# Agnes AI: Cost-Effective LLM Strategy for Emerging Markets

**Podcast:** Tech Lead Journal
**Published:** 2026-02-02

## Transcript

Our cost is like one tenth or one twentieth of that of all the major products out there.
We are cheaper and way faster than most of the big models.
Today's guest is Bruce Yang, CEO and founder of Agnes AI.
He's building Southeast Asia's fastest growing AI platform that's 20 times cheaper than ChatGPT while serving 300,000 daily active users.
There's about 500 million to 700 million users using Gen AI product every day.
If you look at uh look at the products like the Gemini or ChatGPT, that's only about 5% of people are spending money for subscription.
And for the emerging countries like the Southeast Asia, like Latin America, very few people spend any money on these products.
If you look at the population in the Southeast Asia, there are a lot of people speaking minority languages like Bahasa, Thai, Malay.
These languages are not very well served by most of the major large natural models.
Our goal is to create a product which will be everyday AI for everyone.
So what makes Agnes AI a compelling options for people who are used to ChatGPT, Wat, Gemini?
We are very much specialized for the work cards like doing your research, writing a PowerPoint, designing for a picture or generating a video.
All the models are specialized for your daily tasks instead of your daily conversations.
Do you think this is an AI bubble or is this something that can still continue for foreseeable time?
Long term speaking is a big bubble, but most of the models won't really exist after like three to five years.
Hey, quick pause.
My goal with Techly Journal is simple.
Learn from the best in tech so we can all grow together.
If this resonates with you, hit subscribe to follow the channel.
It's the biggest way for you to support the show and help us keep bringing great guests and insights to you.
Thanks for being here and let's get back to it.
Hello everyone, welcome back to another new episode of the Tech Legional podcast.
Today I'm very excited to have someone who is closer to my uh presence, right?
Uh more local than the usual guests.
Uh his name is Bruce Young.
Bruce is actually the CEO and founder of Agnes AI.
So before before this, I haven't heard a lot about Acnes AI, but it's actually one of the you know top most growing uh AI model and AI app at the same time, coming from Singapore.
So it's much more local here uh than you know the ChatGPT, the Gemini, the COD, all the models that we uh hear a lot recently.
So I'm really excited to talk a lot more about this local model.
And yeah, Bruce, welcome to the show.
Thank you very much to have me, uh Harry.
Very very excited to share about Agnes.
So Bruce, in the beginning, I always love to ask my guests, maybe share a little bit more about you by sharing any career turning points that you think we all can learn from that.
Sounds good.
So I I I've been uh um studying in uh Raffles Institution um for my high school and after that I I've been studying in the US at UC Berkeley for undergrad.
And um right after that I've been working in the valley for over four years in uh Microsoft and uh LinkedIn.
And um after that I I've been doing startups in the valley and and decided to come back to Singapore actually right before the COVID.
So during the COVID, I there's actually nothing nothing much I could do.
So one thing which I think is a turning point for my career is I uh started my PhD during the COVID at National University of Singapore.
And I think this is a very good timing for me because um you know uh it's a good time to think about the second curve of my career and it's right before the the start of the Chat GPT.
So during my my study and my research as a PhD, you know, a researcher at um we've been very very well exposed to all the brand new stuff like the agents like large language model and like the the memory agent memory.
And that is the foundation for the starting point of Agnes AI and which I I think is one of the most exciting moments of my life.
So I I've been talking to a lot of people um seeing the opportunity of Agnes and any exciting uh consumer app of uh AI for recent years, if you miss it for once, you missed it for your life so uh I'm definitely getting very much excited about um you know funding this company.
And I think the my my PhD is uh a very important turning point for me.
Yeah.
Yeah, so I think you brought up a very interesting point, right?
If you miss this, you know, moment in time, right?
You kind of like missed um you know, so-called the turning points in the world in in fact, right?
And I mean, as a user, even like this is also like a defining moment for uh a lot of people uh getting used to working with AI, knowing AI capabilities.
I can't say for sure, like uh by being a founder, you know, someone who is building an AI model and AI app at the same time.
I'm sure this is must uh this must be more exciting for you.
Before we continue, I want to tell you about our sponsor, Sweep AI.
I love my JetBrains IDE.
And if you like me, you know the struggle.
When it comes to AI coding assistance, we've had to compromise.
Maybe out tabbing to other tools or settling for autocomplete that just doesn't cut it.
Well, that changes with Sweep AI.
Sweep is built specifically for JetBrains IDE.
And honestly, it's been a game changer for me.
The autocomplete feels instantaneous.
The AI actually understands your code, and there's an agent that genuinely write code faster.
Plus, your code stays secure.
I finally found an AI that works great in my JetBrains IDE.
Normal jumping between multiple tools just to get decent AI assistance.
You can download Swip for free at sweep.dev.
Thousands of developers at companies like RAM, Amplitude, Atlasian, and Cloop are using Swip every day.
See for yourself why they're loving it.
And now let's get back to our episode.
So maybe the first place I would like to ask, because starting from ChatGPT, all this craze about AI, LLM, generative AI, and all that, right?
The last side checked, um, you know, in Hugging Face or maybe some internet um articles out there, there are plenty of models already available in thousands, hundreds of thousands.
Why creating another model with Agnes AI?
Well, there are a lot of models, and this is actually the reason we want to start our own one.
The models have uh a division of two kinds.
One is like the Soda, the closed source models provided by ChatGPT, Athropic and maybe Gemini.
And another set is defined by the open source ones, you know, one of the very famous ones, um Deep Seek and Queen and Lama and a few more.
We've been through a time that you know the soda models are mostly from the closed source one and open source ones are actually lagging very far behind.
