OpenAI Sam Altman Singapore & Jakarta Townhalls, AI Regulation & Job Displacement and Founder Stress, Burnout & Failure Report by Startup Snapshot - E286

In the discussion between Jeremy Au and Shiyan Koh, several key insights emerged. They explored the concept of founders dealing with success and failure, particularly during the winding-down phase of a startup. Shiyan emphasized the importance of founders conducting themselves with integrity, treating everyone well, and communicating effectively during this challenging period. It was acknowledged that not all outcomes will be ideal, but behaving ethically and in good faith can positively impact how investors perceive and support founders in their future endeavors.

The conversation also touched on the issue of founder stress, burnout & failure report by Startup Snapshot. Shiyan highlighted the significance of taking breaks and encouraging founders to prioritize their well-being by allowing themselves to disconnect and recharge. They discussed the benefits of taking vacations, going off the grid, and engaging in activities like hiking to free the mind and gain new perspectives.

Jeremy shared his personal experiences of finding restoration and clarity through hikes and off-the-grid experiences, including his month-long hike on the Pacific Crest Trail. He emphasized the value of giving oneself time and space for personal growth and reflection, and encouraged founders to seize opportunities for off-the-grid experiences, especially during transition periods.

Additionally, the discussion addressed Sam Altman's push for the regulation of AI and the use of "co" language such as "co-evolving" or "co-pilot" in relation to artificial intelligence. This highlighted the need for humanity to co-exist and evolve alongside AI, suggesting a paradigm shift and a call for responsible development and utilization of this technology.

Supported by Esevel

Do you manage your own IT for distributed teams in Asia? You know how painful it is. Esevel helps your in-house team by taking tough tasks off their hands and giving them the tools to manage IT effectively. Get help across eight countries in Asia Pacific, which includes onboarding, procurement, device management, real-time IT support, offboarding, and more. Gain full control of all your IT infrastructure in one place with our state-of-the-art platform. Check out esevel.com and get a demo today. Use our referral code, "BRAVE" for three months, free. Terms and conditions apply.

Jeremy Au: (01:20)

Morning, Shiyan.

Shiyan Koh: (01:22)

Good morning for me. It's a good evening for you.

Jeremy Au: (01:25)

Well, it always works out somehow, but here we are across time zones and excited for this episode. So I think we have two major topics. First of all, it talks very much about Open AI's visit to Singapore and Jakarta. So talking a little bit about that and what it means for Southeast Asia and so the topics are covered. Of course. Secondly, we have had a recent report discussing kind founder stress and burnout and some of the quantitative numbers about something that we know quite well qualitatively.

On that note, just wanna quickly do a shout-out to Anthony from Trihill Capital, a partner who's excited about Indonesian, Singapore and gave us lots of good feedback about our buddy's charisma. And so I think we'll try to incorporate that moving forward. So shout out to you, Anthony. The second person to shout out to is Choo Yi Lin for recommending the startup Snapshot report on founder stress. It was a tremendous recommendation. So definitely good fodder for conversation. So if you're a listener and you're tuning in, feel free to email or WhatsApp or I don't know, smoke signal to us something for us to chew or digest or answer on your behalf. On that note, Shiyan, do you want to share and wanna chat a little bit more about open AI in Singapore and Jakarta?

Shiyan Koh: (02:39)

Many of our listeners are aware that Sam Altman and OpenAI were in town on the 13th. And they did sort of a series of meetings, but also an open forum, I think with a thousand plus attendees where he took audience questions and did a little bit of a small talk. And it's kind of part of this extended, I think like 16-city world tour where they're kind of going around meeting developers, kind of sharing their vision and telling their story about what is open AI up to, it's not the scary boogie man inviting regulation. And more conversation really around what's happening in that field. So I wasn't there myself, but I do have several spies, no lah, friends who are attending. And so, it kind of gave me a little bit of colour on the conversation. And I think, it is what's top of mind for people who are working in the field, whether they be regulators or startups and other technologists. And so I think, it's interesting that he's in Singapore and Jakarta.

I think Singapore is the nucleus of Southeast Asia like tech and policy. He was hosted by AISG and they were moderating the conversation. And so I think, folks are kind of eager to be part of that conversation. And think about also how we ensure that a lot of these foundation models are not only trained on Western data points but also incorporate local data, the local language, and local norms as well. And so I think that is also part of what regulators are thinking about in addition to just regulating the output, is also thinking about what's the input. And so if you think about how increasingly people are incorporating Chat GPT, Anthropic, and other models into their workflows, what else can people do to continue to like to localize, improve the results and build something useful for humanity? And so I think that's like a pretty interesting set of questions for people to consider as they're building.

Jeremy Au: (04:36)

Yeah. I think the simple fact is that they chose to go to Singapore and Jakarta is quite revealing right, in terms of his priorities as well. For Japan, for the US and, Europe to make I think a certain set of conversation and talking points. I think tracking the consistency of his message is quite clear to me that there are two major talking points that he's doing lot to emphasize. The first of course is that he's pro-regulation which is, I think quite an actual proactive stance actually, I think is quite uncommon. I think for early-stage founders, for a company that has grown tremendously, there's a lot of regulatory action, but I think this is probably like, from product public launch to, doing a global tour to advocate for regulation consistently is interesting.

