Meet The Cambridge PhD Building The Tech To Prove What's Real Online w/ Mansoor Ahmed | OpenOrigins

‍00:00
Mansoor Ahmed:
If you don't do this, we can't trust anything online. Five years into the future, I started designing a system around proving photos and videos. In a world where AI has made creating photos and videos trivial, what can I do to help an honest source prove that what they're sending is real? I genuinely find this fun and I find it fun to read AI papers. I find it fun to experiment with new technologies. Our goal is to make photos and videos trustworthy in a ubiquitous manner without any barriers to entry. So anyone anywhere should be able to prove photos. Anywhere, anywhere should be able to trust photos.

00:39
Amardeep Parmar:
Today's guest is Mansoor Ahmed founder and CEO of Open Origins. For all of you worried about AI and how do you tell if something's true or not online? That's exactly what Manny's company is trying to solve. So he's got a long history as a PhD researcher from Cambridge and he's a specialist in the technology side. Right. Of like, how can you tell by looking at the pixels, by understanding the data, whether it's been the real reflection of it or whether it's actually been doctored. So you can see how Manny himself just really loves the technology and now it's now created into a business with a thriving and growing team that's expanding across the world. Really hope you enjoy this episode and very topical. I'm Amma from the BAE HQ. Great to have you there today.

01:26
Amardeep Parmar:
So great to see everything after you but who you want when you're growing up. What did you want to be? What did you think that one day would be your dream?

01:33
Mansoor Ahmed:
I think the first thing that I remember wanting to be is a tennis player. My dad was a the national tennis champion India and I wanted to be just like him. So I started playing tennis when I was like 6 years old. I really wanted to be a Dennis player. And by the time I was 12, I discovered that I'm a big nerd and probably better suited to something bookish rather than tennis. And I started being really interested in physics. So the second thing I wanted to be was a physicist. My parents shot down both of those ideas. Roe standard middle class Indian. So, like, either you can be a doctor or an engineer. And so I ended up not being a tennis player or a physicist and end up doing computer science instead.

02:20
Amardeep Parmar:
Yeah. And it's interesting because in many ways you had a different role model to what many people have.

02:25
Amardeep Parmar:
Right.

02:25
Amardeep Parmar:
Many people had role mode and he was more in a professional career.

02:28
Mansoor Ahmed:
Yeah.

02:28
Amardeep Parmar:
Whereas I think we could say it's more fun to have Like a daddy's a tennis player, right? Because it's.

02:33
Mansoor Ahmed:
Oh, my dad wasn't a tennis player professionally. He played till the national zones, but he was actually a forest conservator. So I grew up in, like, tiny little villages all over India. We used to move every two years. I didn't have any neighbors until I was like 13 years old. So in that way, I think I had a very different childhood too. Most people that you meet outside of India, from India, you know, most of them grew up in an urban area. I didn't really live in an Indian city at all. The biggest place I lived in must have had like a thousand people until I moved to Switzerland.

03:06
Amardeep Parmar:
With Switzerland University or when did you.

03:08
Mansoor Ahmed:
Yes, I. I moved to Zurich for my master's. Did my master's at the Swiss Federal Institute of Technology. It was a great place, but it was definitely a bit of a culture. Going from a forest India to Zurich.

03:20
Amardeep Parmar:
Did you find that adaption hard or was it something which you kind of took in your shot?

03:24
Mansoor Ahmed:
Really hard? I remember the first winter, especially in Zurich, was just like depths of depression. But I kind of forced myself to become much more of an extroverted person. I was a introverted nor shy kid. And I was like, that's not going to fly here. I don't know anyone. If I want to know people, if I want to have a community, I need to put myself out there. So I remember setting myself a challenge. Like, for a month, I'm going to go talk to five strangers. And could be as simple ask me at the time, but for a month, I have to talk to 5B strangers. They've done it.

03:57
Mansoor Ahmed:
I real, like halfway through the month, I started realizing I'm actually enjoying myself talking to his rider people and kind of continue doing that and kind of done myself from an introvert to an extrovert.

04:07
Amardeep Parmar:
Because obviously you've been going from the.

04:09
Amardeep Parmar:
First place six guts, right.

04:10
Amardeep Parmar:
Because you're going out your comfort zone.

04:12
Mansoor Ahmed:
Yeah.

04:12
Amardeep Parmar:
What made you do it?