But right now, the gap is very, very small.
So uh we think that this is the best time for us to find our way to close the gap between the open source ones and closed source ones.
We have the all all the requirements which which we need, all the ingredients for us to do this job because we have a very strong research team.
My advisor joined us um as part of the funding team of Agnes.
I have another friend, um a Brooklyn alumni who uh went to MIT for his PhD.
He's part of my research team, and we have a lot of users using our product out there, especially in the Southeast Asia.
Uh, we have over close to four million register users uh with something like close 300k DAU by analyst this month.
So we we have a lot of people using our product every day.
And that generates a lot of data, a lot of prompts using these prompts as a you know the seeds to train our post-train our model through a reinforcement learning.
We could potentially also distill it distilled our model from um deploying the the soda models in our production.
So there are multiple ways to close a close gap especially we have our you know product out there users using a product generating all this kind of data and this is our opportunity.
Yeah.
Wow it must be very exciting for you seeing you know like four million registered users and 3000 daily active users that's pretty a lot right so maybe one thing to verify so Agnes AI is an open source model is that correct?
Well now it's not open source but we have um you know published the paper which details about how we train our model uh so it's somewhat open source.
The reason we haven't open source it uh is we're preparing for another open source model which will be uh released very soon.
Some of our our smaller models have already reached soda which is state of art for the range of parameters like 7B, 30B.
But if you want to really make a noise and be the king of um maybe um Southeast Asia we have to train an even bigger model.
Right now we have seen very good results.
We have um for the 30B parameter model we are already reaching a very good results for um SEA or quite a few benchmarks from the Southeast Asia region.
But we want to enhance it slightly more before we uh release it to the public.
Alright thanks for clarifying that so maybe let's go into the the differentiator right so what makes Agnes AI a compelling options for people who are used to you know chat GPT, you know Anthropic, Cloud, Gemini, whatever models that are available out there.
So if you're talking about the model itself we are very very much specialized for the work cards like doing a research writing a PowerPoint or maybe uh something like you know designing for picture or generating a video um you know we have both the large language model to generate the props and the image and video generating models too which is the the DIT diffusion transformation, diffusion transformer model to uh generate the content.
And we also have a model to support our group chat.
So you can think that all the models which we are trained are specialized for um your daily tasks instead of your daily conversations.
So this is one thing which we're differentiated from ChatGBT or Gemini or Anthropic.
Another thing which differentiates us a lot is we are cheaper and way faster than most of the big models because we we are specialized for more um narrow tasks.
We don't need to use very big MOE models.
Smaller models will suffice with some proper routing.
Um by routing, we mean that if you know the what the intention of the user, we would not need to direct the request to the big model all the time.
So for for example, i if you um need to uh work on your PowerPoint, the model specialized for the PowerPoint by writing the HTML um content for the PowerPoint is specially trained, which is unbound with that with uh the uh model with with uh soda models like the Gemini or Chat GPT, but it's much faster and much cheaper.
So this is how we differentiate ourselves from from the other models.
So that's very interesting.
You mentioned it's cheaper, it's faster, right?
So obviously these days we don't actually perceive about uh latency much, but uh uh cheaper is definitely in everyone's mind, right?
So um if you talk about having a model that is right, especially you're talking about the emerging countries.
So one of the observations which we had is you know, only like 0.5% of users are paying for generative AI products.
Um so this is the data from uh Manuel Ventures.
Um there's about 500 million to 700 million users using general AI product every day compared to the Netherlands in the world, which is about 6 billion, it's about 10%.
And out of this 10%, if you look at uh look at the products like the Gemini or ChatGPT, that's only about 5% of people uh spending money for subscription.
And this is mostly centered in the high-paying markets like the US, um, the Europe, Korea, and and Japan.
And for emerging countries like the Southeast Asia, like um Latin America, very few people spend any money on these products.
And if you can really reduce the cost for them, reduce the cost for ourselves.
We are, you know, kind of uh block ourselves from serving more people.
Our goal is to create a product which will be everyday AI for everyone.
Serving the uh emerging countries, be very inclusive.
We want to keep our costs really low, much cheaper than that of ChatGPT, uh Publasty, uh or even that of Deep Seek or Queen so the the the the the way we we solve the problem is um reduce the size of the the the model while achieving achieving the the same kind of standard of the the results.
So uh we we we've been doing extremely well for this.
We our cost for um for each of the tasks is like one-tenth or one-tentieth of that of all the major products out there, and that gives us a huge edge to serve um the unserved population in the world.
Yeah wow very very exciting like when you mentioned one tenth or one twentieth I think uh it's um you know starts uh very compelling right yeah so what one question uh as well because you you're kind of like the answer to deep seek from Southeast Asia right do you also support Southeast Asia local context meaning like languages cultures this kind of stuff yeah of course so this is a very important uh thesis for us because if you look at the population in the Southeast Asia there are a lot of people speaking minority languages like um Bahasa like Thai Malay these languages are not very well served by most of the uh major large language models because uh because of the copper the lack of couples of data from the literature um so what we can do from here is um instead of uh trying to make a product which is doing ever doing well everywhere we want to make a product which specially aligned uh aligned for the local interest especially serving for the uh regional languages.
And the way we we do this is post-training from uh open source models, but with more data, more uh language, more literature, and more, you know, uh data from our own platforms, from all the competitions from Agnes.
That way we can bring more interesting content like the slans, the spoken languages to our platforms for people to understand about the culture, understand about the ethnics.
Um that's something which I think is uh a huge advantage for us.
Yeah, you mentioned about the lack of corpus, right?
So the last I use I'm I'm Indonesian by the way, so I speak Bahasa.