Shiyan Koh: (05:25)

Well, there's a cynical perspective to that, right? This is like now that we're public, let's regulate and keep everyone else slower or down. And I haven't seen a ton of specifics on what that regulation would look like. I think it plays well to regulators to say like, Hey, look, we're so open. Come regulate us. But at the same time, I think there was a draft of the European proposed regulations and he's like, yeah, we're gonna try to meet them, but if we cannot then, we'll just shut down in Europe. So that's also like a double-edged sword, right? Which is like, yeah, you wanna regulate, but if you want your people to have access to this stuff, you need to regulate in a way that allows us to still operate, right?

Jeremy Au: (06:04)

Yeah, and I think what was interesting is that we've seen this kind of regulatory differentiation, but also arbitrage plays out. I think one of the interesting analyses I was reading was that I think the country that ends up being the most, you can say permissive or lax, depending on which site you take on AI data and usage, will end up becoming a hub for AI companies. I think there's a very interesting dynamic, right? And I think we saw that to some extent actually with crypto, where there was, in Singapore, a more permissive regime on crypto. And so we saw Singapore become a crypto hub. And now we've seen that with the recent.

Shiyan Koh: (06:36)

Well, no, no. So I would push back on that, right? Which I think was permissive. I think they were just clearer, yeah. So for a long period, the SEC didn't say anything.

Jeremy Au: (06:47)

Okay. That's fair.

Shiyan Koh: (06:48)

So then it was like, people were like, well, am I, is this a security? Does this pass the Howie test? Like there's a lot of like, and so I think MES was on the forefront of like like defining things more clearly upfront and creating a process for people to do it. And so people came, but I don't think it was permissive. And I think you see that in the flow of companies to Hong Kong and Dubai. Because where MAs came down pretty strongly was around advertising. So, they shut off all the advertising to consumers, to retail customers, essentially by crypto, which, really shut off a lot of growth for people.

And so I think there are two things, right? Which is like, one, you want a reasonable set of rules to play by that are articulated. So you can continue and you don't want, like arbitrarily, something will calm down and say, Hey, you're not allowed to do this. And you're like, but you didn't tell me. So I think that's a little bit part of the play here, right? This is like engaging regulators early so you can set out some rules of the road so you can continue because you think there is a potential that down the line. Regulators could go too far and overreact, and that uncertainty is worse for you, right? So you can like proactively co-write regulation together rather than say, Hey, I'm gonna be a cowboy and then someone's gonna come and smack me.

Jeremy Au: (08:05)

I think was else was interesting was that Rachel Lim was a Singaporean that was kind of co-hosting with Sam Elman. So she's part of the Open AI team. She's a, I believe, Raffles junior college alumnus. So kind of interesting to see that dynamic that's there. So some local representation on the stage. I thought the key phrase that they had that became the headline was very much about how humanity must co-evolve with artificial intelligence. And I thought that was a very striking statement. I think coevolve I don't know whether it was the ad-libs or was that prepared. But I think there's, I think a sense of a generational shift, right? An evolutionary term paradigm shift, of further evolution, right? And as if there's a species or some kind of like, an accelerant to, I don't know I don't evolution itself, right? So I thought it was an interesting dynamic to say, yeah.

Shiyan Koh: (08:57)

It's kind of true, right? Like if you think about it, how are human brains have you evolved? A hundred years is not a lot of time in an evolutionary timescale, right? But you think about what has happened in the last hundred years in terms of access to information, and how we consume things, it has been a huge change. And it's not clear that people have trained themselves to, harness it more effectively. It's like, why like access to processed foods and sugar just make people fatter? It doesn't like to reduce global hunger. It's just people who have access even fatter, cause your body is tuned to say like, I want that delicious sugar and fat.

Jeremy Au: (09:45)

So, humanity must co-evolve at McDonald's.

Shiyan Koh: (09:48)

Yeah. No. But like I, I think that's like a, it's like a totally good point, right? Which is like, In one sense, you could say, wow, we have the internet now. Everyone on earth with like an internet connection, and a cell phone has access to more information than anybody in all of history. Right. But what does that happen? Does that lead to more people becoming educated, more people becoming enlightened? Or does it just lead to larger and larger consumption of our day with entertainment and being like broadcast stuff into our heads? Right? And so now we're at this next stage of like co-piloting with AI and that's gonna like transform a lot of our workflows. How do we co-evolve with people so that they don't just blindly do whatever the AI tells them? Like we already had that first case of the lawyer who submitted fake cases generated by a ChatGPT, and he was like, oh, I didn't know that it could hallucinate.

Jeremy Au: (10:45)

What happens? A lot of timing. That's what happens to the 737 MAX, right? Is this like all these pilots, obviously it's so autopiloted and then, without sufficient training, basically, the co-pilot that's, the system ends up becoming an autopilot and the autopilot standard doesn't work. Right. So I think, something like this, right? I think everyone in AI is using the word co to hide something. Like whenever they say something all the time. You always have to watch out because they're trying to hide something the opposite.

