1
- 3 years ago
Kaiju ETF Advisors builds, trains, and employs robust artificial intelligence (AI) and machine learning technologies designed to improve fund management decision-making. By empowering these innovative technologies to curate and provide direct management of our ETF, we’re striving to go places no Registered Investment Advisor has gone before.
Category
🗞
NewsTranscript
00:00 (upbeat music)
00:02 - What is up Zinger Nation?
00:04 Welcome back to another episode of Ben Zinger Live.
00:06 Happy Tuesday, everybody.
00:08 (upbeat music)
00:10 Ryan, how are you doing today?
00:16 - Good, how are you doing, Aaron?
00:18 - Good, it's good to have you back, man.
00:19 It's been a couple weeks.
00:21 - Yeah, it has.
00:21 I mean, I had a break and then you had a break
00:24 and there was future-proof thrown in there.
00:25 So it's gonna be back into the routine.
00:29 - Yeah, it really is.
00:30 So before we get started,
00:32 we were talking a little bit before the show,
00:34 sounded like you had a good weekend,
00:36 you got to get out on the water.
00:38 That's always good.
00:40 - It is.
00:41 I mean, I lived in Barbados for six years
00:43 and I am happiest on, in, or under the water.
00:48 Even really cold water like we have in Switzerland.
00:51 I'll take it.
00:52 Just whatever you got, I'm in it.
00:55 - How cold is it?
00:56 Is it like you need a wetsuit if you want to get in the water?
00:59 Can you swim with a wetsuit?
01:02 - Yeah, it depends on your tolerance.
01:03 Like coming here from Barbados,
01:06 I feel like I need a dry suit to get in this lake.
01:09 But the locals are like, they're in there in a Speedo, man.
01:13 - Really?
01:15 - They'll swim in that in February.
01:17 I'm not joking.
01:18 It'll be like five degrees
01:20 and they will be swimming in that thing in February.
01:23 I would be in like, I don't know.
01:26 Yeah, like a dry suit probably.
01:28 - Yeah, I mean, I know a lot of people
01:30 kind of sing the praises of the cold plunge
01:33 and taking cold showers.
01:34 And I feel like it's become a very,
01:37 very common advice among like the Huberman Lab,
01:41 podcast crowd and like the tech crowd.
01:43 But that's just one thing I have not gotten behind yet.
01:46 Maybe I need to give it a try,
01:47 but can't do the super cold water,
01:50 especially like the river.
01:51 - You gotta give it a shot, man.
01:52 I mean, I was skeptical through the whole summer.
01:55 You know, I moved to the Caribbean originally
01:57 'cause I hate the cold despite being a Canadian.
02:00 And, you know, I started with the contrast therapy,
02:05 you know, the hot tub, cold plunge, hot tub, cold plunge.
02:08 And man, like I got into it.
02:10 I got seriously into it.
02:12 It's how I start every day.
02:14 Now it's like 10 minutes in the hot tub
02:16 and five minutes or sauna
02:18 and five minutes in the cold plunge back and forth.
02:20 And I am like on fire.
02:22 Sounds terrible, but it's actually pretty cool.
02:26 You gotta try it first before you hate on it.
02:29 - All right, well, if you've tried it
02:30 and you're telling me I gotta try it,
02:32 then that means I actually gotta try it.
02:33 So I will, Ryan, I'll report back.
02:35 I don't know, I guess I could just like do a cold shower.
02:38 I don't really have a way to do it.
02:39 - That's awful, don't do that.
02:41 Don't do the cold showers.
02:42 Like on the back of your neck,
02:44 it's like drizzling down your back.
02:45 It's just the plunge is like 10 horrible seconds
02:49 and then you're good.
02:50 You just sit in there.
02:51 - All I don't have access to a,
02:55 maybe I could just hop in the pool at the gym,
02:57 like go straight from the hot tub to the pool and back.
02:59 Maybe that would kind of simulate the same thing,
03:01 but it wouldn't be cold enough, I don't think so.
03:03 - No, you just gotta, well, maybe it would be,
03:05 maybe it would be to start.
03:07 You gotta try.
03:08 - Okay, I will.
03:09 All right, well, let's get into it, Ryan.
03:11 So I kind of teased Kaiju again.
03:14 There may be some people joining us today
03:16 who this is their first time
03:17 and they've missed out on a lot.