Sometimes I also converse with AI using Bahasa, and it seems to be able to understand.
So you mentioned the lack of corpus here.
Maybe give us some use cases or examples where you know maybe things like Gemini or ChatGPT cannot serve, but with your model, actually it can find better corpus, bet better data.
Because I assume a lot of these models just train using you know internet data, right?
Yeah, just yeah, uh crawl the internet and get get all this data, right?
How how do you train differently basically?
Yeah, that's a very good question.
So I I won't say that um GBT or Gemini are not doing well at all in terms of the minority languages, but because they are using the same model to serve every population, every person in the world among all the you know geographical regions, you you can't really do well everywhere, especially uh if you think that the the major languages are English, Chinese, uh, European languages, like French, German, Japanese, Korean.
It's very difficult for you to serve the major language language well at the same time cater to the minority languages with the same model.
So that that's how we are diff differentiated from them.
We it looks like we're all coming from the same pre-trained model, but we're giving uh a more weight, a bigger weight for the for the minority language.
It doesn't have to be different literature.
It could be the same.
And the one ingredient which we are slightly different is we inject a lot of data from our own platforms.
And there are a lot of people using our product in the region.
60% of all our population are from the Southeast Asia speaking Bahasa, Malay, Thai and all the languages and Filipino.
So we this provide something very meaningful to us because if you only get data from online from literature is mostly formal language and you don't use that a lot in your local um you know everyday communication and we we we have a lot of data from this daily use use cases.
We inject that in our post training and that just helps us to make a better you know stand out and make make a better satisfaction for the local audience with our own models.
Yeah.
So one of the definitely challenges when working with this LLM, you know, generative AI is basically about hallucination or you know, giving the wrong inaccurate information.
So, how does Agnes differ, you know, do you do you face the same challenges or do you do something different to minimize that?
Well the the the entire problem of hallucination has been um dealt much better, especially with the the new models.
We we we we do a lot of post-training from uh the pre-trained models like Queen and Deepseek.
For the newer versions, they are already performing much better, so we take some free ride.
Besides that, we also use a lot of you know uh citations with we're sourcing our um putting support from all the sources for all the trends which we provide in our research and search results.
That give us better um you know fact-checked for us.
Uh besides that, we um another way to solve the problem is using multi-agent systems.
So sometimes one large language model um can be wrong, and um the model itself, as one agent do not really know where the problem is.
But it's very easy to be checked by another agent.
So if you have multiple agents, especially multiple agents uh from different models, looking at the result at the same time, different roles.
One is working on the generation, another one is working on the emulation and verification.
The third one is working on providing the feedback for some uh correction.
All three of them working together solve a lot of hallucination problems.
So we we've been dealing with this kind of situation a lot, um, and we uh enable this trio situation of generation, devaluation, and correction.
So um I think that's a very very good way to solve problem cycleization.
Yeah.
Well, very fascinating using multi-agents approach to actually deal with this.
So uh definitely it's something that we all can see in terms of the results, right?
Whenever we use any AI model.
So maybe let's uh dive deep a little bit more about Agnes AI.
What kind of features are available?
Is you mentioned about you know daily tasks, you know, maybe like generating research or PowerPoints, or do you also do it for I don't know, emails, any productivity things?
Like maybe tell us a little bit more.
What are the some of the core features or the like competitive advantage that Agnes provides?
Yeah, so um uh Agnes is both the name of the model and the name of the product.
In terms of product, we we we think we are we're doing pretty well.
Um the number speaks for itself.
And we have uh both two different platforms, one is on the PC and one's on the mobile.
The um positioning of the two platforms are slightly different.
If you're working on the PC, a lot to do with um the productivity.
So um you can do your deep research on Agnes with the same context, you can uh start a PowerPoint, or you can generate images or videos, and you can inject the video and images into the PowerPoint.
It can also help you to uh work on the Excel.
It reads the Excel sheet very well.
Um a thousand you know, lines of uh uh record can be read within half a minute, and if we understand the content, the relationship, and you can draw the draw the graphs for you within half a minute too.
So it it solves a lot of problems of having lifting, we would call it daily work.
If if you don't have AI supporting you, you might spend hours I I would spend hours like five hours, ten hours working at PowerPoint and you have to start with the research it's like uh paying the ass.
With AI it solved the problem very well and the good thing about Apple is it has a genetic memory.
That means the first day after you use it, it knows where you stop um from yesterday and you will start the task with a context with you which you left off.
And you can also pass the context to your uh colleague uh through a group chat uh from a shared collaboration that means uh you don't have to just drop an artifact like PowerPoint to him and he doesn't know what uh how you come up with this uh this this result AI would help your colleague understand where all these results got coming from and answer some questions for your colleague for them for him or for them to pick up from you.
So this is on the on the PC side.
On the mobile side we are studying more on social entertainment slightly more you know uh less formal we really want our uh mobile experience to be uh more inclusive compared to that of ChatGBT and not only limited to uh uh productivity if you look at the retention numbers usage hours um for different apps um you know the social apps like TikTok are doing way better than that of uh the productivity ads like the chat GPT GBT their you know daily you usage time is like by average like 40 minutes if you start with TikTok or WhatsApp normally hours would be spent so I I I definitely think that when we talk about AI right now are uh are still limiting them themselves into the productivity category and if you look at the entire internet in entire you know usage of the internet right from your wake up until before you you get to sleep you will spend most of our time um working on communication like how we we are doing the community right now uh on the WhatsApp and also you spend a lot of time and your free time to look at uh you know entertainment or contents and things like from Instagram, TikTok, maybe Facebook.
So we think that all this content can be very well supported, assisted by AI.
It's just that nobody have have done that just yet.
