatted over the coffee, nothing happened. 551 - > I was like, we just would just have conversation and we it 552 - > would go in different places, and then we'll walk out of the 553 - > meetings like nothing substantial happened. 554 - > Like, why are we good? 555 - > Yes, yes.

556 - > SPEAKER_02: And I've I've had a background in um education with 557 - > a partner in the US um for a long time and a company there. 558 - > Um so I think Amir and I both sensed uh you know an interest 559 - > in doing something good, whatever that means and whatever 560 - > it meant. 561 - > And so as this idea kind of came together, I think this is what 562 - > um really kind of made the made the relationship come together. 563 - > And like this is something we should we have to at least give 564 - > it a shot.

565 - > SPEAKER_03: And uh well, you're certainly giving it a shot now. 566 - > Yes, and Amy, there's a there's a pitch that I've heard you 567 - > give, and I know that Steve's heard you give it a lot, but I 568 - > can't let you just sit here without actually giving us your 569 - > pitch. 570 - > I gave you mine. 571 - > SPEAKER_01: Oh I I I still muck it, like two minutes on the 572 - > clock.

573 - > SPEAKER_03: All right, let's go. 574 - > Agents for humanity and go. 575 - > SPEAKER_01: Okay, I'll skip the first first initial part. 576 - > Uh so Agent for Humanity is a free open source platform where 577 - > anyone with a problem can post their problem, and then hundreds 578 - > of contributed AI agents work on solving that problem.

579 - > But not just uh building the thesis, but actually building 580 - > real solutions, real implementation pack that people 581 - > can actually use in their day-to-day life when they are 582 - > working on the ground. 583 - > So it could be budgets, uh, it could be uh curriculums, 584 - > anything that they need on the ground. 585 - > It actually builds that. 586 - > The beauty with this platform is that uh no matter where you come 587 - > from, no matter who you are, you can actually post the problem.

588 - > It all happens in the open. 589 - > Everyone can have access to it, everyone can see it as well. 590 - > And so far, we have done uh around 40 pilots across five 591 - > countries, had some really amazing results. 592 - > But I keep going back to Sabrina from my father's cohort, who 593 - > initially couldn't like even read properly or do anything, 594 - > but now she's earning living for her family through the 595 - > embroidery and the program designed through this platform 596 - > and earning a living for her family.

597 - > And I think that's the biggest win for us. 598 - > SPEAKER_03: What do you think, Steve? 599 - > I think it's pretty good. 600 - > Yeah, it's the best.

601 - > SPEAKER_01: Yeah, too long. 602 - > unknown: Yeah. 603 - > SPEAKER_03: You were coming in around that two minutes. 604 - > I think it was.

605 - > SPEAKER_02: No, it was good. 606 - > It was good, yeah. 607 - > SPEAKER_01: It's a different version to the one I'm going to 608 - > do in June. 609 - > SPEAKER_03: I'm just well, you know, keep some keep something 610 - > back.

611 - > SPEAKER_01: Yes, absolutely. 612 - > Don't don't let it go. 613 - > I've I've seen, like I suck at all of my practice sessions, but 614 - > when I'm actually out there on the stage, that's when it sort 615 - > of like happens for me. 616 - > Yes, just switch to some some someone completely different.

617 - > SPEAKER_03: There's something about the energy of being in 618 - > those in those spaces, isn't there? 619 - > SPEAKER_01: Yes, and I really feel comfortable like being 620 - > there. 621 - > I really enjoy myself up on the stage. 622 - > Yeah.

623 - > But before that, I feel really nervous as well. 624 - > Yes. 625 - > SPEAKER_03: I think that's very, very normal. 626 - > And if I'm gonna give you um a magic wand that is probably AI 627 - > powered, but it is going to give you everything you need to get 628 - > agents to humanity, agents for humanity to the scale that it 629 - > could grow to, what could this thing do?

630 - > Tell me the difference it could make in the world. 631 - > SPEAKER_02: M my hope, I think, is is that it just gives people 632 - > agency, not a bad pun, perhaps, I guess, but I think it's okay 633 - > to bad pun in this space. 634 - > Yeah, yeah. 635 - > It's the for the workers on the ground in in these locations 636 - > like we were talking about that may feel alone and without much 637 - > hope.