03:19 They've missed out on a lot of good information
03:21 that we've talked about the past few times,
03:23 but for maybe some of those people,
03:24 you wanna just give the quick elevator 30 second pitch
03:28 on what Kaiju is, what you guys do?
03:31 - Sure, so for those who don't know us,
03:34 we're a global ecosystem of companies.
03:37 We use artificial intelligence
03:39 to build trading intellectual property, trading strategies.
03:44 And then if the strategies are possible to be used
03:49 in a 40 act space, so that's public dissemination,
03:53 we wrap them up as ETFs and we launch them that way.
03:56 And if they're not, we use them in private funds
03:59 that we manage on the private fund side.
04:01 So everything starts with Kaiju IP.
04:04 We build everything in-house.
04:05 We've been doing this for like over half a decade,
04:08 long before it was in the mainstream news cycle,
04:13 that is for sure.
04:14 And long enough to have been made fun of for it,
04:17 which I keep saying probably should stop.
04:19 And that's it, yeah.
04:20 When our flagship ETF launched last December,
04:23 it was the first fully curated and directed AI ETF
04:27 we launched on the New York Stock Exchange.
04:29 Moved down to the New York Stock Exchange floor in March,
04:32 only one of six ETFs on the NYSE floor.
04:36 And it's been good so far.
04:38 And we saw some pretty huge flows in
04:41 like the last couple of days,
04:43 which was nice to see after a successful future proof.
04:46 So many thanks to those RIAs
04:49 who came on board after long discussions.
04:53 - Well, there you go.
04:54 I'm glad the conference was productive.
04:58 So, Ryan, I'm glad you were able
05:01 to give a brief overview there
05:03 because my first question today is about the team.
05:06 And I'm sure, we've talked about this,
05:08 how at Kaiju, you guys have been using AI
05:12 and kind of in this technology for a while.
05:14 It's not like, there's a lot of firms
05:18 that are now saying, "Oh, we use AI,"
05:20 'cause they saw it get hot,
05:21 but you guys have been doing this for a while.
05:22 And your team is comprised of experts
05:26 from a very diverse backgrounds,
05:29 applied mathematics, theoretical physics and cosmology,
05:32 enterprise technology, and in your case, cryptography.
05:37 So how do all these,
05:39 having people from all these different areas work together,
05:42 how is that a strength for Kaiju?
05:44 - Well, I think there are a couple of different ways
05:49 to look at why that's advantageous.
05:53 Because of course, someone might think,
05:54 you're coming from these diverse backgrounds,
05:56 like how do you get on the same page?
05:57 Like you're sort of clearly not from the same discipline
06:00 and there's some overlap.
06:01 I mean, maths are part of everything that all of us do,
06:06 but where's that intersection and why is it beneficial?
06:11 Number one, I would say,
06:12 none of us were incubated in a financial space originally.
06:17 I mean, I was an equities trader eventually
06:20 after I was a cryptographer and then a portfolio manager
06:24 and a CIO fund manager before I became the CEO
06:29 of Global Kaiju, but I didn't start there.
06:33 I didn't come out of school
06:34 and like right onto the nisey floor
06:36 or the Merck or CBO or whatever it was.
06:38 And the same with everyone else who works there.
06:40 Everyone else works here.
06:41 And so we don't tend to look at problems
06:44 the way that people who've been incubated
06:46 in those environments look at problems.
06:48 I mean, that's a systemic issue, right?
06:51 When you have like the, take the Tiger Cubs, for example,
06:56 you have a couple of standouts there,
06:57 but you've had some lackluster performance
06:59 and you sort of wonder why.
07:01 And it's because there's a central ideology
07:05 that they're all exposed to
07:07 and they're all using the branching out slightly.
07:09 It's really hard to do something unique
07:11 when it's been bashed into your head over and over again.
07:14 And we're all susceptible to that.
07:16 So when you have somebody coming
07:17 from completely different discipline,
07:19 especially one that uses AI in a way
07:22 in which you haven't used it before.
07:23 I mean, we've all used AI before.
07:26 We came together in our own disciplines,
07:28 but not in the same way.
07:30 There are commonalities for sure.
07:32 We're solving different problems in novel ways.
07:35 So that's a huge strength.
07:37 Like I learn from the team all the time, every day.
07:40 Hopefully they learn from me.
