Join us on this call led by ⁠Insignia Ventures Partners⁠ principal ⁠Yongcheng Ong⁠, tapping into the minds of innovative entrepreneurs who are at the forefront of building the AI transformation infrastructure for enterprise, ⁠Appier⁠ CEO and co-founder ⁠Chih Han Yu⁠, ⁠WIZ.AI⁠ CEO and co-founder ⁠Jennifer Zhang⁠, and ⁠Bluesheets⁠ CEO and co-founder ⁠Christian Schneider⁠.

Editor’s Note: This episode is a edited recording of a panel at the 6th Annual Summit of Insignia Ventures Partners

Summary: Three key elements for AI transformation to work for enterprise

1. AI transformation is really at its core, data transformation.  

This means that the future of AI transformation is not so much providing models but enabling companies to build their own with internal data. As Bluesheets’ Christian puts it: 

“We’re really more concerned about how the enterprise prepares their data for AI rather than how they get more of it. Subsidizing data from the outside world will probably work to a certain extent, but if it’s really meaningful is the other question. 

If it really comes down to ROI and measurable ROI, whether that is in terms of man-hours or whether that’s in terms of efficiency, then usually we would default to internal data to be used for training purposes rather than from the outside.”

2. Business readiness and data readiness smooth over adoption, but expanding from a more nascent market can be valuable long-term.

As Chih Han puts it, “Nowadays in Appier, we really look into two most important dimensions when we decide which market and which application to enter. One is data readiness, whether the market has sufficient data. Second is business readiness, whether businesses are ready or educated with the basic concept of AI. We found that with both factors being very mature, it’s the best sweet spot for us to enter.” 

It helps at the same time to have roots in Southeast Asia, where the diversity and nascency of needs for AI transformation mature AI enablers with their onboarding and customer success, making it relatively easier to enter more mature markets like the US or even similarly emerging market regions like Latin America. 

3. Clear land and expand strategy is necessary for AI transformation in “early adopter” markets. 

It is more practical to introduce AI transformation in departments where wins can be quickly achieved to strengthen management confidence, before replicating it across teams. 

As WIZ.AI’s Jennifer shares from her experiences selling AI solutions from voice AI talkbots to enterprise-specific LLMs, “You are selling the whole [solution], but you need to have a very clear land and expand strategy for each enterprise. That’s something you should do, especially in Southeast Asia, as the region is still in the early adoption stage for enterprise.”

Timestamps and Highlights

(02:12) Introductions

“…in 2012, I then started the company in the period when AI was relatively unknown. So I think now we were really experiencing a period when AI started as an academic subject, then saw really early adoption. I would call 2015 to 2020 the early adopter period. Today is a mass adoption period where everybody has already experienced the power of AI, and a lot of applications are being developed.” – Chih Han Yu

“This year, we opened a new chapter and deployed all these generative AI models, specifically a lot of larger models, into enterprise applications. On one side, we enhanced the customer experience by turning call centers into multi-language multimodal models. We also use the same models to enhance not only the customer side but also in-house employee working experiences like helpdesk, HR, and IT.” – Jennifer Zhang

“We are using and building AI models for automation in enterprises. We specifically focus on automation of unstructured data, which means wherever in the enterprise there’s unstructured data coming towards you in the form of applications or internal procedures, we help facilitate and structure that data. Eventually, we make it available for your internal systems much faster.” – Christian Schneider

(07:46) What Does AI Transformation Really Mean?