04:14
Mansoor Ahmed:
This is. This is going to sound like such a beautiful Sydney answer. Okay. I wanted to get to the best university for the sick. I want to study, and at that time. And I'd get a scholarship. and it was basically eth, Cambridge, Stanford. Those are the three options I'd given myself. Got it to eth, so I went there. But I'm really thankful that I went to ETH because being there made me really question why I'm doing the work I'm doing. Until that point, I was just being like way pragmatic. I'm like, okay, I want to do computer science. I'm going to do this field. I was doing natural language processing research, but back in 2014. .Before it was cool. Yeah. And the only reason I was doing it is because I was like, this is the next thing.

04:57
Mansoor Ahmed:
This is what's going to be popular. And then at around that time, I came across Edward Snowden's work. I don't know if you're familiar with it. Yeah. But let's just watch this for the audience. Edward Snowden was a whistleblower. He's a whistleblower who debuted a lot of mass surveillance that the US and other its allies were doing on basically everyone. And I found the story really inspirational. You know, for him to give up his nice cushy life just because his conscience caught him to and basically become a fugitive for life. He's basically in house arrest in Russia right now. Made me really question why I am doing the work I'm doing.

05:34
Mansoor Ahmed:
You know, the pathway I could see for myself is if I'll do NLP research, I would go work for an advertising company or Google or something and kind of contribute to that surveillance network that I am not a big fan of. And that's when I decided to switch my focus and start working on security. The first security thing that I actually started working on was allowing missile blowers to contact journalists to anonymously release documents even if they're under surveillance. Yeah. And that's how I got into my PhD as well.

06:08
Amardeep Parmar:
Because obviously you chose that academic route initially.

06:10
Amardeep Parmar:
Right.

06:10
Amardeep Parmar:
And you mentioned how your parents wanted to be a doctor or an engineer or up accountant.

06:14
Mansoor Ahmed:
Yeah.

06:14
Amardeep Parmar:
How did that go down or how did you make not understand...

06:16
Mansoor Ahmed:
Only doctor, only a tutorial. I didn't really see myself ever doing that job. I am pretty bad at taking instruction, so I knew that if I'm going down this road, the only eventuality I'll be happy with, at least that's how I conceived of it when I was like 18, is I want to be a professor, I want to be a lecturer, I want to be able to do cool research that I find cool, not because someone told me to do so. And I also had role models at that time. I was a big fan of Michio Kaku who was a physicist. Stephen Hawking, obviously, just like every self respecting nerd. And those characters really propelled me to adopting this sort of goal for myself without really thinking about it. You know, they're just cool people.

07:08
Mansoor Ahmed:
I remember watching Carl Sagan growing up and that created a model of this ideal academic. And I wanted to be like them.

07:19
Amardeep Parmar:
And in that space. Right. Because worth hearing from people who are academia yet is that often restrictions can be quite limiting as well. And where people want to do a certain thing because that's what they believe in. But then there's reclamations and there's the university wants you to do one thing and you want to do another thing. How is jug in that? Because I think often people underestimate. That's quite a skill to be able.

07:39
Mansoor Ahmed:
To survive in that environment. Yeah. I think for the most part I didn't face the horrors of academia that you hear about. I was really fortunate to have an amazing PhD supervisor who's sadly passed away last year, Ross Anderson. And he really gave me full freedom to do whatever I wanted to do as long as it was new. He was like. He used to have this saying, he would say research is. Research is done with a shovel, not with a spoon. So like if someone is already dug up the earth and you're just trying to find the tiny little optimization, that's not research, that's just you doing basic optimization. You need to go and date new Earth. And so for the three years of my PhD, I just did whatever I wanted to do. I had no meetings.

08:32
Mansoor Ahmed:
I had just absolute freedom and access to all of the resources that peer which had to offer. It was genuinely one of the best periods of my life as far as work was concerned. I'm pretty much the same during my postdoc. I knew at the back of my mind that the moment I get into a lecturership, the moment I stop having students, this is going to change. Yeah. Most of my time is going to look like, you know, applying for grants and doing exams and things like that. And I knew that wasn't for me either. So it's. It's very fortunate timing that I left academia when I left academia. So I still have a very rosy nostalgic feeling about.

09:13
Amardeep Parmar:
Sometimes there's that thing where opportunities come about at the time you need them. Right. You just don't know why. Yeah. But I think a lot of people feeling that it's maybe where. Anyway, it's like, okay, I've done well enough in what I'm doing, but I don't now know what to do.

09:25
Mansoor Ahmed:
Yeah.

09:26
Amardeep Parmar:
And at this stage, did start mirroring company ever come into your mind or is it just.