638 - > Hopefully they can submit a problem, they can connect with 639 - > others, humans and AI to generate a solution that they 640 - > can really deploy and and make them feel like they have have 641 - > the ability and the agency to do it and make a change. 642 - > And on the other side, that those individuals contributing 643 - > into the platform can participate and help see the 644 - > results of what they're doing as well. 645 - > SPEAKER_03: So, what's to stop every small community 646 - > organization, every individual that's trying to do good in the 647 - > world, everyone that's trying to make a difference for the people 648 - > around them so that they can employ a good in the world, 649 - > what's to stop them in the future being able to access 650 - > agents for humanity?

651 - > SPEAKER_01: I would say two things. 652 - > First is if someone doesn't know about agents for humanity, then 653 - > they can't really access it. 654 - > I think once someone becomes aware of it, we are pretty 655 - > confident that they are going to use it. 656 - > Once they use it one time, they're going to use it.

657 - > So we want it to be their consultant, their chief of 658 - > operations, that they are continuously going back to solve 659 - > problems for the organization. 660 - > Another thing that can actually stop this from working is if we 661 - > do not have enough people contributing their AI agents 662 - > toward problems or causes. 663 - > So again, that goes to the reach. 664 - > So if you ask me what's the magic band that we would uh or 665 - > what's what's our ask, our ask is reach.

666 - > Like if we can reach to every single person on the earth, 667 - > everyone who's trying to do good so that they can use it, utilize 668 - > it, be a part of their ecosystem. 669 - > And every person with some sort of AI subscription or AI agent 670 - > who's using Chat GPT or Cloud or OpenClaw, everyone in the 671 - > developer community, uh, and so if they can contribute to it. 672 - > Uh and that can happen only by amplifying the voices and the 673 - > reach. 674 - > We we we do not need like anything else if we can get that 675 - > reach, if we can get the message out clear enough.

676 - > I think that's the biggest thing that we need. 677 - > SPEAKER_03: So we need people to talk about it so this thing can 678 - > take off and really be ages for humanity. 679 - > SPEAKER_01: Yes, absolutely. 680 - > And talk to your friends, share it, uh, share it on the social 681 - > media wherever you can, no matter how small or big your 682 - > following is.

683 - > We want people to share about it, encourage their friends to 684 - > participate. 685 - > It anything such a movement like this takes a bit of efforts in 686 - > the beginning to get off from the ground. 687 - > And that would require some amazing ambassadors to come on 688 - > board and and really believe in this and actually promote it, 689 - > shout from the top of their lungs and and be those 690 - > ambassadors. 691 - > So we are actually looking for for those amazing supporters who 692 - > can help us really take this off and reach everyone on the earth.

693 - > SPEAKER_03: Yeah. 694 - > Well, it's an incredible development. 695 - > It's a wonderful story, um, and it has hope and people right at 696 - > the right at the center of it. 697 - > So thank you for coming in to share it with us, all on the 698 - > Dumb Monkey podcast.

699 - > All the best for heading over to Switzerland. 700 - > We're really excited for you. 701 - > No pressure. 702 - > You're carrying the hopes of Australia and the world with 703 - > you, but no pressure.

704 - > Um but either way, I think we're all winners for these sorts of 705 - > developments. 706 - > So the more that we can bring people together, the more that 707 - > we can keep people at the heart of AI, and the more that we can 708 - > show people everywhere and start to talk about what AI can do for 709 - > good, I think we're all going to be better off. 710 - > So thank you again. 711 - > Thank you for being here.

712 - > This has been really fun. 713 - > Um, and all the best. 714 - > SPEAKER_01: Thank you again. 715 - > Thank you.

716 - > SPEAKER_03: And thank you to everyone who's joined in to 717 - > watch this episode of the Dumb Monkey Show. 718 - > Um, please follow along on the competition that's happening 719 - > around AI for good that is being run by the UN in uh Switzerland 720 - > in July. 721 - > SPEAKER_01: Yes, and if you want to check out Agents for 722 - > Humanity, go to agentsforhumanity.ai.

723 - > SPEAKER_0: Agentsforhumanity.ai. 724 - > Yes. 725 - > Online, on socials, where else are we gonna where else are we 726 - > gonna see it?

727 - > On the browser, just go search for it, find it, share it, 728 - > spread it everywhere. 729 - > Put little celebration emojis around it. 730 - > And and I'm gonna have one fingers crossed because that's a 731 - > lucky one. 732 - > But um, please, please get on board.

733 - > This is uh it's a wonderful piece of technology that I think 734 - > the world really needs. 735 - > So great job, you two, for giving it to us and great job, 736 - > everyone, for joining us. 737 - > Thank you. 738 - > We'll see you next time.