And it's very difficult for GBT or for Gemini for any other products to do the same thing because they have already have a very strong use base.
And if they want to make any change, they have to start a new app.
So we are able to integrate all these social features into the productivity category because we're still early to the market and we want to really cater to the young audience um the the ones which we called the AI native population, the ones which you know have their high schools or college uh while during the COVID, stuck in in in in their own room and you know, have to jump on the virtual world right after they get you know unblocked.
Um the it's a year of the the birth of chat GPT.
So they they're very native to AI.
Any product without AI seems odd, seems uh um but not for them.
So I I I I think this is something which um um our product will will will help will get a chance if we are able to serve this AI native generation.
If we get them used to uh using our product instead of uh that of WhatsApp or uh or Snapchat, we get a chance to become the next meta.
That's objective pilot.
Very exciting.
I still cannot um you know wrap my mind around you know productivity and entertainment at the same time because I feel those two are opposing things.
It's like one thing distracting from the others.
But definitely it will be there will be a lot of innovations looking forward for Agnes to do that.
So one thing in particular that you seem to focus a lot is the group chat, right?
Um I rarely use an AI that is within a group chat.
Um so one thing that I know uh it's like in WhatsApp you have this meta AI.
Maybe in Telegram you have bots, right?
But I rarely see an AI that actually participates in a group chat together with you and a few other people, right?
So tell us why this is the next thing that people should uh know about and use uh more.
What kind of things that this AI can do in a group chat?
Well, so uh you know, we are not the first players to uh include group chat in AI.
ChatGPT has just launched their group chat recently and um Snapchat as an incumbent player, they are working with Populacity.
Uh I think they are going to spend something like 500 million to gather services from Popularity to support search in their chats.
So uh I definitely see the trend of uh you know AI and chats working together, and the chat which we which we are we're meaning uh uh a real chat, not a chat with AI.
I mean um like like the chat GPT.
So, you know, AI in the chat can be very uh uh disruptive, I mean very um innovative because during your during a normal chat uh with your friends, there's a lot of uh loss in memory and loss in context.
If you miss the communication in the morning, you will spend a lot of time to look at all the communication in the morning, but it may not really contain a lot of information.
And AI can help you summarize the content and and let you um jump start with a topic.
First of all, AI serves as a common memory for all the people in the group.
Second, you know, sometimes uh people won't really communicate very well.
There's a lot of misunderstanding.
AI sometimes can understand people better.
Uh one sentence may mean two different things to different people, and some of the people are very inverted and other people would be more expressive.
We would hope that AI would fill the gap to make all people very expressive.
If the person can read an expression for for himself, AI will fill the gap.
There are a lot of fun fun things.
One one of the things which we are thinking about is like a game, some of the AI games in the group chat like the the Netflix show of Squid Game, right?
Squid Game, we think about you have to convince AI to not kill you, but kill another player during the during the session and every once well once once for a while um you know half of the people will be built until until you find a winner.
And there might be a some reward like in all our product we're give giving the credits.
So we give like everybody will commit for 10 credits and at the end of the game only the the the the two players or one player will get all the credits.
So this kind of cool things, fun things would only be enabled with AI.
And we we we're also thinking about other games like guessing who is human in in the group chat.
The AI have to act like human, or humans have to act like AI to misguide other players.
So this kind of cool things um is very well enabled because AI has gone into a stage that is very similar to the logic of human when they when they speak.
Almost passing the Turing test.
So a lot of cool things, fun things can happen in a social scenario.
And that's something we want to really want to explore.
Definitely those are fun games to try.
So if people are interested, I guess they can sign up to Agnes AI.
So you mentioned about the a lot of uh context missing in a group chat or misunderstanding happening.
You know, it's trying to also be more inclusive, and especially with the Southeast Asian context, right?
We we are kind of like different for different countries, right?
Different languages, different culture, different, even like the way we speak is different, right?
So I guess it will help a lot if we have this kind of AI to help us to, I don't know, smoothen the conversation, make sure no misunderstanding happening.
But what about the privacy that is ongoing between all those chats, right?
Uh, how does Acnes actually protect privacy and security for these kind of conversations?
Yeah.
Well, so um one thing which we can uh support um compared to all compared to all the other you know application apps is we have our own models and we have all our own you know database.
So that means we are able to um solve the problem from our own our own end instead of playing, you know, um switching between different apps, different models, and that could introduce a lot of complications.
So uh privacy is a huge problem for us.
I mean, uh a very important topic for us.
Uh, we're working with all the you know, GDPR, all the uh the guiding guiding rules for different regions.
If you we talk about um, you know, uh enterprise solutions, we're able to, you know, deploy our service on our clients, you know, their own servers that also reduce any risk of uh information leaking.
So I would say, you know, uh besides all the um all the important measures, we can handle this uh slightly better because we have our own models and uh we have our own database.
And that means we don't have to deal with you know data exchange problems.
Um but anyway, I think AI would definitely help with uh a lot of uh governance in the community in the social media in order to reduce any kind of data leaking.
Yeah.
Yeah.
So you mentioned that uh someone can host acnes model inside their premise.
Is that correct?
That's right.
Okay.
That yeah, definitely that's uh one also cool thing, right?
To protect privacy and security, especially if you deal with, you know, I don't know, like very sensitive data or highly confidential.
Yes, yeah.
You you mentioned that people can get help from Agnes AI to produce PowerPoints and uh documents, research, uh images and all that, right?
So definitely that's uh like a huge help for productivity and all that.
Right.
But the other aspect of that is like some people are concerned about the critical thinking aspect that might be affected because we leverage AI or we just outsource everything to AI, you know, AI will produce everything for you.