09:31
Mansoor Ahmed:
It really wasn't one of the things that I was thinking applied to me. I consider myself a fairly reductant entrepreneur. I wanted to make cool things and I went into academia trying to make cool things. And then I could see that the future for me involved less and less making cool things and more paperwork. And it just so happened that I met one of your former guests, Sakshi Mittal, at a networking event and was talking to her about my research and she goes, you know what, you really should talk to my husband, Shravin. He's got this VC fund. I think he'd love to fund your research. And I was like, I don't know about that. We'll see. Okay, fine, we'll have a chat. I have a call with Shravin, we chat.

10:16
Mansoor Ahmed:
We were supposed to talk for 15 minutes, we end up talking for an hour and then we have another call. And at the end of the call he's like, you know what? If you want to start a company, I'm ready to write the first check and we'll co found the company. So you're putting me in a really difficult spot here. Put a lectureship by Yup. At Cambridge and I've got this. Now what do I do? I remember having conversations with my family and friends and they're like, the lectureship is probably going to be waiting for you. You could probably make your way back. But this thing that you've got an offer, this feels like a once in a lifetime kind of opportunity. So then I said yes to Shravin and we co founded Open Origins together.

10:53
Amardeep Parmar:
And even with that as well. Right, because it's one thing to start a company, but you have to have an idea that you really want to pursue. So what's the story there? What made you decide to.

11:02
Amardeep Parmar:
Yeah, Tackle.

11:03
Amardeep Parmar:
Put me with it.

11:04
Mansoor Ahmed:
Yeah. So going back to my PhD, I was really interested in helping journalists do their job. Like that's kind of been a recurring passion inspired by Edward Snowden. And when I initially made that system for whistleblowers to contact journalists anonymously, I worked quite deeply with a bunch of large you can use organizations and we actually deployed that system. So now if you go to, let's say the Guardiand you go to their contact us page, it will send you to anonymous portal where you can upload things. But that's part of the system that I designed. But as I'm talking to these journalists, they start telling me, look, anonymity is great. You've made the whistle door's life much easier, but you made the journalist's life a lot harder.

11:49
Mansoor Ahmed:
Because part in journalism is establishing trust in what you're receiving and the way traditionally journalists would have done is by establishing in the Boston Trust with the source that's sending your documents, that becomes impossible if there's real anonymity in it, there's no way to find out who sent it.

12:06
Mansoor Ahmed:
So all you've got are the photos and videos that this person has sent you. I'm like, okay, now I've kind of, you know, made the situation a lot harder. What can I do to help them? I started thinking about, okay... Exactly. So I started thinking, okay, fine, I don't want to compromise on anonymity because we live in a mass servants world. What can I do to help an honest source prove that what they're sending is real? Can I put something on their device that does not compromise their identity, that does not compromise their location, but proves for a fact that the photos that they're sending have not been touched upon, have not been modified in any way, and that what the reporter is seeing on the other side is an accurate representation of reality.

12:57
Mansoor Ahmed:
Yeah, and that's kind of the initial genesis of Open origins. This is 2018 is when I first wrote the academic paper. And during that time in academia at least there was already growing awareness of how generative AI is going to affect photos and videos. Like, we knew that there was a writing on the wall for trust in photos. We knew that there was going to be a time where you couldn't just look at the pixels and believe it. So kind of singing five years into the future, I started designing a system around proving photos and videos in a world where AI has made creating photos and videos trivial. Wrote a couple of academic papers. They ended up being quite controversial.

13:44
Mansoor Ahmed:
I remember getting into a few arguments at the time my supervisor refused to appear co author on my paper because he's like, I fundamentally don't agree with this approach. Oh, wow. But I was like, no, this is the only way we are going to reach a point where photos will be trivial to create and humans will not be able to distinguish them.

14:01
Amardeep Parmar:
I'm sure there's many people listening right now who is thinking, wait, this is exactly what we're seeing, right?

14:07
Mansoor Ahmed:
Yeah, yeah. I think I was about five years ahead of the girl there. And once I published all the papers, I realized that there was nothing else I could do within academia. You know, I could keep doing some open source code, but that's never going to reach actual users, you know, the sources that I'm trying to help, or even just citizen journalists who are trying to report what's happening in their neighborhood, it's never going to reach them. Or what needs to happen is this needs to be packaged as a product that people can use. And that's sort of been the impetus for me to explore entrepreneurship, which eventually led to Shravin and Open Origins.