And what I'm noticing a lot for AI stuff is they're using a lot of the word co, so copilot, because they're saying we're not removing control. It's a copilot, it's co-evolve. So this co-word is coming up over and over again. Basically to imply that there's a shared status, right? There's a peer, right? And I'm just saying like, I think as you said, I think there's gonna be a lot of autopilot folks. There are a lot of people automatically, but I think it's only a matter of time before.

Shiyan Koh: (11:44)

So I think the 737 MAX is a really interesting example, right? Because that bug, right, that caused the plane crashes, had been reported before, but it had been reported in incidents that were not fatal. And the fatal incidents were in budget Airlines that had, I think, pretty deliberately tried to reduce the amount of training to increase the supply of pilots to staff their flights. And so, yeah, like if people try to take shortcuts like that, Right? It's sort of like co-pilots work for the two standard deviations, right?

Like the number of instances, you need to have like better judgment and more expertise for the tails. But at that moment when the tail happens and you didn't train the person enough, right? Then you're screwed. So I think that maybe you say co-evolve is to hide something, but I think co-evolve is an important word to say. Like, Hey, you can't just abdicate all responsibility to the AI. You have to put effort into thinking about how people should responsibly interact with this.

Jeremy Au: (13:05)

I think capital and asset owners are trying to like, scale and I'm gonna make the air sound. And the reason why there's the air sound is that you make it sound as if individuals get to choose to be on a peer basis. I think a lot of asset owners are saying, how do I rebuild this business model? Whereas it's mostly AI and some humans, which makes sense. I mean, use a factory model, right? The industrialization of production, of content. And I just think that for a lot of folks, I think there's gonna be a law effect where folks are gonna be churned out because of these productivity improvements, effectively taking over their jobs. And one discussion I was having, was at the Ace Awards Gala. We, I received an excellent award for I guess, Ace aged 35 and above.

Shiyan Koh: (13:53)

You're so Jeremy.

Jeremy Au: (13:54)

So fancy. But I thought it was a really interesting conversation. I gotta have a little bit cause we're discussing our political ramifications, right, of this. And what was interesting was that we had seen this giant wave of globalization. Blue-collar jobs were moved towards China and obviously, China saw the fastest rise in living incomes for poor folks, et cetera. But we also saw the hollowing out of the blue-collar, middle-American industrial class, right? And so that has dramatically reshaped the political coalition, right? Of blue-collar workers. So there's a weird alliance between business Democrats and business Republicans, but it was okay for different reasons.

And so we saw that different change, but that was all for blue-collar. Right. And now we have white-collar professionals where I don't think they get a choice to be a copilot or not. I think a lot of their roles are gonna get reduced change and sure, eventually, I think in 10, 20, 30 years, we are all gonna re-skill ourselves in different roles. But I think there's gonna be a lot of jobs that. It's gonna go away. Right. I think we see the writer strike if they are insistent on the AI side of it. They're worried about their jobs being automated out to some extent. So I think there's this interesting dynamic where I think we're only seeing the very beginnings of this. Yeah. I don't think it's a copilot. I think it's mostly autopilot with some checks. Yeah.

Shiyan Koh: (15:13)

Well, we should like to make a bet in 10 years.

Jeremy Au: (15:19)

On what? Like?

Shiyan Koh: (15:20)

Like what jobs? What percent of jobs have been automated away versus not? Maybe. Okay. We will make the bet right now. We should think about this next week.

Jeremy Au: (15:29)

Well, the interesting part was that McKinsey reported on what jobs would get automated away. That was done about a couple of years ago and I thought it was a really good report, but they did not consider this at all. So actually they wrote about how a lot of these professional jobs that require communication were like a good skill to learn to avoid getting replaced by robots. And now I'm just laughing a little bit because this report has to be updated now. So now I'm like wondering like maybe the only job that's left is like a celebrity and maybe a doctor and like the guilt, right? Lawyers. Soldiers.

Shiyan Koh: (16:02)

I mean, there are auto-generated celebrities too.

Jeremy Au: (16:07)

Oh yeah. I met once today. Startups and I was just yeah, that's another thing altogether. For all the lonely people out there, you don't need to meet someone special. You can just have AI.

Shiyan Koh: (16:16)

I mean, yeah. Replica 20 bucks a month.

Jeremy Au: (16:18)

I think in China some reports of folks who are falling in love with the AI's. I don't know. Interesting times, right? I think it'll be a countercultural movement. It was like real humans, not AI. I don't know. I can't think catchy protest.

Shiyan Koh: (16:33)

No, it's just like artisanal honey.

Jeremy Au: (16:35)

Yeah. Organic humanity. In UC Berkeley, like we always had a lot of protests and free speech movements, right? And I was like the robots have gotta go. Down robots. I don't know.

Shiyan Koh: (16:53)

Don't quit your day job, man.

Jeremy Au: (16:55)

ChatGPT, can you please maybe some slogan writers? Slogans for the anti-AI and anti-robot. I don't know. So I read this really interesting article in the Economist, and I think there's an opinion piece about how at the end of the day, this isn't artificial intelligence because, there's, it's not thinking, it's not conscious, but what it's done is it's hacked language, right? As a model, right? You and I, I'm conscious, you're conscious, but we communicate through these words that we hacked on top of our biology, and so for the first time we have into non-intelligent beings in a sense, who can talk like us. And so we look at language as a form of sentience and consciousness.