07:41 I mean, one of the fun things that we do every quarter
07:44 is I fly up to London,
07:46 which is where the sort of core AI team is.
07:49 And we talk about trading mechanics,
07:53 and that's where we solve all of the mechanical problems
07:56 that exist when you're bringing a model strategy
08:00 into the real world.
08:01 It's like, great, we're super
08:03 in a validated model environment.
08:06 And it's absolutely possible.
08:08 Now let's see how this works real world.
08:11 And that can be a collision sometimes.
08:14 Like the machine's telling you to sell 1400 spreads
08:18 of like ISRG and the counterparty is like,
08:22 "Yeah, I don't want that."
08:23 The data scientist is like,
08:27 "What do you mean you don't want that?"
08:28 The machine thinks, "Yeah, that's the Theo value.
08:30 You should be able to sell that."
08:32 And you're like, "Yeah, he's just in a bad mood today.
08:35 Sorry."
08:36 And that has to be quantified and fed back into the machine.
08:39 So coming at it from different perspectives
08:41 help us work through those problems.
08:43 - Yeah, I mean, I find that fascinating.
08:46 I was wondering if it's one of those things
08:48 where it's almost like hard to get everyone
08:51 on the same page
08:52 'cause everyone's coming from these different backgrounds
08:54 and might not see things the exact same way.
08:57 But at the end of the day, I think as in nature, right?
09:02 Like in an environment, you want diversity,
09:05 you want different things working together.
09:08 And I could see how that could also be a strength for Kaiju.
09:12 One thing about the, for the dip ETF and a while ago, Ryan,
09:16 this may have seemed like a hypothetical question,
09:20 but how would the ETF prepare for a black swan event?
09:24 I mean, the COVID-19 crash,
09:25 we actually saw a black swan event,
09:28 which those don't come around too often.
09:32 I think that's the definition of a black swan event
09:34 is they don't come around too often.
09:35 - Yeah, exactly.
09:36 And if you haven't designed the ETF
09:39 specifically just to predate on black swan events,
09:42 then just like a human portfolio manager,
09:45 there's risk there.
09:46 I mean, I would love to be able to build a system
09:49 that made a lot of money in normal market conditions
09:53 and then, hey, miraculously identified a black swan event,
09:58 flipped the entire portfolio construction
10:00 and then made money when it collapsed or surged
10:03 or whatever it is.
10:04 But if I did that, I'd be on a very large yacht somewhere
10:09 and probably not doing this show
10:11 as much as I love it though, Aaron.
10:12 But what we do with black swan event prediction is,
10:17 we run these two engines at the broad market level,
10:23 sector and industry level
10:24 and the discrete underlying level.
10:26 And the two engines are regime classification
10:28 and regime change detection.
10:30 And so classification at any level we're running it at,
10:34 and if you've done the math,
10:35 there are like hundreds to thousands
10:37 of classification engines running from like Disney
10:41 and Netflix all the way up to sector, industry
10:45 and then broad market.
10:47 And classification is just trying to say,
10:49 okay, based on what I'm seeing in the moment,
10:51 real time at the tick level,
10:53 what regime do I think I'm in?
10:56 And it's bull bear neutral or I don't know.
11:01 And that matches up with decisions that it's going to make.
11:05 So let's take an example.
11:07 So I just mentioned Disney, we'll use Disney, fine.
11:10 So it identifies that Disney is suddenly shifted
11:15 from neutral to bear, from bull to bear.
11:19 It's gonna talk with the sector
11:22 and industry classification engines.
11:24 Is this like a sector and industry thing?
11:26 Is it a broad market thing or is it just Disney?
11:29 And that will help the machine make decisions
11:32 at those different levels.
11:33 Is it taking like broad protection or is it like,
11:36 oh, well, that thing's just tanking over there.
11:38 I guess we should manage that and move on to something else.
11:42 And then we have change detection.
11:43 Change detection is looking at the information
11:47 that's coming in at the tick level.
11:48 And instead of historically, where do I think I am?
11:52 It's based on what I'm seeing in the moment.
11:54 What do I think is coming next?
11:55 Is it a continuation?
11:56 I'm neutral, I'm gonna continue neutral.
11:58 I'm neutral and I'm bull.
12:00 Neutral, it's going to bear.
12:02 And there's a percent probability that it guesses it has
12:07 in terms of predictive capacity with that.