14:46
Amardeep Parmar:
Hello. Hello. Quick interruption to let you know about BAE HQ. We're the community for high growth British Asian entrepreneurs, operators and investors and you can join completely free at thebaehq.com/join. There you get our CEO pillars, sets, content and events and opportunities direct to your inbox every week so you can get involved and it can help you to further your business and your career. We also have a free startup course called BAE Startup Foundations where if you think about starting a business someday or at the early stages, it gives you all that information to help you hit the ground running and to thrive in this new world. Back to the show. So obviously it's great help time. So I haven't met the beginning yet, but you've got to move from academia and the goal setting there to a completely different goal setting in entrepreneurship.

15:35
Amardeep Parmar:
How did you find out all the.

15:36
Amardeep Parmar:
Adjustments you had to make in terms of the way you worked with where you were coming about?

15:39
Mansoor Ahmed:
Yeah, I don't know if I've still sort of caught up to all the dummies that I need to. I think I'm still an academic at heart. Deep down I still want to just make more things, which is a dangerous impulse to you're running a startup. I think it's been a lot of lessons learned the hard way. There aren't enough books you can read to note this. I obviously had a lot of help, not just from Shravin, but from a bunch of entrepreneur networks and founder friends who've given me good advice. But I have made a lot of mistakes and I'm sure I'll continue to make a lot of mistakes. I just hope that none of them kill that company.

16:29
Amardeep Parmar:
So let's say someone this thing is like doing a PhD right now or they're in the academic world, what advice would you give them before they move across?

16:38
Mansoor Ahmed:
Oh, that's a good question. I would say first of all, if you're doing a PhD on something, you've probably already spent three to four years on this topic. And I would advise them to introspect very carefully. Think are you happy to spend another 10 years working on the same thing? Because I think PhDs are explorers at heart. We want to dig up new things. And for me, one of the Frustrations I have with entrepreneurship is that you kind of do have to stick to it. We pivot, but it's not the kind of pivot that you would do in academia where you would just literally go and tackle a different field of literature. The pivots are minor by comparison. You don't get to learn as many new things from a purely technical point of view, let's say.

17:22
Mansoor Ahmed:
So my advice would be, are you happy to stick to this thing for the next 10 years? And also, are you happy with being distributed person in a room? Because when you're a PhD or a postdoc, you're kind of used to being an expert in a field. People come to you for your expertise. When you become a startup founder, you're probably the stupidest marketer in your company, stupidest ops person in your dla. You need to have a lot of humility to just be like, okay, yeah, I don't know anything about what you know. Please teach me. Like, I'm five and I have to catch up as quickly as I possibly can. And then there's the third thing, which is, I think I didn't really realize the weight of responsibility.

18:06
Mansoor Ahmed:
And like, when you're a PhD student, you've basically the only person whose life you can mess up is yours. The worst thing that can happen to you is you don't get your PhD. And you could. That's fine. You'll still probably get a good job industry if you're doing it stem. When you're running a startup, other people's ideas depend on you. And that really gave me a lot of sleepless dice for the first couple of years at least.

18:29
Amardeep Parmar:
Yeah. And like I just said to you, said that very humbly, that you're a stupidest person in terms of marketing smud things. But obviously having that background from Cambridge and the PhD work you're doing there, it meant that you're on the forefront of that technology.

18:41
Mansoor Ahmed:
Yes.

18:41
Amardeep Parmar:
And as, I guess now, as time has gone on, you've got to keep that forefront.

18:45
Mansoor Ahmed:
Yes.

18:46
Amardeep Parmar:
And how do you manage to do that?

18:47
Mansoor Ahmed:
I think that is one of the most difficult things and I really do struggle with that. So there are times where I feel like, oh my goodness, this industry is moving so fast and it really has gotten supercharged in the last two years. So I, the way that I try to do this is I guard my one, not one, my solo time, very carefully. So I'll like, literally schedule time in my calendar, I'll switch off my phone no one can contact me at this time. And I know that I will never be at the same level I was when I was doing my PhD. I just don't have the time to actually write an academic paper that pushes human knowledge forward. Like that is not my place anymore.

19:27
Mansoor Ahmed:
But what I can do is read all of the research that's out there and know where things are going. And I do see that as like a foundational part of my role as founder. I am a technical founder and my company only survives if as long as we are on the cutting edge of technology.

19:41
Amardeep Parmar:
And I said that about the commercial side too. So you've got to one, keep the technology at the top. Also make sure that people know that your technology at the top, so then they continue to buy and use it.

19:51
Mansoor Ahmed:
Yes.