But we look at dolphins and we look at monkeys and we look at pigs and you're like a hundred percent smart, but we're like, Hey, you don't speak English and you can't put out apostrophe where it needs to go. So guess what? You don't count. Right. So I think there's an interesting dynamic where at the end of the day, these language models are not consciousness, but they're this statistical probability language. But yet I think it's so good and it's really good enough that people are falling in love with it, starting to raise AI little kids. It's just kind of bonkers to see that. I think that's why I think it's transformational because not about intelligence, right? It's about language generation. From my perspective.

Shiyan Koh: (18:19)

I've always said my dog was smarter than my kids until about the age of two.

Jeremy Au: (18:28)

So, ChatGPT four at the age of two is how much smarter than your kids and the dog.

Shiyan Koh: (18:37)

I mean, yeah, but like the dog, you never have to spend time convincing it to go to sleep. You don't have to feed the dog. The dog kind of knew how to eat kind of from day one. It was really fast to potty train. The dog follows directions. Yeah, mean like the dogs like actually outperforming. We always tell the kids like, Hey, the toaster is like the best child, the most obedient child.

Jeremy Au: (19:02)

My favourite use case of ChatGPT so far. Someone said please pretend to be a loving spouse.

Shiyan Koh: (19:11)

Do you process your comments before your wife?

Jeremy Au: (19:16)

He was telling me that basically like, he has a long day at work, right? And the wife doesn't want to hear well obviously, so he just writes to ChatGPT like, oh, had a tough day at work and blah, blah, blah, ChatGPT is like, Oh, sorry, dear. How are you feeling about that? He just, he just, and then he like processes that on his cab right home, and then he's like, doesn't do that data dump to the spouse, and I'm like, that's hilarious because somewhere between a therapist and a really good friend, a spouse maybe, but I don't know. It was just.

Shiyan Koh: (19:46)

I can't decide how I feel about that.

Jeremy Au: (19:49)

I know it's interesting.

Shiyan Koh: (19:50)

Does it just reduce the resolution of conversation you have with your spouse or it's just sort of like you think your spouse is not interested in it, so you're like creating space to have conversation about things that they're interested in?

Jeremy Au: (20:02)

Oh, is this emotional cheating?

Shiyan Koh: (20:04)

Oh, I don't know. I think cheating is a hard word. Like that's a pretty harsh word, but I don't know. That's kinda interesting, right? Like, I mean, I guess you have different friends who serve different functions in your life, right? might have sports friends or business friends or whatever it is. Interesting.

Jeremy Au: (20:21)

Adultery with the chatbot. And you know what? It's not impossible. I think flat out, I think there'll be a, there's a headline that's gonna happen in the next five years.

Shiyan Koh: (20:30)

No, I think in replica they've already talked about it, that people use their replica friends to emotionally process things and it can feel like emotional cheating, they're also positive use cases where it's like, Hey, it helped them think about how they were relating to their spouse and do it better, even if it did start as like a venting thing.

Jeremy Au: (20:55)

Yeah. Talking about venting and stress and processing your feelings. So speaking about processing feelings and feeling stressed out about life we had an interesting report called Start Snapshot which is available and linked in the transcript on the podcast. But here are some key statistics about how founders are processing mental health. So I thought it was interesting to see how they said 72% of founders report that there's a negative impact on their mental health. 81% of founders are not open about the stress, fears and challenges. And I thought my favourite two pieces of that were like only 23% of founders go to a psychologist or a coach to talk about those feelings. And only 10% of founders turn to their investors to talk about their stress. Yeah.

Shiyan Koh: (21:40)

I often feel that part of my job is being a therapist.

Jeremy Au: (21:44)

About to only 10%. So if you are truly trusted by everybody, you have 10 times more.

Shiyan Koh: (21:51)

Do I want 10 times more?

Jeremy Au: (21:54)

You see, people can feel their energy, so they don't want to give it to you. But I think it's fair, right? I mean, the truth is it's tough, right? I mean, when I was a founder, I honestly trusted some investors more than other investors, right? And even for investors, I did trust I was thinking about, Hey, I only have one hour a week, right? To process some stuff, maybe one hour a month, right? And so am I going to be talking about my feelings, right? I'd rather just talk about work or the decisions, there are so many things to talk about. So I was stressed as a founder. And yeah I think I'll probably put myself in one of those buckets of, I was one of 90% who did not talk to my investor about my stress.

Shiyan Koh: (22:39)

I think it comes out in different ways, right? It's not like they call you intending to talk about their stress. It's more like you talk about the business and then you can kind of sense that they're a little bit down and so you kind of do a check-in and you're like, Hey, how are you feeling? Like what's going on with you? And kind of normalize it because. You know what we tell people when we write the first check is like, Hey, most of the time in a startup things are kind of going badly. So you don't need to put up a front for us, right?