So what is your thought about this critical thinking aspect and anything that you think Agnes AI do differently to help users not to, you know, like soften their critical thinking, but actually also uh collaborate together.
Yeah, I I think Jensen Huang has mentioned that in the you know symbiosis of human AI, humans will need to give a lot of inspiration while AI doing the have lif having lifting work.
I don't really think that um AI would serve the purpose of uh um you know introducing all the um inspiration and entire end-to-end work for human because if you look at Sora, it's it's a huge success at the beginning, but nobody is using it right now because everything is AI.
It's there's no real human content in it.
And uh the density of human um UGC user generated content will be very important success for uh community because it's like uh a mutual respect.
If you if you this is content generated by AI, why do I need to read it?
If it's uh content generated by human, there's a lot of compassion in reading the content.
So I uh I I would say that um Agnes serving as a product of uh uh AN native social um social app, you know, integrated with the productivity, we will value a lot into human inspiration, human content.
And we put human in the loop at all the major milestones.
So for example, if you really want to produce a very good PowerPoint, you would do a very thorough research yourself and you would go for multiple rounds of prompting to get the content you needed, and it would really ask Agnes to help you look into the content, uh verify the content, and put up a something like uh page by page uh you know layout, and we enable people to have multiple rounds of communication to edit the content, uh add each of the page page content before uh start the generating of the HTML, which is the end part of the uh of the slides.
And even the first round of slides are generated, you are able to edit very well.
We support, you know, advanced editing function, which is almost like manual editing on uh on PowerPoint, but it's much much more smoother, and you can also have Agnes help you change the page, like change uh the the the logos, um uh change the image, change the the background.
By the end of the process, we would hope that there's very little, you know, uh signals of AI.
Very little, you know, uh uh you you won't really see the the the AI content um within these slides because AI is doing everything which humans wants it to do, instruct AI to do.
So so this is how how how how how our philosophy of the product uh is changing the development of of all these features.
Yeah.
Well I think uh that's very important right we don't want to lose our critical thinking at the same time leveraging on AI but definitely AI helps to boost a lot of productivity right and in in one aspect it actually can boost our critical thinking as well if let's say we provide a lot of inspirations just like what you said right so if we provide a lot of in inspirations yeah yeah so this is nothing to do with Agnes but I want to also want to share a story about myself while I I I I I was doing my PhD uh one of the classes which I took the robotic class I I was a plus student uh a plus like top student in in the class and the reason I was able to to do that with well I I I started my PhD uh not right after my undergrad I'm like 10 years older than all my peer classmates but I'm still like the number one in class which is very surprising and the the the secret which I I I I share to all the all my friends when they ask about uh the reason is I use a lot of AI.
There's a lot of, you know, uh to use like the noble AIM, um, Chat GBT to help with, you know, curating the content of my paper and doing the research, doing uh doing a survey.
And this is all you know very well supported by our professor.
He starting from the first day he he's he's very open-minded.
He said that you should use the AI tools.
Just let let us know.
If you can generate a very good results uh with AI tools, you should share that in the class because everybody should know about using AI tools to boost that their creativity, boost their the per productivity.
But but one of the things which we we don't really want to see with this kind of A you know facilitation is if AI is diluting the entire community, that could be a huge problem, like right now the Surah app.
Um Sura right now, it's k kind of chasing people away because all the AI content is all looks the same.
It's there's a homogeneity, homogeneity, which means all the images, all the videos are looking uh all look the same, very AI.
So I we we don't want this to be the same kind of situation when we uh involve of human productivity.
Yeah.
For example, if you don't put any inspiration, if you work on a paper, it will all look the same.
And if you work on a survey, you will give a very very general responses instead of your insightful, you know, critics.
So that's my comments um about this topic.
Yeah, thank you for sharing your personal journey with AI during your PhD study, right?
So which brings me uh a very interesting topic, right?
Because I think there is a dilemma in, you know, leveraging AI for students, right?
Especially as early as maybe primary school uh and secondary school and things like that.
What is your view about this?
Are you more pros towards leveraging AI even early in the education?
Or do you have a different thought?
So my daughter she uh is exposed to AI because I'm uh AI um entrepreneur but I also give her a lot of uh heavy lifting uh schoolwork like the mass olympia he just got in the gifted program uh in Singapore which is like top one percent uh of the batch and she uh at age of nine she's she has a vocabulary of um 10,000 which is uh quite a lot she does all the math Olympia um she was like top 40 national level so I I I I definitely see the the advantage and the benefit of training the intellectual without the assistance of AI at early early stage because if you start to rely on AI a lot of things can be very easy but you you will just lose the the the kind of uh habit or lose uh critical thinking or you you might might not even know how to prompt.
That's a problem.
If you know the know the end results, you know, human at least have to be the evaluator at at the end.
You will know what kind of result would be a high quality result.
Uh and the only way to to appreciate to have the kind of taste is going through the process manually without dependence on AI.
Even though I I know the benefit of I know the disruption of AICOS, I still instruct my kids to work really hard, to be a critical thinker, to understand about you know rudimentary logic and high quality critical thinking because this will be very very important prerequisite for you to to be a AI adopter for you to really highly uh utilizing AI in your future work.
That's my belief.
Yeah.
Very interesting takes uh coming from uh you know an AI founder, right?
AI entrepreneurs, so definitely maybe in early age you should not leverage AI too much, right?
Right.
And what you mentioned is uh definitely very true, right?
So if if you don't know how to prompt if you don't know what good results, good quality look like, I think that's uh definitely a big problem, right?
Even though you have AI, but if you cannot produce high quality results, that's also kind of like defeating the purpose.
That's right.
So I think thanks for sharing your take.
Yeah.
So how about coding?