12:10 So the most interesting for us, of course, is I don't know.
12:14 And that's the black swan event, right?
12:16 And when it kicks back a whole series of I don't knows,
12:19 it means that in 220 terabytes of tick level data,
12:24 it's never seen this before ever.
12:27 It has no reasonable correlation with any pattern
12:32 that's existed in any market anywhere.
12:35 And that'll give you some pause, right?
12:39 If you look at what would it have said
12:41 as the market started to heave off the cliff
12:43 after Feb, 2020, you mentioned COVID.
12:47 So that was an I don't know what this is.
12:51 And what it at least allows you to do
12:53 is take protective action, right?
12:56 You're not just hitting your same favorite strategies
12:59 the whole way down.
13:00 Yeah, baby, this is gonna turn around.
13:01 Let's go, let's go.
13:03 You're gonna say, okay, obviously we're not going to try
13:07 and corridor anything when it's collapsing
13:09 through the floor.
13:10 You're getting exchange circuit breakers that are tripping.
13:13 You're getting global halts.
13:15 You're getting like, that's a whoa, hold on.
13:19 And it's one of the reasons that, you know,
13:21 especially in the dip ETF prospectus
13:23 and in private fund prospectuses,
13:25 you have to say stuff like,
13:27 or we may hold cash in certain circumstances.
13:31 Nobody wants their money not to work.
13:33 There's no trading strategy where it's like,
13:35 yeah, I'm not gonna put this money to work.
13:36 But you identify that if you have one of these moments,
13:40 you know, I don't think anyone wants you to pretend
13:42 that nothing's happening and you just keep going.
13:45 So, yeah.
13:48 - I was kind of curious about that.
13:50 Would there almost a number, you know,
13:53 not that the AI would be trained to do this,
13:56 but could there ever be a case where, you know,
14:00 say there's like a flash crash and something's going on
14:02 and the AI on its own wants to just sell everything?
14:06 'Cause I mean, at least I would figure
14:08 that wouldn't be actually like the worst, you know,
14:10 break glass in case of emergency
14:13 if there is some sort of like crazy event happening
14:17 and the AI doesn't recognize exactly what's going on.
14:20 I feel like that wouldn't be in all the realm of things
14:23 it could do, wouldn't be like the worst thing in the world.
14:26 - Yeah, no, and it depends what you program it to do.
14:29 You know, like if you're predating on specific signals,
14:33 they basically disappear and you get hit with it
14:35 across the board, I don't know,
14:37 return from all of these engines, you know,
14:40 690 different engines or 2200 engines,
14:43 depending on which sphere you're looking at,
14:45 suddenly all fire off, I don't know.
14:48 Like what it'll do is it'll look to mitigate risk,
14:51 close positions, and then it's gonna wait
14:54 until it does have a reasonable degree of certainty.
14:57 And it doesn't take very long for it to do that.
15:00 I mean, the regional banking collapse for us
15:02 was something the machine had never seen before
15:04 in the US in March.
15:06 So, you know, there's a collapse, markets go down,
15:08 dip traded lower than its benchmark did,
15:11 but then it took like two days for it to identify
15:15 what it was and said, oh, I got this.
15:17 You know, unfortunately at the time we had
15:19 sort of partner constraints that prevented us
15:22 from doing 100% rebalance.
15:23 That changed as of like three weeks ago.
15:26 So we can just rebalance completely.
15:28 But, you know, we have this historical information
15:30 that at that moment, it certainly wanted to.
15:33 It wanted to dump the entire portfolio
15:36 and reload with what it identified
15:38 was going to be profitable.
15:40 So, but you have to tell it to do that.
15:42 If you don't tell it to do that,
15:44 sometimes these things can just freeze.
15:46 And that comes down to like, you know,
15:48 I'm gonna be boring and be like a broken record here,
15:52 but that's the responsible AI, right?
15:54 That you and I have talked about before.
15:55 What's the difference between just, you know,
15:57 painting with the AI paintbrush and the responsible AI?
16:00 - Yeah, I guess, and that makes sense.
16:04 I guess I can just foresee a scenario
16:08 in which we're talking about a like black swan event,
16:11 where if, you know, the algorithms
16:15 are kind of getting these, I don't know, signals.