You don't have to pretend like everything is up until the right and crutches it, because that's just not normally the state of affairs. And so it's kind of permitting people, like establishing that social contract to say like, it's okay to be like, it's not going that well. It doesn't mean it's the end of the world, it just means like, you gotta figure out how to, solve something locally or whatever it is. But I would say like 90% of the time you kind of feel like things are not going well. Something's always on fire, right?

Jeremy Au: (23:38)

Something's always on fire. That's a hundred percent right. And I think there are some numbers here, right? They're saying that out of like, a hundred percent of the founders, 44% have high stress. 37% have anxiety, 36% have burnout, 13% have depression, and 10% have panic attacks. So that's a sure, pretty sizable chunk. I mean, the way I think about it's like, yeah, half of the folks therefore are stressed out and one-third have anxiety, a burnout, right? And one in 10 have depression.

Shiyan Koh: (24:07)

Yeah, I don't, I mean, think it's think they also don't take care of themselves, right? Like they eat overeat, or they drink or they don't sleep enough or they're not exercising, right? And so all those things I think are important for being able to go to the distance. Like you need to have strong emotional resilience, but your body needs to feel good. So like you do need to sleep you need to exercise and you need to like, have places where you can like hang out with people and connect with people who matter to you. Because the worst is also if things are going badly on the home front, right? Like if your spouse is mad at you cause you're not spending enough time or they feel like you're absent then yeah, it can just be bad news bears all around.

Jeremy Au: (24:50)

Yeah, I remember my spouse was a, well, then girlfriend was frustrated with me because she was like, Jeremy you're not taking care of yourself. Right. You're not taking a haircut you're not shaving, right? You're putting on weight. You're stressed out all the time. You can only talk about work. And I was like, yeah, because all I think about is work. That's why I can't talk about anything else other than work. It was terrible. And I was, I think it was quite validating to see the report and be like, oh, okay. Thank goodness. It like, looks like, yeah. 59% found a steep lesson in starting a startup. It's like, oh, I don't feel like I'm part of the minority. Right.

Shiyan Koh: (25:22)

Yeah, but I mean, I think that's why co-founders are important. Yeah, because having someone who like really understands what's going on, I think it's hard for your spouse cuz they're not in it, right? They're forced to endure it, but they're not in it the way a co-founder is. And having co-founders who can like to balance you out and sort of also laugh, I think is super important. So I don't know. I just remember in the early days of NerdWallet, we would like to work and then we would all work out together. We would like exercise. So it's like we'd be in the office and then like there were like 10 of us or whatever, and we shared like, we went to the same exercise class and we'd all work out and then we would all eat like greasy Chinese food together and then we'd go to bed and we'd like to start all over again the next day.

Jeremy Au: (26:12)

I missed out on the step about exercising, so I would just work. I would drink, so for lunch, I would know it was terrible. It was terrible in retrospect, but I used to chunk like caffeinated soylent, so, I created it next to my desk.

Shiyan Koh: (26:29)

What kind of Singaporean are you?

Jeremy Au: (26:31)

Then, for dinner, I would get home and then I'll order like, a platter sushi. I wanna work workout. So I gained 20 kilograms actually, so I gained from normal weight. I became medically obese by American standards, not by Asian ABMI standards. It was just horrible actually. And then I remember after I sold the company and then I did my physical and then basically, my blood results were like, oh, you're pre-diabetic, you're pre-hypertension, you have bad cholesterol, you have no vitamin D, you have bad testosterone, you are obese.

And I was just like, look at your liver enzymes are terrible. And I was just like, look at this thing. And I got this report. I remember I was in line at a ramen shop. In New York and I was just like so sad and walked out of the line for this Michelin star ramen shop. And then I had to, I remember I, I walked to a dim sum shop and got some like steam stuff and I was just like, such abysmal thing. I was like, oh, I finally have time to do a health physical, and I got these terrible results and I kind of knew it was happening. Yeah.

Shiyan Koh: (27:44)

It's okay. Now you've got yourself on track, right?

Jeremy Au: (27:47)

Yeah, because I'm on venture capital. Yeah, I know. Because It's much less stressful and I think there's a lot more control. Right. I think and I think I have the benefit of a good social support network as well. Now I think I'm much more intentional now like I have to sleep harder in terms of like, Melatonin and the valerian root and the soft pillow and all this stuff.

Shiyan Koh: (28:10)

Weighted blanket, man. Yeah, It’s all about the weighted blanket.

Jeremy Au: (28:14)

I know, right. And then yeah. So I think another thing that we talked about a little bit was also yeah, 42% of founders report the fear of failure as the major source of stress. I thought that was a really interesting number. I thought it was interesting because actually over the past week I hosted the Phoenix offsite. So basically it's a virtual offsite for founders that are moving on to the next chapter in their life. Cause they exited their startup because they had to wind it down because there was a co-founder departure.

And so we had about kind of like a dozen folks just kind of like discussing their feelings and I think the, I think it was interesting to see that the language about what is success or failure was a really tough thing, and I think one person framed it up as like, okay, so you're successful, you've become a unicorn and I'm not that right for whatever reason, so I'm a failure. Is that right? Then I was like, whoa, like, business statistics, but that's a statistic on a company level. But to embed, company success versus personal failure was interesting, I don't know what dichotomy, right? And so I think there's an interesting dynamic between what the right language is for founders.