Because a lot of uh use case these days, um, you know, using AI is for coding, right?
So you can generate code, front end code, back end code, whatever code, right?
So is this something that Acnes also can support?
We could support coding, but we do not support it in a formal way.
That means if you ask Agnes to code for you, maybe yeah give you some uh code snippet in response, you will do it well, just like how the ChatGBT is doing.
But we do not uh provide a function for you to uh code with within IDE just like a cursor.
There are reasons for that I can share in a bit um coding is very difficult task.
It's one of the tasks which uh you see a huge difference between the levels of um smartness of AI.
That's why you know Athropic is doing extreme extremely well in coding despite all the other you know AI models are catching up on everything else.
So we don't think that it's a very good category of um topics or area that we should focus because that means we need to maintain at the top level of the model all the time.
And that could be very competitive if we talk about um product point view.
We want to you know for some a short period of time we want to be a soda best in the world but it's very difficult to maintain at the top level all the time.
And if we build a product which um sits on the dependency of a soda model, there are very few, you know, uh choices.
And we have to uh be very uh cautious about making that of uh decision.
So right now, you know, a lot of other functions like research, uh PowerPoint, uh maybe uh um, you know, image generating or video generating, or maybe uh group chat.
Even if our model are not soda, you might not see the difference.
Number one.
Number two, uh, you won't really appreciate if we change to a better model because the network effect, the memory, the habits, the habitual use use cases would help you to uh dig on our platform.
This is something which we um we definitely think like will make a better fit for us at this point.
Yeah.
Yeah.
So yeah, definitely um very difficult, right, to compete in uh coding.
Um you we have so many benchmarks, right?
And I think over the time we will see uh eventual winners, but it seems like these days there are a few options only.
Yeah.
Um definitely it's one one as you mentioned about enterprise that can host AI models.
Right.
Uh for consumers, right?
There are plenty of options.
You can subscribe to any models you like.
You can even uh use an open source model deployed on your laptop and all that.
And there are again, like there are hundreds of thousands of models available out there.
So, what do you think the future looks like?
Because as an end user, I might be overwhelmed with a lot of uh options.
I might not know what's best for what use case.
And ma many of these things are over the internet, right?
So that means I will send a lot of my data, and especially if you use it for you know, like private conversations, you know, confidential informations, there will be a lot of leakages happening.
So, what do you see the future of this?
Do you think everyone in the end will have their own model host hosted by themselves?
Um, I don't know about that, but you know, uh most of the models won't won't really exist after like three to five years because the only reason they are existing right now is there's not a huge dominance and we are sitting on very shaky land there.
Everything is disrupting, just like how the Chat GPT is is focusing a lot on all different kind of things and we just see that um the model itself is not a huge barrier for a product to grow um down the road because you know, one year ago, ChatGPT is still like number one in the world.
Within the last one year, there are quite a few new players which is competing with Chat B like Anthropic, Gemini, and X AI.
Within the um you know half a year from now, GPT might not even be the best model in the world right now after the uh release of Gemini 3, everybody think that Gemini is doing better.
So a good product definitely has to be uh you know, a merge of both a good model and a good traffic should used product.
And that that means that it's not about the model which makes people you know doing the doing the choice and making the choice.
It's the product which providing all the features, utilizing all the models which get people stick on it.
So that that that's why I I I I think you know ChatGPT is definitely doing well.
There are quite a few other players like um maybe uh Agnes, us, uh maybe a few more.
But most people won't be really uh um knowing what what what's the model behind it, because the product, the features, the use cases will definitely be more prominent for people to understand to resonate.
Yeah.
So uh um in terms of whether people would have a product built by their own model down the road in the future, it's a potential I don't know.
I I I I think it it's pretty wild imagination.
It's something very different from what people are using the apps right now, but it's about a lot of new things.
I won't say it definitely won't happen.
But as as mentioned, um I I don't think that there are going to be a lot of models down the road.
There are very few models supporting very few products which will dominate in the world.
And all the other smaller models will be on the long tail.
Very few people use it.
Um they might still exist, but it won't make a huge impact.
Yeah, so I think everyone now is competing, you know, like all the big boys, right?
Trying to be the best model available out there.
And I I I personally see also like these AI models seemingly doesn't have a lot of mode, right?
So like what you mentioned, right?
It's not the model itself that makes the whole difference, right?
It's the application of it, the integration with the app, product, uh the use cases.
And I can like for me myself, right?
Even though I sometimes see those extremists trying to like compare one model versus the others, I personally find it less uh compelling for me because yeah, I just use auto in some of the products, you know, they will they'll choose the best model for me and I'll just leverage on the output instead of you know tweaking which model I should choose based on you know what temperature, what parameters and all that.
So I think what you mentioned is definitely very, very uh true.
So I I am I'm a little bit interested in when you say that uh a lot of models will die, right?
So I know some people these days are talking about AI bubble.
So there are a lot of investments, you know, high valuations.
Do you think this is an AI bubble or this is something uh that can still continue uh for foreseeable time?
So it's defin definitely not a two-nip bubble, which there's no substance at all.
AI is something which is very well supported by real economic impact.
And if you look at the the growth of Chat GBT, it becomes so widespread, not because it's a bubble, but because it does really make some uh you know use cases.
I'm using ChatGBT to um save a lot of time when I trying to write a write a paper, um, trying to write a you know email.
It give me a lot of very very good support, same from our product at NSCA.
So uh with AI support, a lot of um people are empowered with making a stronger impact.
And this is something which I I definitely see that um is a huge, very important growth for the economy.
Um that's why I I I don't think it's it um long-term speaking is is a it's a big bubble, but within a short period of time, if you look at um, you know, the transitioning right now, a lot of money is betting on the models, and just as we mentioned that the application definitely will win the game down the road.