19:51
Amardeep Parmar:
And obviously, as you grow that technology, what are some examples, if you can give to them, of where people are using this right now and it's really made a difference?

19:59
Mansoor Ahmed:
Yeah, I can talk about this big deal that we signed with idn. So IDN is a media production house. They make the videos behind channel 4, channel 5 and ITB. I think they're the second largest news video archive in the UK. And what we did is went through about 70 years of their archival footage. We did an audit of that entire archive. We proved that entire archive is free of synthetic AI generated content. And we put all of that onto a blockchain so that in perpetuity. Now we'll be sure that anything that was sourced by it and is real is not synthetic, has not been modified since then. Similarly, we worked with AMH Media. They're the largest news organization in Zimbabwe. And Zimbabwe, they were having national elections about two years ago and they had a very interesting problem. Zimbabwe is a massive country.

20:53
Mansoor Ahmed:
Elections there can get quite chaotic. It's not necessarily safe to be a very easily identifiable journalist in a remote hidden and Zimbabwe. But you still want to make sure that nothing unethical is happening at election rallies or at election voting booths. So what we did is instead of having them send our photojournalists to all these remote areas, we gave them our secure sourcing tool, which is our camera application that allows you to prove your photos and videos for free. And we had them list that on their website. So whenever anyone went to their news website, they'll see a link saying, if you want to submit stories, please use this link. And were blown away by the reception.

21:40
Mansoor Ahmed:
Throughout that election cycle, we captured hundreds and thousands of photos and videos of election rallies or voting booths and all of that content from all these remote parts in the. It could be used by the editors directly without having to spend any time trying to vet whether this is real or not, because our technology was proving it. I'm really proud of the fact that we helped make that election so much more secure.

22:00
Amardeep Parmar:
And it's obviously a very topical thing right now.

22:03
Amardeep Parmar:
Right.

22:03
Amardeep Parmar:
All the elections going to go on and various things from the various governments in the world about all this information. And as Reggie knows, this can be such a great issue in the future. And again, like how. How you've been able to scale that. Your operations in the team to be able to keep. Yeah, that game too, because there's so much we've done so far, but there's so much more. It's going to be coming up with new problems. Right.

22:25
Mansoor Ahmed:
Yeah. I think foundationally the way we've managed to do it is that we have a really strong why. And like, I think the problem is so obvious. It's like, oh, if you don't do this, we can't trust anything online. We're going to have made these filter bubbles worse. An already divided society can't have any cross communication anymore if you can't agree on what are the facts. So we have a very strong why. The other thing that we've done, I think very well is we've engineered the technology from day one to be scalable to the entire Internet. We weren't out there trying to make just an insurance app, like just something that helps your insurance claims go through. We weren't there making just a citizen job from day one. We architected this technology to be used for anything and anywhere.

23:15
Mansoor Ahmed:
So that really has helped us a lot. And then on the sort of business side of things, I mean, that's. There is no formula for it, except that you have to put in a lot of hours and you need to talk to the stakeholders that are really feeling the way. Right. So I am speaking to journalists every week. I started speaking to insurance people regularly because they're now seeing an increasing amount of fraudulent claims because of synthetic images. So you've got to be in touch with the people actually facing the problem so that you can evolve the product to meet their needs.

23:49
Amardeep Parmar:
And let's say. So what we did yesterday is we had the first person who came on podcast before.

23:54
Mansoor Ahmed:
Yeah.

23:55
Amardeep Parmar:
And then 200 episodes later, we put them on again. So let's say you're sitting in two years time different. So it's a data.

24:00
Mansoor Ahmed:
Yeah.

24:00
Amardeep Parmar:
What would you love to say you've been able to do it.

24:02
Mansoor Ahmed:
That's a good question. I would love for B2B were to say that we've made photos and videos trustworthy again. And like I want that to be table sticks. You know, the same way that if you type facebook.com in your browser, you know you're just going to Facebook. You don't even think about all the security that's gone in to make sure that facebook.com means facebook. I want that to be just as Table 6 for photos and videos. Right. Like there needs to be no gatekeeping. There needs to be. You don't need to sign a license with any company. You just trust photos and videos because we've built technology that helps you do that.

24:37
Amardeep Parmar:
And before talking as well about how the team has grown quite significant the last few years and where do you see that side of things? Well, in terms of geographic expansion and where the focus of your work is going to be.