Shiyan Koh: (29:30)

Well. Do they think about it as a one-shot or they kind of wanna keep playing the game, like they wanna take more shots on goal?

Jeremy Au: (29:44)

I think they're just thinking about the last shot, the last game, right? And all the various mistakes or things they did well. So I think a lot of emotional processing and also the processing of the facts and the different circumstances. So I don't think they are thinking too hard, honestly, for the cohort of founders that was coaching, wasn't thinking too hard about what's happening. Yeah. I mean, is this one step down the road right before they can process that?

Shiyan Koh: (30:12)

You talk to a lot of founders in the valley and it's a funny thing, right? You kind of need to have two things in your head simultaneously. One is like, you kind of know that the probability of getting a unicorn is pretty low, but at the same time, you think you could do it right? That's why you're doing it. And you feel kind of like you love being in the game. And so even if this shot isn't the unicorn, right, you sell it, it's not an amazing outcome.

Whatever it is, you still learned a lot and you're gonna take those learnings and roll it into the next shot because you're playing the game, right? And the game is like, I get to be in control of my destiny. I get to build cool stuff, I get to work with people I enjoy. And part of the game maybe you can take breaks and go to Fang. Big company and rest for a little bit before coming out and playing again. The winning, I guess the success is more about the ability to play the game rather than necessarily generating a unicorn outcome every time you take a shot.

Jeremy Au: (31:21)

Yeah, I mean, I think that rhymes with another number in the report, which is about how 93% of founders would potentially start another venture despite feeling very stressed about the current venture. And I think there was a very strong sense as well, which is I think most founders in that group wanted to keep doing it cause.

I think there's a real strong sense of autonomy of control or freedom, but also I think there's a strong sense of mission and purpose that, and so a lot of the questions were like, we did have a panel of founders who either built another startup or joined corporate or became a VC, and a lot of the questions were about what you just said, which was like, Hey, how should I think about doing another shot? How should I rest? Would I be accepted by corporate, right? As a founder, right? And so there's, I think, an interesting dynamic of like dunno, what's the word? Like, feeling awkward about whether they would fit in that corporate mould, right? And be able to join that.

Shiyan Koh: (32:19)

But I think also, bigger companies, not necessarily corporate, but bigger later stage startups, hiring former founders are great because you kind of get someone who's like, Hey, I know how to make things happen with limited resources. And here I am, I'm a bigger company, I can give them some resources. Right? Like they don't need to go and raise money themselves. You're just saying like, Hey, run with it. Go make stuff happen. And those can be amazing for culture, especially if, you're bigger, maybe like things have slowed down a little bit. You're not feeling as much of that, go urgency. You can kind of inject these people into an organization so you're not going full-on like, I don't know, working for Acme Paper Company or whatever. But something that's still sort of in, in a tech-related innovation space, so.

Jeremy Au: (33:06)

Yeah. I think it's a tricky thing and I think the companies that are willing to be changed. Oriented and trying to fight for the future, I think I absorb founders better. I think there are a lot of companies that are more focused on preserving the status quo. Right? And I think that find it, it's not gonna be a culture match, let alone be a role opening for someone.

Shiyan Koh: (33:27)

Oh yeah. I mean, it'll be very frustrating, right?

Jeremy Au: (33:29)

Yeah. I think there is, I think a stigma still around, I think founder failure, right? I think Southeast Asia got to that first wave of like, now founders are profession and the career. But I think the stigma around failure and I think also varies across the region. I think there's a sense that I think Singapore's getting a bit better, but I think it's still pretty tough in Indonesia and Vietnam right now because they don't have that wave right of startups' success and failure and the recycling of that talent across the ecosystem. So I thought it was interesting to hear from folks in the ecosystem be like, oh, like, I can't find a job. Here at my local who was able to understand this, right? And I thought it was not an easy part, right? I mean, Singapore wasn't hospitable 10 years ago for a lot of this talent as well, right? So.

Shiyan Koh: (34:16)

But now there are other startups too, right? So you should have a bigger base of places to recruit into. Yes.

Jeremy Au: (34:23)

Yeah. But a lot of startups are also struggling as well, so, interesting.

Shiyan Koh: (34:27)

Was there a good conversation, kinda like reflecting on like what they would've done differently or kinda like what were their learnings or takeaways from the experience?

Jeremy Au: (34:35)

I mean, I think it's, long story short is no. I also don't think it was the right space to do so because, even for myself, like for my two companies, obviously took time for me to be like, okay, these are the things I did awesome and these are the things I didn't do great on. Right. And it took me, honestly, I think for both companies took me two years I think to fully process and get that full learning, right? So I would have Google. I think you just need space and time, right? And yeah, I did well and it was okay, right?

But still, the things I would've done differently in. I have Google Docs, right? For each company where I'm just like this is weird, right? It's like both of them are about 10 pages long, right? But these are the bullet points, right? It's like, these are the things I did well and these are the things that I could have done differently. These are the reasons why I did them wrong, right? On a fundamental level, right? So then there's a little bit recursive and I'm like.