Um so for the companies which was a lot of money, also wasting a lot of money making the models, uh, but not really being well adopted by a lot of users.
They might be um stand at the center of the bubble.
Um, but the the wealth, the the value of AI will definitely grow all together.
Just the transitioning from the model companies, just we will see the transitioning from the model companies to the application companies.
And the application will be very uh universal, not only limited to the large language model, but also um like the robotics, like the enterprise, the automations.
Um so that that's that's my belief.
Um, so in short, I I don't I don't think that AI itself is a bubble, but micro level speaking, maybe some of the model companies because raise so much money at the beginning, but because of completion, because of the application traffic um problems, they might be a bubble.
Interesting thing, definitely, right?
And and I recently read this um Southeast Asia Economy Report.
So, you know, every year, you know, they came up with this report, right?
And one aspect that is very interesting in the report is definitely the introduction of AI into some of these applications within the region, right?
So definitely it's one major variable uh in the economic growth for Southeast Asia at least.
And I I believe it is happening also in the whole world as well, right?
In big bigger parts of the world.
So definitely AI seems to disrupt every industry.
I would say maybe, maybe if not most uh industries, right?
So I think definitely is something that we can foresee for quite some time.
Uh AI producing a lot more values integrated with uh so many other products.
So, which brings me to the next question for us individuals, right?
I know that these days many people are already talking about AI, but I can still see some people are not into using AI yet.
So maybe they are scared about hallucinations, they are scared about AI governance, security, and all that.
So, and especially for Southeast Asia, I'm sure there are a lot of other people within this region that are not tech savvy, right?
So they they don't know what is AI, they don't know what it's capable of.
What is your advice for us to be more ready, uh more integrated with AI in our day-to-day life?
Well, this this is something which I uh Agnes is trying to do.
One of the problems with people not using AI is it's only limited to productivity and is there's a lot of you know um cool features, which is only provided by the high high um pay tier.
And Agnes trying to solve the problem by provide a lot of cool features and pay ts of other products for free.
So for people of the region who wants to understand about AI and try AI, you should try it with Agnes today.
Um I think we are we're wealth of the products which provide all the features from ChatGBT from ProPlasti from from Gemini uh to one product.
And we provide a lot of free quota.
Almost you don't have to spend money to enjoy your use of the deep research, PowerPoint, email generation, and video generation a lot.
Besides Agnes, uh you can definitely also try all the other products like perplasticity and GPT.
Um try to, you know, integrate your life, um, your work, your studies, um, your communication within you know AI scenario.
Sometimes it looks uh uh a little bit tack savvy.
I mean, but I I I would definitely say that it's very easy to enter the door because the communication with AI is not using a um it's like a GUI, it's like command, it's it's like natural language.
For a lot of uh other tools like Excel or Adobe Photoshop is very difficult to understand all the instructions, but AI, you just have to be a natural language speaker.
You just have to know how to prompt.
So it it's not difficult at all.
Yeah.
So I think the the cool thing about AI LLM um definitely is the natural language interface, right?
Which makes it easy for any people, right?
Even like with the local model like Agnes maybe in their language, they can just speak to the AI as natural as speaking to a person.
And yeah hopefully they can get a good results right especially if it's like a common problem, right?
So I'm sure the the answers will be much more accurate and in a good in a good standard.
And yeah hopefully we can see a lot more people within this region uh being AI trained being AI enabled uh and hence uh impacting the economy a lot.
So Bruce I think it's been a great conversation is there anything else that you want to mention about Agnes, maybe cool use cases from you know your journey so far that people have um in terms of using Agnes.
Uh yeah, it could be anything else as well.
Yeah, we have we have received emails from a lot of our users talking about how they use our product.
Like we have received emails from um um teachers from Abu Dhabi who have mentioned that they have transitioned from gamma to us because we're making equally good sites with much more quota at a much cheaper price.
Um they're asking for for more quota.
And we have also received uh you know uh emails from uh um Philippines um with emails and with uh Gov.ph, which is probably from the government sector, asking us to provide more quota for for deep research because they are doing a lot of deep research relying on AI.
So we have seen a lot of people from emerging countries adopting AI from our platforms.
And one of the things which we we definitely observe is they're not afraid of AI.
Um they're willing to try, and they're willing to try with more advanced features.
Um but the problem that uh with all the other products is they they have a barrier of uh um uh paid tier, which if you don't spend money, you only try with uh very basic conversations.
Like if you don't, if you if you're not subscribed to ChatGBT, you won't have deep research, and that deep research is only um 15, a quota is only 15 for the entire month.
For us, our research is we're giving like close to 30 daily research for everybody for free.
And I definitely see that this is um very important um effort for us to get people from emerging countries to adopt to adopt AI.
And uh and we we want to have the AI priority in the world, and the willingness to pay should not be a barrier.
And if people want to pay, um you can we we provide some of the features like I mean some of the some of the options with starter subscription, like $3 a month.
People from emerging countries want to try our product with even high quota.
Even if they don't want to uh try with a high quota, we provide something like one-time um, you know, one one-time credits for them to complete the the slides if they already finish half of the slides, uh want to finish the second half.
But basically the the quota is quite a lot for everyday use usage.
And we also believe the one thing which we we believe is that the traffic itself is a value.
So just like how TikTok is able to gain a lot of use users using their product very quickly every day with a huge amount of hours, uh very high retention.
So one thing which we one of our measures to uh um help you know adopters um to try our advanced features without paying is to bring more other friends or all the family friends to using a product and you know, form a group chat so that they can talk, they can keep it, keep everybody active on the platform because we believe that the high usage, um, the the habits, um, the high traffic would be uh even more valuable than the AR than the paying customers.