24:48
Mansoor Ahmed:
Yeah. So I think for the foreseeable future at least, we're going to keep doing our engineering here in the uk. I think we've got a great time. Good here. My connections with Cambridge obviously help. We've got researchers coming in from there. On the business side of things, we are seeing a lot more interest from the States. It is obviously a bigger market. So I think we'll naturally have to gravitate a little bit more towards that side. We'll hire more on the business side in the States. But yeah, engineering, primarily UK based. I love my Cambridge connection. I want to keep that going.

25:19
Amardeep Parmar:
And do you go for about three years now, right?

25:21
Mansoor Ahmed:
Yes.

25:22
Amardeep Parmar:
And if you were to like right at the beginning of this, what could you tell yourself three years ago? If I wish if you could shortcut some of this hard lesson, that pain you had?

25:31
Mansoor Ahmed:
Yeah.

25:31
Amardeep Parmar:
What would you have told yourself or what would you have maybe looked at in a different in life?

25:34
Mansoor Ahmed:
How many things am I allowed to done all day? I thought I made sure. I think the most important thing I'll tell myself is around hiring. I think I. I had the wrong lens when I started doing a lot of hiring. Initially, I think I was much more focused on skills, on CB or brands, on people's resumes than productivity. I think the more time I spent running the company, the more I've realized that the thing I appreciate the most about my employees, my co workers, is people who just take ownership of the problem and we just go ahead and solve it. They might solve it wrong and that's fine. I much prefer a wrong solution than no solution and low progress. So these days when I'm hiring, some primary thing I'm focused on is, can I trust this person to just keep going?

26:22
Mansoor Ahmed:
Like, do I do handhold them? If I do, then I don't want them on the deep. So that's probably the primary advice. The second one, Meditate more. Yeah. Spend more time doing journaling. Because it is so easy to just get caught up in the frantic nature of our work and have that frantic chaos get into your mind. I'll stick with those two for now. Yeah. Yeah.

26:48
Amardeep Parmar:
I guess with that second part, how. How do you keep the boundaries in place? Because AI, even as somebody who's not in an AI company, is seeing you're just constantly updated, can feel quite evolving to so many people.

26:58
Mansoor Ahmed:
Right.

26:59
Amardeep Parmar:
How do you stay on top of the curb? How do you. If someone was gonna wins that AI or that AI, should I talk to them? Should I take that call? Yeah. How do you transparitize that? Of what are you going to pay attention to? What action it is not for you right now?

27:11
Mansoor Ahmed:
I'm quite lucky in the sense that I genuinely find this fun. I find it fun to read AI papers. I find it fun to experiment with new technologies. So that doesn't help me with having boundaries, but it does help me do a lot. Right. So I would just do a lot. I'll just read a lot. I'll just try out a lot of new technologies. so that's number one. But the second thing is when I'm trying to do research specifically for, let's say, a problem statement, as I've been trying to figure out, I try to go old school. Like, I love pen and paper. I'll print out things and I'll read them. I collect typewriters. So I literally type out things on a typewriter. I'll write things. I feel like going analog really helps crystallize your thoughts.

28:01
Mansoor Ahmed:
I think we're losing that skill to some degree. Just being able to stop the ingest, take a breath and crystallize your thoughts on paper is like, for me, it's a game changer. I don't think I can do any of my work if I didn't have that habit.

28:16
Amardeep Parmar:
So with me, how I actually started all this up was as a writer. And it's exactly said sometimes we think how much people are going to view and all that stuff. And it's. It's really about can you get clarity of thought. Because when you write something down, you have to think so much more, whereas if it's in your head, it can swirl around and get confusing. But you say if you're putting something down on paper, it becomes very obvious to you. Does this make any sense or not?

28:39
Mansoor Ahmed:
Yeah.

28:40
Amardeep Parmar:
And is there anything that maybe you think you had. So obviously you had said you put it to this track several years ago in terms of the situation.

28:46
Mansoor Ahmed:
Yes.

28:47
Amardeep Parmar:
Is there anything that's made the surprise you has been maybe different than you had said you mentioned like small pivots.

28:52
Mansoor Ahmed:
Yeah.

28:52
Amardeep Parmar:
As we had Any small pivots you think were particularly interesting?

28:56
Mansoor Ahmed:
I think I underestimated the pace of development. Right. Like if I remember correctly in my original paper from 2018, 2019, I predicted that this tipping point of photos being AI generated photos being photorealistic would happen in 10 to 15 years that have terrified. So I underestimated the baseline development which also meant that we had to sort of quicken our development cycle quite a bit. My original plan was to do R and D for a much longer time. So we hadn't caged in that we had to strip down some of the features, strip down some of the devices we supported to make up for the fact that the world was a much more chaotic place than I had anticipated.