Shiyan Koh: (35:29)

So like, can, would you share some of those learnings, like if you were gonna start a third company, like, kind of what principles would you bring into it?

Jeremy Au: (35:40)

Think for the first company, I think what I learned was that I learned that for myself. We were a consultancy, selling to the social sector, and then we were a consultancy that started selling to the government. And what I learned about this was that. Selling to the government was something that I was, found myself needing more energy in. Right. And so that was a personal realization and when I realized that, that's unlocked me to go find a successor to do that. Right. And so I think, but I kind of did that strategy without knowing that, it was just a discovery process. It wasn't a bad thing, it wasn't a good thing, but it's just like, that's not my thing to sell giant projects in that sense. Right.

Even though I think, of course, think Singapore government is probably one of the best governments to work with, and I think they're tremendous people. But I'm just saying like, that wasn't my energy sweet spot right then. For the second company, I think one big difference I would do differently was just thinking about the long term, right? I think I enjoyed education tech, and I enjoyed the startup journey. It's just that I was building it in America. And my heart was in Southeast Asia, right? In Singapore, right?

And so there's this interesting geographic realization to be like, oh, I would like to be home. I know this sounds cheesy, but I was like, yeah, I miss home. Right? I miss my parents, I miss my sister, I miss my friends. So I think it was an interesting dynamic, right? So I was like, yo, I'm glad if us to build something out, we'll build this out of Southeast Asia. Right. But I think it's not obvious. Right. Until you do it. So these are all realizations that take time to process. Yeah.

Shiyan Koh: (37:07)

Okay, cool. Well, thank you for sharing.

Jeremy Au: (37:10)

Yeah. How about you? When you think about it, obviously you help a lot of founders, right? Go through both the success stage, but also go through the failure stage in terms of winding down or moving on to the next chapter as well. So how have you tried to help or how have you tried to help them process?

Shiyan Koh: (37:28)

I mean, I think the wind-down thing is pretty rough, right? Because there's like the personal sort of how you feel about it, and then there's just all the like paperwork and cat herding. And so the whole thing is just not that fun. And so I think part of it is like, not giving permission necessarily, but sort of saying like, 'cause they kind of feel bad, right? So they wanted to win. They wanted to make you a ton of money, and now they're winding down and you know they've lost all your money. Or they'll go use cents on the dollar or something, right? So it's like no one's like excited about this outcome and part of it is like being able to say like, Hey man, it's okay. Note to founders, this doesn't mean you should set my money on fire, but if you made a good faith effort, you behaved with integrity, you tried hard and you like to treat everyone well on the way down like that's all we can ask for as investors, right?

Like we, we know the asset class we're in, we know the risks we're taking. And it doesn't mean that we're not gonna back you on your next thing, right? So it's kinda like how you conduct yourself in that moment, I think also determines how your investors see you and like, do you get to play the game again? And so some of it a little bit is just like, sometimes people are like, I'm so sorry, like, I wanted to make money and da, da, da. And you're like, yeah, this isn't the outcome we wanted. Like, I appreciate you treating everyone fairly.

I appreciate you communicating well and like doing this in a good way. And, give yourself that space and permission to like, take some time off before you kind of go rush into something else. Just kind of talking them through that. Right. So, so I think on the sort of like wind downside, that's kind of where it is on the stress side. I think when I do see founders getting burnt out, I will check when I say, when's the last time you took a vacation? And they'll be like, I don't have time for a vacation.

Jeremy Au: (39:22)

Yeah, that's what I was thinking. That's the answer.

Shiyan Koh: (39:26)

And you're like, but you don't have time to be burnt out either.

Jeremy Au: (39:29)

I do not permit myself to be burnt out.

Shiyan Koh: (39:32)

Yeah, but it's sort of just like, saying like, Hey, I want you to go take a weekend off. And don't, like, literally tell everyone you're turning off like you're not gonna check your email and just rest. Bring a book, bring a journal, but don't you just have to do it. And I know this feeling. I like it when I was in the thick of it, my coach was like, please just take four hours today and go away. Like, go for a hike. Yeah. It was just too much. I was like, I have too many things to do. I cannot take a day off. And my coach was like, you have to four hours, just go for a hike. Yeah, I know it sounds crazy, right? And then when you do it and you free your brain to have some new thoughts, you're like, oh my God, yes. Why didn't I do this? Yeah.

And so sometimes it's just like reminding people to do that. And I think a core corollary of that, especially in the remote world, is reminding founders to do it for their teams. Right. So, Like I told a founder, I was like, Hey, you guys have been driving hard for the last 12 months. You need to take your team on an offsite and it shouldn't be super work focused. You guys should just have fun, right? But like, you're gonna burn your people out if you do this. Right. So yeah, I think sometimes it's just like reminding people, permitting them even though they don't, they didn't seek my permission, but just sort of normalizing it.

Jeremy Au: (41:03)

I think that's talking about hikes and that time out. I think one of the things that I shared was that I think if you are a founder who's departing for a new chapter, this is a great opportunity to go off the grid, right? So after my first company, I went for a Vipassana retreat. That's like 10 days outta meditation, no talking. I went with my co-founder. So it was interesting cause we talk to each other all the time. It's not even both, like not talking to each other, but we're in the same space. Very conscious uncoupling, But after that I went on a hike.