“Starting in 2022, the generative AI era became even more interesting. For marketing, in addition to precision execution, we began to see creativity emerging…Currently, only 3 percent of marketing technology involves AI, but we anticipate exponential growth in this area over the next nine years. This is an exciting period where we’ll see growth not only in precision marketing but also in creative marketing.” – Chih Han Yu

“People have experienced AI personally, but adopting it in an enterprise setting is different. There’s a journey from consumer apps to enterprise solutions, and we need to ensure reliability. We should focus on existing use cases and solve pain points from the past. Don’t just sell the future; solve the problems from the past.” – Jennifer Zhang

“Many companies will build their own models, but they will face hurdles and may not venture into all categories within their business. It’s an exciting time, especially for Southeast Asia, to leapfrog some competitors as well.” – Christian Schneider

(17:53) Leading The Third Phase of AI Adoption

“Customers’ experiences with using AI, especially through GPT-like models, shifted their mindset. They now understand that while computers can make mistakes, they are generally correct and can enhance their capabilities.  This mindset shift unlocked potential in the creative side of marketing technology. Nowadays, in addition to precise decision-making, AI also contributes to the creative side of marketing, as marketing combines both creativity and execution.” – Chih Han Yu

(21:28) Commercializing AI for Enterprise

“I believe it’s not one time use and you expect them to adopt everything. You are selling the whole [solution], but you need to have a very clear land and expand strategy for each enterprise. And you need to really handhold. For most customers, sometimes they want to jump, but they don’t know where the ladder is, and you need to offer that. That’s something you should do, especially currently, in Southeast Asia, as the region is still in the early adopter stage for enterprise.” – Jennifer Zhang

(23:56) AI Transformation Means Data Transformation

“We’re really more concerned about how the enterprise prepares their data for AI rather than how they get more of it. Because subsidizing data from the outside world will probably work to a certain extent, but if it’s really meaningful is the other question. If it really comes down to ROI and measurable ROI, whether that is in terms of man-hours or whether that’s in terms of efficiency, then usually we would default to internal data to be used for training purposes rather than from the outside.” – Christian Schneider

(27:40) Advantages of Being Global AI Companies with Roots in ASEAN

“Nowadays in Appier, we really look into two most important dimensions when we decide which market and which application to enter. One is data readiness, whether the market has sufficient data. Second is whether the business readiness, whether businesses are ready or educated with the basic concept of AI. We found that with both factors being very mature, it’s the best sweet spot for us to enter.” – Chih Han Yu

“​​The good learning for us is actually because we tapped first into the Southeast Asia market, when we actually used the same customer for piloting similar solutions it landed pretty well in Latin America…What we realized is that even in the Middle East, the contracts are bigger, but there are less challenges when we’re deploying in multiple countries there.”  – Jennifer Zhang

“And I think for any companies looking into coming up with B2B enterprise solutions, there is an array of opportunities now, and I think we’re going to see a lot more level playing fields in segments and industries, companies going after the POCs from different parts of the world, and we don’t really feel a restriction by being a Southeast Asian-based company.” – Christian Schneider

The content of this podcast is for informational purposes only, should not be taken as legal, tax, or business advice or be used to evaluate any investment or security, and is not directed at any investors or potential investors in any ⁠⁠⁠⁠⁠⁠Insignia Ventures⁠⁠⁠⁠⁠⁠ fund.

Transcript

Note: Transcript has been edited for concision and clarity.

Introductions

Yongcheng Ong: Let us welcome my esteemed panelists for this afternoon. Joining us will be Chih-Han, co-founder and CEO of Appier, Jennifer, co-founder and CEO of WIZ.AI, and Christian, co-founder and CEO of Bluesheets. 

Alright, to kick things off, we can have each of the panelists give a quick introduction and maybe share with the audience a little bit about what your business is doing.

Chih Han Yu: Thank you for inviting us. So my name is Chih Han. I am the CEO and co-founder of Appier, a leading AI for marketing technology company. 

So our company was actually established long before AI became a thing. The company was established back in 2012. Now we have scaled to 17 different countries operating across three key continents and servicing more than 1500 customers with annual recurring revenue of more than US$150 million. Appier is also a listed company on the Tokyo Stock Exchange. We became a listed company in 2021. 

And now we have started the company in an era when AI wasn’t even a very popular word, and experienced the first scaling phase of AI through the popularity of AlphaGo. And recently, this is what we call a third phase of acceleration due to the popularity of ChatGPT and others. 