29:38
Mansoor Ahmed:
I think the other thing that I kind of guessed at, but didn't realize how effectively it would be done is how quickly the sort of massive tech giants would come in and try to monopolize the space. And to me that is a scary scenario. I think we already have. We already live in a world where some people have a stronger voice than others. Like if you're in Britain, you probably have a stronger voice than someone in Zimbabwe, as I've seen working with the people in Zimbabwe. And my fear is that because photos and media can't be trusted anymore, we're going to reach a point where large tech giants will say, okay, we'll trust photos and videos that have been created using our tools, we'll trust photos that have been uploaded to our networks.

30:25
Mansoor Ahmed:
And if you have a contract with us, let's say you have a contract with Adobe and you use our authorized edition of Photoshop, then we'll trust your photos, it will show a tick mark on it and everything else gets blocked. And I fear that will make the voiceless even more voiceless. And that has shaped quite a lot of our society going forward. I'm like, that is the worst case scenario to me. I want to do everything in my power to make sure that doesn't happen. Which is why we've embraced open sourcing so heavily. So we open source most of our Tooling, we really re encourage people to go and judge the dueling on its own basis. We don't want people to trust the photo from Open Origins just because it says it's from Open Origins.

31:07
Mansoor Ahmed:
We want them to trust it because the mathematics is good. And to me, mathematics is a great leveler. That's how we get past these monopolies. That's how we get past the skatekeeper.

31:16
Amardeep Parmar:
You mentioned about open sourcing, and I think many people early in their careers or when they start their journey, they really worry about open sourcing, about is that the right decision or not? Can you talk, obviously from the mission perspective of why you've done it, but it's also how does that make sense in terms of the company strategy and how that means you can grow faster and make it a big difference as well?

31:36
Mansoor Ahmed:
Yeah, from mission perspective, it's completely crystal clear to me. Our goal is to make photos and videos trustworthy in a ubiquitous manner without any barriers to entry. So anyone anywhere should be able to prove photos anywhere should be able to trust photos. Open source is the way to do that. From a business point of view, I think open source keeps you honest. I don't know, like, it forces you to innovate, it forces you to keep a pace of development. Yes, it does mean that you won't be able to do like patent licensing or, you know, create a standard closed ecosystem and does push your business to a different angle than what most VCs would try to push you towards.

32:20
Mansoor Ahmed:
But I think in the long run, that's the route in myth by which you make a better product because you just have so much more, so much, so many more cheques and balances on what you're doing, right? Anyone can go and modify your code, anyone can replicate it and build a better version. So you better be ahead of the curve and build the best version possible. And honestly, as long as you're. You are building the best version possible, having it open source makes these conversations so much easier. I've been in so many client meetings where I'll be having a conversation with the CISO like a Chief Information Security Officer.

32:51
Mansoor Ahmed:
And if our code wasn't open source, that meeting might have turned into two months of deliberation where their team will go into our code base and blah, blah, blah, and we might have lost the sale, right? But because it's open source, all I have to do is like, here's the GitHub link, go wild, and I never have a conversation with them again. We'll just get the Sale. Right. So open sourcing really helps give your clients a lot of peace of mind that nothing shady is happening, that you've done your homework and that this is a quality product.

33:20
Amardeep Parmar:
It's all about fresh fries that they can see for themselves.

33:23
Mansoor Ahmed:
It works.

33:23
Amardeep Parmar:
Yeah. And it's very interesting because before I knew about what you did, it's something which almost worries me. Right. It's like when you're watching things online, it's like wait, is this actually personal or not? And actually there's so many companies now who are encouraging people to do AI generated content and people are like, oh you could just have an AI podcast you need to get the guest on and those kind of things that are happening. But it's that humanity, anyone is lost and where classic one right now you don't have to agree with this, but the Google Gemini when you search something, what's going to be at the top? Often it's not true.

33:59
Mansoor Ahmed:
So bad. That's so bad.

34:01
Amardeep Parmar:
And just like wait more. But if you just read that you don't know any better then you might just think oh okay, that's the truth. And the classic one has Reddit.

34:10
Mansoor Ahmed:
Right.

34:10
Amardeep Parmar:
Is that it gets it from Reddit where somebody's made a joke or someone's just being saying something stupid and it's no use as a source of truth. And it's almost like some of the insert is getting harder to use.

34:20
Mansoor Ahmed:
Yeah.

34:20
Amardeep Parmar:
Because you just don't know what to trust anymore. So yes, you can Google things or yes you can chatgpt it but if.