Shiyan Koh: (41:36)

You're such a hippie.

Jeremy Au: (41:38)

Don't tell anybody. and after that, I went on a hike on the Pacific Crest Trail from Los Angeles to Yosemite. And I was reading and listening to Wild by Cheryl Strayed.

Shiyan Koh: (41:47)

How long is that you hiked the PCT?

Jeremy Au: (41:50)

Yeah. For one month. Yeah. I mean, and not the whole thing this month.

Shiyan Koh: (41:54)

That's so cool. I wanna do that.

Jeremy Au: (41:56)

Yeah, I should do that. And so I was talking with another founder of the Botanic Gardens, who is going for a walk because he's departing a successful company. And one of the conversations that we had was just like, Hey, there's an opportunity to do like the Camino De Santiago, which is you walk Spain to France, there's like a one month walk the chance to take a train. Right. For those long train rides across America, across Europe, all those things are like kind of off-the-grid experiences that you get to do. It's a small window of opportunity to do Right. And I think giving yourself time to do that.

Shiyan Koh: (42:26)

Did you hike by yourself for a month?

Jeremy Au: (42:29)

Well. I went with my then girlfriend and, after that hike, I was like, oh, maybe she can be my fiance and eventual wife. So, yeah, that was a nice moment to walk and hike for a month in a shed tent and all that stuff. It was great.

It was awesome. But then of course, once we came back to New City, we argued. I was like, no, I was feeling so lovey-dovey for a month. And then the city, we had some squabble about something. I was like, boo. And after that I was like, oh no, this is fine.

Shiyan Koh: (43:01)

Wait, did you get married? Is it your current wife or not?

Jeremy Au: (43:04)

Yes, it is my current wife.

Shiyan Koh: (43:05)

Okay. That was, confusing, which is like?

Jeremy Au: (43:08)

Confusing. It was a good story but became bad. My now wife was who I hiked the PCT with, and we were just, a boyfriend and girlfriend then. Yeah.

Shiyan Koh: (43:20)

Oh, that's amazing.

Jeremy Au: (43:22)

Yeah. It was fun. It paid off. Yeah. My funny story here very quickly is that to get out of the trail hit you had the hitchhike and I realized that having my wife be by the roadside hitchhiking, and then the car will slow down to pick her up and I'll just jump out of the bushes effectively, and I'd be like, Hey, there's two of us. We need a hitchhike. Right? So, no I thought it was a wonderful hike. It was very restorative. And that was the big difference because, after my first company, I had that window of time, but for my second company, we sold in 2019 Q4, and then it was a pandemic or after that, right? So, I just didn't have any of that space. But also, I mean, it was a terrible time of tragedy and I was just glued to my tv Right? Seeing all this stuff. So it was just a very tough time,e right? For everybody. Right, so.

Shiyan Koh: (44:14)

Well, I wanted to hike the Annapurna Circuit for my honeymoon. And my wife was like, no. She's like, I want to have hot showers and a bed for my honeymoon, and I was like, that's reasonable.

Jeremy Au: (44:28)

Yeah. Well, that's why you can do the Camino De Santiago. You get to do that with the hot showers, right?

Shiyan Koh: (44:34)

Yeah. So I still, I've yet to hike the inner person, but the PCT is on my list. I definitely wanna do that.

Jeremy Au: (44:39)

I know it's such a wonderful hike and it's beautiful, but that can be a hard hike.

Shiyan Koh: (44:41)

Would you ever do the whole thing?

Jeremy Au: (44:43)

I think there was a period when I wanted a section hike the rest of the Pacific Crest Trail, but now that I have kids, I kind of realized that it's not doable. Like, I mean, it's just a long chunk of time you gotta go to SF, the West Coast. So I think I'm hoping that maybe when my kids are teenagers, hopefully I can like, Half, I don't know. Coach them in the ways of nature so that I have a new hiking buddy I guess maybe for the rest of the trail.

Shiyan Koh: (45:07)

My friend, I have a friend who's hiking the PCT right now with his wife. But they're retired. This year the snow is, there's a lot of snow still because of all the rain and snow that happened over the winter. So they have to skip around to try to avoid the snow sections and then come back and do them later in the summer.

Jeremy Au: (45:24)

Thankfully I didn't have to do that during our timeframe. So, well, on that note anything else you wanna think about or share to wrap up this?

Shiyan Koh: (45:32)

I think the wrap-up is everyone should just go outside. Go for a hike. Turn off. Disconnect for a bit.

Jeremy Au: (45:41)

Don't let your replica model, bother you while you're out of of nature.

Shiyan Koh: (45:44)

Don't do it.

Jeremy Au: (45:46)

Yeah. I think for me, I think the key takeaway is, I don't know, I think the word for me is humanity. I think it was interesting that I think there's humanity in the context of open AI and Chat GPT, also humanity is in a phase of stress. Right? I think that's the one-word summary for me, humanity.

Shiyan Koh: (46:06)

Cool.

Jeremy Au: (46:08)

Awesome. Well, I'll see you next week, Shiyan.