And I think it’s a very exciting period, an era because I was an AI scientist before starting the company, really starting when AI was very academic. My work prior to starting Appier was applying AI and machine learning in robotics applications, including robotic dogs or self-driving cars. Yes, there was a self-driving car in the early 2000s already. 

And then finally, with a vision of turning AI into a return on investment or democratizing AI technology back in 2012, I then started the company in the period when AI was relatively unknown. 

So I think now we were really experiencing a period when AI started as an academic subject, then saw really early adoption. I would call 2015 to 2020 the early adopter period. 

Today is a mass adoption period where everybody has already experienced the power of AI, and a lot of applications are being developed. So I’m very excited to join the panel and also operate in a team that continues to deliver value for customers in this exciting period.

Jennifer Zhang: My name is Jennifer Zhang, and I’m the CEO and co-founder of Wiz.ai. Usually, I break our company’s introduction into three phases. 

Number one is actually five years ago, what problem were we solving and the product we offered was voice recognition for local accents and dialogues in Indonesia, Bahasa, Thai, and Tagalog, through voice AI bots. 

The second phase was actually three years ago, when we expanded not only into the voice channel but also other communication channels like email, WhatsApp, and all the other customer communication channels into one solution. We also analyze all the conversation data and decide what kind of profile the customer can delegate to humans [for engagement]. So it’s a human-bot collaboration model. That’s phase two. 

And this year, we opened a new chapter and deployed all these generative AI models, specifically a lot of larger models, into enterprise applications. On one side, we enhanced the customer experience by turning call centers into multi-language multimodal models. We also use the same models to enhance not only the customer side but also in-house employee working experiences like helpdesk, HR, and IT. 

That’s our three phases of growth, and this year, I think it’s quite exciting for us to see that everyone can touch upon how powerful ChatGPT can be. This is giving us more chances to talk with all our enterprise customers, since we have this kind of [tool], what exactly do you want to solve, and give us a better understanding of what exactly this can solve, like what kind of problems exist with current enterprise users.

Christian Schneider: My name is Christian. I’m the CEO and co-founder of Bluesheets. Our company is about four years old now, and we’re in the automation space. We are using and building AI models for automation in enterprises. 

We specifically focus on automation of unstructured data, which means wherever in the enterprise there’s unstructured data coming towards you in the form of applications or internal procedures, we help facilitate and structure that data. 

Eventually, we make it available for your internal systems much faster. There’s usually a chain of events that has to happen before data can be processed internally, whether it ends up in an ERP system or an internal legacy system. So we really help facilitate and automate that process end to end. 

We’re also Singapore-based. We are currently expanding into the Asia-Pacific region. We have customers in about 17 countries right now. I’m very excited to be here, of course, as well.

What Does AI Transformation Really Mean?

Yongcheng Ong: Thank you for the introductions. So for the first question, I’ll address it to all of the panelists, right? Thankful that you guys have shared a little bit about how the businesses have evolved over the years and all this.

But these days, when we talk about AI transformation it is often quite broad, right? Perhaps you can break this down for the audience. What does AI transformation actually mean for your business? And, if we take a step back, how do you see the state of AI transformation in Southeast Asia relative to the rest of the world?

Chih Han Yu: Because we focus on marketing digital marketing I can classify it into three stages. 

Prior to 2015, people simply didn’t care about AI at all. So any applications for marketing, people only cared about whether it actually brought true value. So our company was established during that period of time. For enterprise customers, they didn’t really care if it was AI, AR, or VR. As long as it delivered value, they adopted such a solution. 

And the second period was the early adopter period from 2015 up to 2022. That’s the period when the advancement of AlphaGo and the popularity of deep learning technology, along with the reduced cost of GPUs, made people start to believe that AI might have superhuman capabilities.