16:18 And like you said, it might not take it that long
16:20 to figure out, you know, what to do, what's going on.
16:22 But like, you'd want it to act quickly, I'd assume.
16:26 You know, you don't wanna have that like inaction
16:28 that could, you know, so I guess it'll just be interesting
16:31 to see over time how these AI things like deal
16:34 with these events.
16:34 'Cause I mean, going back to COVID,
16:37 I don't think like, when you hear how a lot
16:39 of professional traders and investors,
16:41 like a lot of them had never seen a market
16:44 respond the way to COVID,
16:45 and especially coming out of COVID,
16:47 like the recovery when the market recovered,
16:49 like a lot of people were like,
16:50 we've never seen anything like this.
16:52 So with different AI algorithms,
16:54 the way they're trained by things that they've seen before,
16:58 I could imagine it'll be like,
17:00 I'm just curious to see how it's gonna work,
17:02 you know, going through something for the very first time.
17:07 - I mean, it's seen a pandemic now, right?
17:09 I mean, like it's at least got one reference point,
17:13 you know, but when you say, you know,
17:14 a lot of investors or traders, I mean,
17:17 none of us had ever traded through a pandemic before, right?
17:19 That the last traders that have done so
17:21 are like freaking dead.
17:23 So, you know, that was 19, 19.
17:24 - But specifically, just not, you know,
17:26 pandemic or not, I think just like that,
17:28 you know, the crash, first of all, in March, 2020,
17:32 but then that rapid recovery from that V-shaped recovery,
17:35 how quickly like the S&P and then like,
17:37 specifically there were days,
17:39 I remember looking back on it,
17:41 like I think there was a day that Nasdaq closed up 7%
17:44 or something crazy on the day.
17:46 And they were like, you know, professional traders
17:48 that weren't bad traders or dumb people that were like,
17:51 you know, maybe shorting a little bit
17:52 because they were like, okay, this is getting out of hand.
17:54 The economy is closed.
17:55 And the next thing they knew,
17:56 they wake up the next morning,
17:57 the market's up another 4%.
18:00 So I mean, I think it was just a crazy, crazy time.
18:02 - So I could see, I don't know how much,
18:05 I mean, obviously there was a lot of quant trading
18:07 and algorithmic trading going on at the time,
18:09 which some of it was probably doing really well
18:11 and some of it probably wasn't doing well,
18:13 but I wonder now with the way AI technology has progressed,
18:17 even in like three years,
18:18 how AI would have been dealing
18:21 with those types of movements back then?
18:23 That would have been interesting to see.
18:25 - It's the slight, it's not, I don't wanna make a statement
18:29 that AI is substantially predictive
18:31 because it can help us try to identify
18:35 what we think we're seeing with a little more certainty.
18:38 There's not a crystal ball, it can't be a crystal ball.
18:41 But one of the reasons that we did well in that collapse
18:45 is that we were fully set up net long units short side
18:50 for that collapse.
18:52 Now, we did not know it was a pandemic,
18:54 like no freaking idea that this was coming.
18:58 We just in that week before the collapse,
19:02 we were getting a lot of I don't knows, like constantly.
19:06 And after a two month melt up, you look at the news cycle
19:10 and this bug is starting to make its way around the world
19:13 and it's starting to command more and more space.
19:16 So what's the safe thing to do?
19:18 Being that long South side.
19:20 So, you don't know why you're doing it,
19:24 but you know that that's the move to make
19:26 in this particular circumstance.
19:28 And then when the collapse came,
19:31 you're like 30 cents spreads are blowing up to like $25.
19:35 March is great for us.
19:36 And so it can do that, it can do that.
19:40 To your point though, the non dead cat, dead cat bounce
19:45 that came afterwards and like the rocket upwards.
19:48 Yeah, nobody, I had like safety nets another 20 points down
19:52 and I don't think anyone thought we were going
19:55 straight back up to a new all time high
19:57 in the next four weeks.
19:59 It was not a-
20:00 Yeah, I mean, I especially too,
20:01 once the market had recovered,
20:03 say like most of the way back to where it was,
20:06 then it was okay.
20:07 We're like basically back to where we were,
20:08 but the economy's still closed.
20:10 Cases look like they're, cases are rising.
20:13 Yeah, exactly.
20:14 Like how it's not gonna keep going up.
20:16 And then a bunch of, yeah.