So that that that's some uh uh strategy and some philosophy of us building Agnes AI.
That's something I want to share right now.
Yeah with you guys.
Yeah, so out of curiosity, because you seem to be very generous providing in higher quota, not even paying for people to uh leverage some of the cool features.
How how is your business model like?
Or are you focusing a lot more on acquisition at the moment?
And even like you mentioned about the traffic, right?
Are you leveraging on the data that comes in for further training and all that?
So tell us a little bit more about this aspect.
Yeah, um, that's a very good question.
So um do we definitely focus a lot of um user acquisition at this stage because we are we're latecomer.
Well not like one year ago or two years ago, when or three years ago when ChatGPT came up.
So we want to catch up with their traffic.
That's why we do not focus a lot on the AR on the on the payment right now.
But the reason that we're able to do that is we're able to support the high usage with our own models with very low cost.
We're like 20th, one twentieth of that of our competitors, um, other players.
And how we want to make money, um, twofolds.
One is still on subscription, but if this is on the high-ping uh markets like US, um Japan, Korea, um, Europe.
For these regions, we want to enter, but we we will give a very uh very big discount in the um subscription, provide equal or more features, uh better quota, uh, but the cost will be one-third of that of the other players.
The the other fault is on the emerging countries.
We understand that most people are not paying, even not only limited to our products, but also ChatGPT, Gemini, nobody wants to pay, especially if it's uh you know monthly subscription.
One of the reasons is you you everybody is afraid that they're they're missing about I mean they miss the time of canceling uh unsubscribe, but the product is uh is not used anymore.
So we we we try to solve the problem by giving a lot of one-time payment, um, which is very low cost, like one dollar or half a dollar uh for people to complete a task.
If you don't don't really want to spend any any money, any dollar, you can also get people um to our platform by invite more people or you know, be very active on the group chat because the youth acquisition is also cost from outside.
By saving costs, it's making money for us.
And long-term speaking, we we're definitely um you know believing uh thesis of uh traffic, just like how Google, uh Meta, TikTok become so valuable b because everybody is talking about it, everybody believes about the brand.
Because of brand, because of high traffic, they are able to monetize with other kind of methods like uh ads, um like IP, like e-commerce, ChatGB is trying similar things with the apps in the GPT, uh, with instant checkout.
We could potentially do the same thing, but we we'll form our own ecosystems because we have higher traffic on the emerging countries.
But our business model definitely will be very different from all the other players because of our our own our low cost because our uh selection of the market because we uh we are originally from a different place of the world.
Yeah.
Yeah definitely um you know for those listeners who come from Southeast Asia we can be proud of having like a model that comes originally from you know Southeast Asia, Singapore specifically, right?
Because like the world has seen Chat GPT.
The world was disrupted by Deep Seek, you know the open source model from China.
And I hope one day I can see Agnes also making a noise in the world, you know, all these AI models.
And definitely thanks for sharing those story with us today and uh I will try my best to support Agnes AI.
So as we reach the end of our concept conversation I have one tradition that I always ask my guests which is a question I call the three technical leadership wisdom.
So you can think of it just like advice that you want to give to the listeners.
Any kind of wisdom or advice that you want to give the listeners today.
Let me go one by one.
Number one I I think is um focus on one metric at at one time.
I think this is very important to us because a lot of people can can't really stand out at the beginning, and for startups like us, we need to survive.
We have to show one thing very strong before we can prove anything else.
So I I I think we focus on DAU a lot.
That's that makes us stand out.
Um we have like one of the fastest growing uh product in in the in the region.
Um so this is number one, focus on on one metric at one time.
Number two, I I think rely a lot on your team.
I I think we we I rely a lot on my team, my advisor from AOS, my schoolmate um from uh my alumni, um Berkeley Club um who is also MIT alumni um doing our research because I can only focus on one thing, and if I focus on too many things as a team, I can't do well for anything.
So uh uh rely a lot on your team, find a good team so you can do something big together.
Um the third part about technical, I I think let me think about that.
Um I would say learn from your mistakes.
You know, in my process of building my startup, I keep reflecting on myself and keep um asking whether I I could uh whether I I did wrong and whether I could do better.
And this become uh um uh a ritual for me to um keep changing myself, um evolving myself.
It's something you know, um which I think is the beauty of doing a startup because if you don't have the kind of pressure, if you don't have the the kind of ambition, people do not tend to evolve or change by themselves.
By doing a startup with so much opportunity and such high pressure of handling a lot of things um at one time, I just become you know a better person myself.
I just be more uh disciplined.
I I just um try to find my own problems, problems and trying to um better myself.
That's one thing I I I I definitely um feel very valuable to everybody doing tech technical, especially in doing startups.
Yeah, keep evolving.
Yeah.
Well, thank you for such a beauty wisdom.
So Bruce, if people love this conversation, they want to connect with you, they want to find out, reach out to you online.
Is there a place where they can find you?
Uh of course I can give my my my email, which is Bruce at Spiens-ai.io.
I I can send out to you.
Yeah, you can maybe put down the comments.
Yeah, sure.
Or you can or they can also find me on LinkedIn with Bruce Bruce Yup.
Yeah, I'm gonna say yeah.
Okay.
Cool.
So thank you so much for sharing today.
So I wish uh Agnes AI uh, you know, a much greater success, a more prominent um, you know, usage in the world, right?
Not just Southeast Asia, definitely.
And I think we all can root uh for the success of Agnes AI simply because yeah, it's one of the local models that came from Southeast Asia.
So thanks again, Bruce, for sharing today.
Thank you very much, Harry.
It's a very pleasure talking to you.