That’s when we began to see AI applications emerging, especially in precision marketing, where AI could select the right audience and optimize marketing budgets effectively. This was a phase of technology-focused companies, including our own. 

Then, starting in 2022, the generative AI era became even more interesting. For marketing, in addition to precision execution, we began to see creativity emerging. AI could generate the right images, text, and marketing plans. AI started exhibiting creativity. Currently, only 3 percent of marketing technology involves AI, but we anticipate exponential growth in this area over the next nine years. This is an exciting period where we’ll see growth not only in precision marketing but also in creative marketing.

Jennifer Zhang: I find that Southeast Asia’s thinking about AI is quite similar. 

The first thing I noticed a couple of months ago when I met one of the largest banks in Thailand was their five-year digital transformation plan. 

I realized that this plan could be summarized into three parts: abstracting data to structured data, workflow automation, and improving service efficiency to maximize income flow. These are the three goals they set for their digital transformation journey. 

What we try to do is align our solutions with these goals. It’s like upgrading from a regular Mercedes to a sports car. You offer a solution that they can afford, but it must deliver the desired outcomes. This is a crucial aspect.

The second observation I made is about the reliability of AI solutions. People have experienced AI personally, but adopting it in an enterprise setting is different. There’s a journey from consumer apps to enterprise solutions, and we need to ensure reliability. We should focus on existing use cases and solve pain points from the past. Don’t just sell the future; solve the problems from the past.

The third observation is about getting closer to the nature of the business. Previously, everything was structured around ERP, CRM, and other frameworks. With AI, we break through these frameworks. To be successful, you need to understand the business deeply. Understand why the business can make money, and then provide solutions that align with those goals.

The fourth observation is about Southeast Asia’s talent. In the past, many companies had to outsource IT talent to India or China. Now, the bar for development capabilities has lowered, and Southeast Asia’s talents can play a more significant role. For example, people who were doing data annotation can now become prompt engineers. This is an opportunity for Southeast Asia’s talent to participate in the transformation.

Christian Schneider: I couldn’t agree more with Jennifer. One more point maybe on the role of Southeast Asia that we see right now is there is an opportunity for Southeast Asian businesses to tap into AI early on, no matter where they are in their digital transformation journey. It will likely still come from the big tech cloud providers, but there’s definitely a great opportunity now in terms of efficiency. 

Specifically, how we at Bluesheets support that is by cleaning up and preparing data for these use cases. We find ourselves in a spot where we can support existing initiatives in companies. We see the first POCs happening toward AI, no matter where in the world, actually.

And then we can also build our initiatives on top of our data models. We can propose projects for efficiency within a company, whether that’s in procurement or HR processes. We clean the data and then offer up services. This could even include services provided by other companies like Appier or WIZ.AI. This is how I think this will evolve in the future. 

Many companies will build their own models, but they will face hurdles and may not venture into all categories within their business. It’s an exciting time, especially for Southeast Asia, to leapfrog some competitors as well.

Yongcheng Ong: I appreciate that all the panelists have shared different perspectives on how they view AI transformation in general. 

Appier has had a front-row seat since well before AI became a thing, which I find very interesting. 

To address Jennifer’s point, I believe speaking the client’s language is crucial. We’ve encountered cases where companies have great products, but they struggled to communicate effectively, leading to lost sales.

I’d like to share an example of an AI fintech company that primarily sells report writing automation. They were offering a complex technical solution to banks. However, they eventually realized that what they needed to tell the banks was simply that they could write reports more efficiently and cost-effectively than using their own BPOs. It was a simple message that took them a while to figure out.

I also believe that generative AI has changed the buy versus build equation. Building solutions has become easier, and it may reshape how SaaS businesses are viewed in the future. 

Lastly, to Christian’s point, there’s a wealth of data available, but many companies have inertia when it comes to understanding and adopting new solutions, especially for unstructured data. Companies like Bluesheets will play a role in tidying up this part of the world.