20:17 And then so, I don't know.
20:18 I mean, it was like, as like we said, just a,
20:20 I don't know, at the time,
20:23 it was big to say it was unprecedented.
20:25 It truly was unprecedented, but now it's not.
20:28 Now we've been through it, like you said.
20:29 Now the AI's got that in the bank,
20:30 but it'll be interesting 'cause like the next time
20:33 there's an event like that, it'll,
20:35 you know, it's like history-
20:36 Market will react like that next time.
20:38 Right, history doesn't repeat itself, but it rhymes.
20:42 So there may be like things that are similar
20:44 to the COVID crash, but might not be the exact same.
20:47 One thing that I've always been like shocked at
20:48 looking back on the COVID crash is,
20:52 and I feel like this is kind of shows that
20:55 even with how advanced trading is and how smart,
20:59 you know, or maybe not smart a lot of people in Wall Street,
21:01 I don't know, but that it's still,
21:03 the market is still not completely efficient,
21:06 is that looking back on the COVID crash
21:09 and why the market crashed.
21:11 And, you know, I know hindsight is 20/20,
21:13 but the fact that stocks like Amazon
21:16 and stuff even dipped at all,
21:18 let alone like 20% on the onset
21:20 when the whole reason the market was crashing
21:22 was, okay, people are this disease.
21:24 You know, I just feel like it was weird.
21:25 People didn't have that foresight to be like,
21:27 well, if the market's crashing, who's going to benefit?
21:30 It's going to be Amazon,
21:31 'cause everyone's going to be at home
21:32 staying away from this disease.
21:34 But I could, you know, again, hindsight is 20/20.
21:37 Yeah, yeah, hindsight, like you didn't know that-
21:39 It was the uncertainty aspect.
21:41 You were going to be at home.
21:42 Were you going to be at home on a respirator?
21:43 I don't think you were ordering anything from Amazon
21:45 if that was the case.
21:46 What if the controls and restrictions were like,
21:49 well, I'm sorry, but food deliveries only,
21:52 you know, I'm fine.
21:53 So grocery gateway or, you know,
21:56 Whole Foods at home or whatever the hell it is,
21:59 they would benefit.
22:01 You know, Zoom certainly went on a tear, right?
22:03 But you know, in those early days, you just,
22:06 it was this really happening?
22:08 Is this the zombie apocalypse?
22:09 Like people just didn't know
22:11 what services were going to be interrupted
22:13 and then for how long, like, are you getting paid?
22:16 And, you know, I mean, you could boil it all down,
22:18 that crazy reversal to a preponderance of sports bettors
22:22 in the world, no sports,
22:25 and $1.2 trillion of stimulus money
22:28 that went to a lot of people who didn't actually need it.
22:30 So you get your 1200 bucks and you're like,
22:32 sweet, I'm still getting paid.
22:34 I'm going to gamble this on Tesla.
22:37 You know, I'm going to pump this into the stock market.
22:40 You know, there's no football, so I'm just going.
22:43 And that's like, you know, I mean, you see the result.
22:48 - Yeah, I will say personally,
22:50 I was in the crowd of people who like made more,
22:54 was making more money during COVID.
22:55 'Cause I never like stopped working.
22:57 I was working on a golf course in college
22:59 when COVID first hit, or I was not in college.
23:01 I was at, done with school,
23:03 but still in my college town where I went to school
23:07 and working at a golf course.
23:08 So we was like working outside,
23:09 never stopped working out of like a single day.
23:13 And they, I still got stimulus checks.
23:16 So I like was still getting my paycheck
23:18 and then on top of that.
23:19 So yeah, I'm definitely in that crowd
23:20 of someone who was getting more money,
23:23 didn't have any sports or anything to gamble on.
23:25 So it was in the markets a little more
23:27 than probably I would have been.
23:28 But now I'm here and it all worked out.
23:30 So.
23:31 - Give you a new career.
23:33 - Exactly.
23:34 So Ryan, Marco Argentia, I hope I'm saying his name right,
23:38 the chief information officer at Goldman Sachs
23:40 recently said in an interview
23:41 that there are two camps right now in AI.
23:45 The first is that AI helps you do things that you do today,
23:48 but faster and better.
23:50 And then the other campus saying that AI helps you do things
23:53 that you're just not capable of as a human being.