34:25
Amardeep Parmar:
You don't know this. True.

34:27
Amardeep Parmar:
And the old thing used to be the videos. It's much harder to make, to edit or to make to make fortunate. Whereas now that's not the case anymore.

34:36
Mansoor Ahmed:
I, I, I give it a year before videos are also just as photorealistic as photos already are. Yeah.

34:41
Amardeep Parmar:
Yeah. And obviously that's going to be even with my own peace of mind by knowing these things are out there, it helps so much. And I guess in terms of like with where you build this out, is it you mainly looking at do customers, the end customers need to know about the product much or is it more about other businesses or who do you need to know about this?

35:00
Mansoor Ahmed:
Yeah. So ideally the end user, like a person scrolling a website shouldn't need to know that Open Origins is certified as well. They should just be able to trust that the thing they're looking at is re. Okay, that's just how a healthy Internet looks like. But I also understand that is a long term vision. Like there are many steps leading up to that. So right now we're working with end customers who are already facing the problem in their day to day work. So these are editors for news out editions. They get, they might get footage claiming to be from Gaza. How do they know it's from Gaza? They use our technology. Our technology has a portal where they can go in and validate things.

35:40
Mansoor Ahmed:
This might be insurance companies who are getting more and more fraudulent claims where someone has used Gemini to put a dentist their car and claimed some damage. Right. Which is one of the reasons why premiums hit or not is because there is a massive surge in the amount of fortune NVMs being made. So we work with insurance companies to help make sure that the photos we send of the damage are actually real. And my hope is that as we go from these sort of high risk, high impact industries to lower risk to order backed industries, let's say like dating profiles. How do you make sure that you are getting catfish? Slowly, slowly.

36:20
Mansoor Ahmed:
The expectation will be that unless there is a reason to trust the photos are not going to be trusted to the point that we eventually stop having any untrusted photos on the Internet at all. Yeah.

36:31
Amardeep Parmar:
So we're going to get to quick fire questions now.

36:33
Mansoor Ahmed:
Cool.

36:33
Amardeep Parmar:
But really enjoyed that. So first one is who are free British Asians or Asians of Britain you think do incredible work and you love to shout them out?

36:43
Mansoor Ahmed:
Yeah, so one of them is a good friend of mine, Amar Shah is the founder of Wave, which I think recently raised the biggest VC round in British history. Brilliant guy, not involved with Wave anymore, but he's working on a couple of stealth startups. But I would not spoil which you will hear about pretty soon. The other person I'd like to shout out is another exited founder, Bonnie Chung. She has now started talking a lot more about entrepreneurship. One of the few people on LinkedIn whose posts I follow that are actually to say they're actually formative. So you should follow her. And then the last one is Johan Bosch. He was a fellow PhD student with me at Cambridge. He's got a really interesting way of looking at technology.

37:29
Mansoor Ahmed:
He works on the intersection of policy and technology and he's one of the people who's made me realize how technology is always political. It just always is. If someone says that their technology is a political key to they're unaware of it or they're trying to convince you about something that you wouldn't be convinced of. But. So he's a great person to read if you just want to understand how technology's political underpinnings shape our life.

37:52
Amardeep Parmar:
So obviously I met Amar Shah before.

37:54
Mansoor Ahmed:
Yeah.

37:54
Amardeep Parmar:
And it's incredible how humble he is because Bazooka had no idea he founded way and then when he added building he was like oh Wei crap. Like. And he was like one of the mentors on our program as well. So it just went to help out. It's amazing guy I met a little too Patrick sound direct too.

38:10
Mansoor Ahmed:
Yeah.

38:10
Amardeep Parmar:
And then next one is if people want to find out more about what you're doing and like follow your story, where should they go to?

38:15
Mansoor Ahmed:
openorigins.com is probably the best place to go. Otherwise I am humanity. I was very happy of that one. On all the social media platforms.

38:26
Amardeep Parmar:
And the people could potentially help you or reach out to ecosystems to help out. What could they do?

38:31
Mansoor Ahmed:
If you are any organization that needs to trust the photos and videos they're seeing, whether that be insurance, news, dating social media, reach out to us. We can help you prove your photos.

38:42
Amardeep Parmar:
Thanks so much. Com. It's great to see the work you're doing as well. Any final words?

38:46
Mansoor Ahmed:
No, it's been great. Thanks for having me.

38:49
Amardeep Parmar:
Thank you for watching. Don't forget to subscribe. See you next time. 

‍