Leading The Third Phase of AI Adoption

Yongcheng Ong: For my second question, I’d like to ask Chih Han first. You’ve been building AI-powered marketing SaaS solutions for more than a decade, and you’ve seen how the industry has evolved. How has the resurgence in AI interest impacted Appier’s growth, especially since you’re a public company?

Chih Han Yu: As mentioned earlier, in the initial phase, people were looking for solutions to boost their business performance. Our solution was built around that framework. In the first period of AI adoption, especially in marketing technology, it was focused on decision-making. 

I can share a story from 2017 when we started pushing generative AI applications to our customers, allowing them to create their AI-generated content. However, customers back then were more hesitant due to concerns about control and brand compliance. This slowed adoption, and we ended up only publishing academic papers about the technology.

Fast forward to 2023, and the same technology is being used with high adoption rates. What changed? Customers’ experiences with using AI, especially through GPT-like models, shifted their mindset. They now understand that while computers can make mistakes, they are generally correct and can enhance their capabilities. 

This mindset shift unlocked potential in the creative side of marketing technology. Nowadays, in addition to precise decision-making, AI also contributes to the creative side of marketing, as marketing combines both creativity and execution. This is an exciting time, and we can achieve end-to-end automation while addressing customer talent concerns.

“Customers’ experiences with using AI, especially through GPT-like models, shifted their mindset. They now understand that while computers can make mistakes, they are generally correct and can enhance their capabilities.  This mindset shift unlocked potential in the creative side of marketing technology.”

Commercializing AI for Enterprise

Yongcheng Ong: Thank you, Chih Han. Jennifer, it has been a landmark year for WIZ.AI. You’ve been pioneering AI-powered customer engagement solutions and building foundational models for enterprises. What insights can you share about driving product adoption at this early stage within the industry?

Jennifer Zhang: That’s actually quite a good question. As I mentioned, when I usually [pitch] to customers, I always [initially] address quick wins because all the management wants is an increase in their confidence level [in the technology]. 

There are actually some cases where they already have, for example, previously used a lot of chatbots. They’re not doing pretty well, but they have enough data sets. So how you actually use the relatively small in-house [infrastructure] to make their chatbots much more powerful. That’s actually the first easy thing you can do. 

In other parts, I think usually we’ll try to lock down a territory where [the customer] can easily feel the change in the business value. Then we also try to find the customer with one, more structured data, and second, an in-house team with more understanding and more sense of digital transformation. Then we will show them one case. And then we expand that into the different departments, from one department to another department. 

I believe it’s not one time use and you expect them to adopt everything. You are selling the whole [solution], but you need to have a very clear land and expand strategy for each enterprise. And you need to really handhold. For most customers, sometimes they want to jump, but they don’t know where the ladder is, and you need to offer that. 

That’s something you should do, especially currently, in Southeast Asia, as the region is still in the early adopter stage for enterprise. And later, once you have one or two cases ready, then you can quickly standardize and make product lines and go sell in the same industry. 

This is actually five years ago when we actually promoted the voice AI solution, starting with one technology, then later expanding into products, then an entire solution, then extending the use case to all the types of business departments. 

I think the same will be happening here with the generative AI solutions, that from the one simple function use case, it will be applied to different departments and then the whole company.

“I believe it’s not one time use and you expect them to adopt everything. You are selling the whole [solution], but you need to have a very clear land and expand strategy for each enterprise. And you need to really handhold…That’s something you should do, especially currently, in Southeast Asia, as the region is still in the early adopter stage for enterprise.”

AI Transformation Means Data Transformation

Yongcheng Ong: That’s a very good point because in Asia itself in general, there is a higher level of service elements that enterprise companies often have to deal with, and hopefully, productization and bringing about a new standard of service level agreement. 

Christian, for you, my question would be, when it comes to AI transformation, the very foundation of it really lies in the data infrastructure layer. How do you see this data arms race evolving across various industries? And how do you see Bluesheet’s role in the grand scheme of things?