23:56 Which to me, I don't think those two things
23:59 necessarily have to be mutually exclusive, I guess.
24:02 But what camp would you say you were in
24:04 if you did have to pick one or the other?
24:06 - I would pick the not mutually exclusive camp for sure.
24:10 Those are, they really are just not mutually exclusive.
24:14 I mean, there are all kinds of things
24:15 that AI is gonna make a lot easier.
24:18 Generative AI already does that to a large extent.
24:22 It's gonna be better.
24:23 Imagine being able to use generative AI and say,
24:26 okay, so where, go look up 13 Ds
24:31 where insiders have increased their position
24:35 to more than X right before large moves have occurred.
24:41 Find out the insiders who have done that
24:44 prior to the most positive large moves.
24:47 Then give me a list of those insiders
24:50 who have done that in the last quarter.
24:53 You know, like that would take you forever
24:55 to look all that up.
24:57 You know, you have a research team of analysts,
24:59 but you could probably do that within like 15 minutes
25:02 using appropriate prompts with generative AI
25:05 and access to those materials.
25:07 So that's something, you're still gonna make the decision
25:09 at the end of the day,
25:10 the AI is gonna really help you cut the time from that.
25:14 And then, you know, on the flip side,
25:16 things that humans can't do.
25:19 Yeah, before we rebalance dip every day,
25:22 that AI conducts 2 billion discrete examinations.
25:27 Like I can't do that.
25:30 I don't know anyone that can do that.
25:32 You know, and I mean,
25:34 like we don't trade in a nanosecond timeframe,
25:36 but we absolutely analyze in that timescale.
25:38 So, you know, that's going to,
25:40 I'm supposed to say regulators require,
25:42 I say may exceed human capabilities,
25:45 but I mean, come on.
25:46 You know, there's stuff like that.
25:48 You know, the AI that flies your plane
25:51 through insane hurricane turbulence
25:54 is going to be able to do that
25:57 with more agility than a human pilot
26:00 because it's able to interpret the data
26:03 at a vastly superior scale.
26:06 That said, do you really want to just fly in a plane
26:09 where there's no person there,
26:10 just in case this thing decides
26:11 to not show up for work that day, there's a short?
26:14 No, like, you know, the entire air crew
26:16 is there to save your freaking life.
26:18 So, you know, it's a balance, but there's certainly,
26:21 there are directions where it'll go both ways,
26:23 but there are some that'll be really attractive early on.
26:27 Oh, have you seen the original art it can create?
26:29 And then you'll find it generally soulless
26:31 and that part of it will die out
26:33 because that's something that humans do really well.
26:36 Yeah, exactly.
26:37 Yeah, and exactly.
26:38 That's, I think, my point too, in saying that it's not,
26:41 you know, I think there are things that it can do faster,
26:44 but in that same route,
26:45 what like that speed itself is something that we can't do.
26:49 So yeah, I can analyze a hundred different stocks
26:52 for me in two minutes.
26:53 I can do that technically.
26:55 It would take me a lot longer than two minutes though.
26:58 Well, Ryan, so let's, you know,
27:01 we've gone a little bit over time today,
27:02 but it's been a great conversation.
27:04 So I'm glad, I just don't want to keep you too long
27:06 'cause I know you're a busy guy.
27:07 That's why we put up there and we just,
27:09 we ramble sometimes.
27:10 Sorry, folks.
27:11 It's all good.
27:13 It's like, we've got our own AI almost.
27:17 Yeah, and it goes in circles.
27:19 Like I'd take that one back to the lab, frankly.
27:23 Let's get over, let's go to our Kaiju kicker
27:28 and let's talk about exchange traded products today.
27:32 So, you know, we know about ETFs, exchange traded funds.
27:36 Well, let's talk about ETPs, exchange traded products.
27:38 What are the most important things retail traders
27:41 should know about ETPs before they consider trading them?
27:44 What are the key benefits of trading ETPs
27:48 and what are the biggest dangers?
27:49 Okay, that's a lot to unpack.
27:53 Let me see if I can do this in like three minutes.
27:55 Okay, so what should, ETPs, that's everything.
27:58 That's ETFs, ETNs, and a variety
28:02 of other structured products.
28:04 What is it that people should know before they trade them?
28:08 I think the one thing that I don't see a lot
28:12 in the retail space is an analysis
28:14 of what the core goal of the ETP is.