Christian Schneider: So in general, what we feel inside Bluesheets is that there are currently many legacy systems implemented within a company. They usually host the data. 

The way we see it in the future happening is that many of these legacy systems will become redundant because a lot of computations are either already done by AI before the data even hits the company or when it’s inside the company itself. 

That’s our premise, and therefore anything that is not currently tied to compliance, meaning reporting structure, meaning internal compliance, as a legacy system, might actually go away. 

For us, the education process starts there. If you need an ERP system, that’s clear. You need to be compliant. Do you, however, need three other systems that are sitting, say, in a procurement process before the ERP system that are actually just there to facilitate a problem that might not exist in five years anymore? 

So that’s where we come in usually, and we explain how the data should be structured. Some companies have already invested very heavily into internal processes, into automation, and they were, in their defense, restricted by the technologies that were available back in the days, which usually was RPA. There were a few other companies, of course, here, a few front runners in certain AI tools. 

So what we will see in the future, for us hopefully, is that many companies will completely restructure and use AI tools only until that moment when you reach that legacy system that will then be tied to the compliance, right? 

So that’s how we see the data arms race playing out. You actually do not want to manually process data or host data in many different systems at once. You will be seeing more data lake-like systems that will then be able to easily release and trigger the transformation of data or send data left and right.

When it comes to sharing data with cloud providers, we currently know that many companies are connected to almost all the cloud providers anyways. That means they will be able to pick and choose, and I think that’s good. There will be most likely more providers for specific tasks popping up. They create and train their own large language models. 

But in general, with the data arms race, we’re really more concerned about how the enterprise prepares their data for AI rather than how they get more of it. Because subsidizing data from the outside world will probably work to a certain extent, but if it’s really meaningful is the other question. 

If it really comes down to ROI and measurable ROI, whether that is in terms of man-hours or whether that’s in terms of efficiency, then usually we would default to internal data to be used for training purposes rather than from the outside. 

That said, I feel that there’s a lot of opportunity right now for companies that focus specifically on promoting data transformation. That is really what’s lacking right now. You have so many different silos. It’s so hard right now to even come up with a project because so many different parties might be involved in this. There’s no easy access to the data that you need. Sometimes it’s not even in the cloud. 

And that brings me also back to my earlier point about Southeast Asia. So that’s an opportunity because if you have not yet gone so far in the automation journey, you might actually see yourself now in a position where you can get right to the top because you can skip a few steps. And I think that’s quite exciting for businesses here in the region.

“We’re really more concerned about how the enterprise prepares their data for AI rather than how they get more of it. Because subsidizing data from the outside world will probably work to a certain extent, but if it’s really meaningful is the other question. If it really comes down to ROI and measurable ROI, whether that is in terms of man-hours or whether that’s in terms of efficiency, then usually we would default to internal data to be used for training purposes rather than from the outside.”

Advantages of Being Global AI Companies with Roots in ASEAN

Yongcheng Ong: I think the role that you play is actually quite pivotal as well because aside from access, I guess what Bluesheets is also trying to prevent is garbage in, garbage out, right? 

If we take a step back right now and look at the competitive landscape, with AI and all this, the competition will inevitably be global from day one. So do you guys think having your roots here in Southeast Asia, would that bring you any unique advantages when it comes to competing at a global stage?

Chih Han Yu: Our key operation is actually in Northeastern Asia and also the U.S. market. In South Asia, we also have a presence for our regional talents. But let me talk about the Appier experience of when we entered the U.S. market as an example. 

So we entered the U.S. market only in 2021, after we went public. Back then we thought that the U.S. market is very sophisticated, and probably it’s a market that is relatively harder to crack. So we really prioritized the Asian markets first, so we have more resources as we went to the U.S. market. 

But it turned out that the U.S. market became historically the fastest-grossing region for our company. We scaled from 0 percent of revenue to 15 percent of our revenue only within two years. That has been a tremendous growth driver for our company. 