28:18 So when these things came along originally, right?
28:21 They weren't, ETFs just weren't constructed
28:23 for retail traders.
28:25 They were institutional hedging instruments.
28:27 So they were packaged together to serve a specific hedge need
28:32 against a human constructed portfolio.
28:36 That's how they, okay, so I want to hedge off my tech risk.
28:41 I'm gonna use the Qs.
28:43 I wanna hedge my AI or my chip risk.
28:47 I'm gonna use SMH, right?
28:49 Like at the sector ETFs, it was a way where
28:52 if a sector was getting hit and you had to make
28:54 a very agile intraday decision, you could pump capital
28:58 into positive or inverse ETFs with a specific construction
29:03 to instead of discreetly hedging all
29:05 of these individual positions.
29:07 Like, you know, our vol art portfolio can have
29:10 like 290 discreet positions.
29:12 If that all had to be hedged individually,
29:14 like there's no way our traders, awesome as they are,
29:18 could handle all of that.
29:19 And so you have these products, which are like, cool,
29:22 I will use 12 of these and that'll sort my whole book out
29:25 really, really fast.
29:27 The problem is if you don't understand what these things
29:30 are designed to do, and you're literally
29:32 just trading price action, then you're gonna get whipped
29:35 here and there by something that you didn't understand.
29:37 I hate to admit that I did this
29:41 when I was a professional trader, but, you know,
29:45 occasionally we would predate on retail using
29:49 some of these ETPs where they didn't know
29:52 what the product was for.
29:54 There were a pair of natural gas derivatives.
29:59 They were ETNs, so exchange traded notes.
30:03 They were Credit Suisse's 20 year forward looking
30:07 natural gas bonds.
30:10 But we became aware of the fact that most of the traders
30:13 who retail that were active in this thought
30:14 it was natural gas.
30:15 They thought it was a natural gas proxy.
30:17 And so you'd read in these Yahoo chats stuff like,
30:22 oh, I think it's gonna be a cold winter.
30:23 I'm gonna buy more of this.
30:25 And we sort of looked at it and went, holy shit,
30:28 these guys have no idea what this is or why they're,
30:31 the swings were crazy.
30:32 Well, the swings were crazy.
30:33 They were like, they would open up or down like 30%, 17%.
30:38 So you could make a killing if you were directionally right
30:41 on the close to open, which is how we traded them
30:44 because we were aware of how the bonds
30:46 were going to be priced.
30:48 So it was a little unfair, but I mean, God,
30:52 buyer beware in these conditions, right?
30:54 So understand that triple leveraged gold bull
30:58 or whatever it is you think has this sexy price action,
31:02 understand how it moves
31:04 and what its interdependencies are, right?
31:06 How many components are in these things?
31:08 You have like a short component list in tech.
31:12 You have overweight and I don't know, Nvidia and Apple,
31:15 and they both sell off.
31:16 It's going to take that ETF with it.
31:18 There's nothing wrong with the ETF.
31:20 It's just overweighted in a couple of areas.
31:22 So it's really important to understand
31:25 what the purpose of this thing is before you trade it.
31:29 I think that's, so I think that might've covered everything.
31:33 The pros, cons, risks, whatever.
31:35 - I mean, there you go guys.
31:37 I mean, if that's not like, you know,
31:38 kind of some inside baseball, some behind the scenes stuff,
31:42 pulling the curtain back, if you will,
31:43 then I don't know what is.
31:45 But some great insights for us, Ryan.
31:47 Thank you for sharing with us today.
31:50 But yeah, man, great conversation as always.
31:53 It's great to have you back on the show, Ryan.
31:56 And yeah, folks, if you guys want more,
31:59 where can they find you?
32:01 - Global Kaiju is kaiju.ai.
32:06 The ETF part of what we do is Kaiju ETF Advisors.
32:11 And if you're interested specifically in dip,
32:14 you can just go to dipetf.com.
32:16 I think that covers everything for now.
32:18 - Beautiful, dropping those links in the chat, y'all.
32:21 If you guys want to go check that out,
32:23 check out the dip ETF ticker DIP.
32:26 Ryan Pinnell, as always,
32:27 thank you for joining us on Benzinga Live.
32:30 Enjoy the rest of your Tuesday.
32:31 - Always a pleasure.
32:32 Have a good one.