But when we think back about global competition, that also means global opportunities. And sometimes the most sophisticated market doesn’t mean that it is very hard to crack because a lot of competitors already help you to educate your customers. 

I think when we operate in Asian markets, a lot of time we have to handhold our customer’s staff on Machine Learning 101 to introduce why using AI is better for marketing and then why this type is better and why this type is superior. 

So that actually is a very long education process. That also means a very relatively long and also tedious education deal closing cycle. When we launched in the U.S. market two years ago, we found that our buyers already knew the basics about machine learning. They also have team members that are semi-experts in these domains. 

Then we can go straight to the conversation to talk about why we are better. And that self-education or vendor-educated process has allowed us to advance to more mature conversations that actually make us shorten the selling cycle quite significantly.

So education is actually a very high toll for startups that are entering the more sophisticated markets. But that actually helps as an advantage of having a customer that’s ready for your solution. 

Nowadays in Appier, we really look into two most important dimensions when we decide which market and which application to enter. One is data readiness, whether the market has sufficient data. Second is whether the business readiness, whether businesses are ready or educated with the basic concept of AI. We found that with both factors being very mature, it’s the best sweet spot for us to enter. 

So just a bit of thought about our entering the U.S. market as an example and how what we thought was global competition were actually also global opportunities.

Jennifer Zhang: So I cannot speak too much about the U.S. and Europe market because we’re just exploring. 

The good learning for us is actually because we tapped first into the Southeast Asia market, when we actually used the same customer for piloting similar solutions it landed pretty well in Latin America. 

So all the companies we talked to before in Southeast Asia, the old rich telecoms banks and also the new rich, the fast-growing fintech, e-commerce, all use cases can literally be copied into LATAM, and actually with our Portuguese and Spanish capability. And the lucky part is that LATAM actually pays higher than Southeast Asia.

I think it’s just because those two markets are still in the growing phase of the digital transformation. Then it gave us a good advantage to show how you leverage the data, the whole process automation to improve their growth in e-commerce and fintech, those industries, and also doing AI transformation for some use cases in telecoms or banking. 

Another way that what we’re building here in Southeast Asia actually helps a little bit is when we explore the Middle East market, but we haven’t expanded too much there. What we realized is that even in the Middle East, the contracts are bigger, but there are less challenges when we’re deploying in multiple countries there. 

So I think that’s actually two advantages we’re building here in Southeast Asia and expanding into other territories.

Christian Schneider: For us at Bluesheets, we are actually a data processing company that can technically be applied anywhere, which is great for us. 

In general, the foundational models, especially the expertise around building products on top, will probably originate in the U.S. We see that already happening right now. But what we feel is the opportunity now, it’s a new playing field. 

So how we use them, whether we use open source or not, is up to the companies that are at the forefront right now. And I think for any companies looking into coming up with B2B enterprise solutions, there is an array of opportunities now, and I think we’re going to see a lot more level playing fields in segments and industries, companies going after the POCs from different parts of the world, and we don’t really feel a restriction by being a Southeast Asian-based company. 

As Jennifer also said, the complexity here tends to be higher from the start, which I don’t think is a problem. I feel with AI, solving the easy cases has become much easier anyway, and so really what you need is companies who have solutions that go broader and also cover more ground when it comes to complexity. 

So I think this is very exciting for us. But I also have to say, of course, I think most of the technology will originate from most likely the U.S. side, I believe, and that is also where a lot of strong products will originate from. I hope that a lot more products out of this region will, however, make it to the global stage for sure.

Yongcheng Ong: Thank you, everyone, for all your insights. I think it’s quite interesting to note that, for all three of you, you guys found Southeast Asia a relatively hard market to conquer. And basically, once you are able to conquer it, going global is actually quite an easy task for you guys. I’d like to take this opportunity to thank my panelists once again for your time and all your insights.