Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
AMA w/ Chief Research Officer @DrSonicScholz
Recorded: July 21, 2025 Duration: 1:00:09Space Recording
Short Summary
Sonic is making waves in the blockchain space with its ambitious plans for growth and innovation, including the integration of AI in smart contracts and a strategic focus on enhancing its technological capabilities. The recent transition of Dr. Bernard Charles to Chief Research Officer marks a pivotal project launch, while partnerships with tech giants and a commitment to addressing industry trends position Sonic as a leader in the evolving blockchain landscape.
Full Transcription
Thank you. Thank you. Thank you. Thank you. Hello, can you guys hear us?
Hi, can you guys hear us?
Is that a yes?
Yes, yes, yes it is.
Oh, we can hear you.
Okay, awesome. Hello everyone, yes, yes, it is. Oh, we can hear you. Okay, awesome.
Hello, everyone.
Thank you very much for joining.
I am here joined by Dr. Bernard Charles.
Thank you very much for having me.
Great pleasure to meet you.
I'd like to start this conversation rather than asking you the typical question of,
can you tell me a little bit more about yourself?
And knowing from listening to other interviews, how humble you are,
would you allow me to sort of tell you what I know about you?
I have like three pages of notes, but I'm not going to refer to them.
I'm going to use my brain.
And please tell me if I misrepresent you in any way, or if I sort of forget something.
So Bernard, you've started university.
You've started your undergrad studies at the Vienna University of Technology in 1997.
You've graduated from that university in 2001 with a PhD.
In the meantime, you've worked there between 1999 and 2003 as an associate professor?
Well, I said an assistant professor.
Assistant professor.
And thanks for having me here.
Then in 2004, you started at the University of Sydney in February 2004,
and you've worked your way up pretty quickly. Now, what most people know is that you've worked as an associate.
You started as an associate, worked your way up to a professor.
But what people don't know, like me,
is that you've occupied really interesting positions,
such as director of research and chair of research,
and you've sat on executive committees,
such as the research committee,
and that was up to 2022.
If I may interrupt.
Yes, of course.
Academic life is very, how can I say,
secret to many people outside of the academic world,
and it's very hard to understand what's going on.
So in 2004, I started really as a junior academic at the University of Sydney as a lecturer.
And until 2022, when I took leave, starting the career here at Phantom Foundation Sonic Labs.
And when you're an academic at any university, you have basically three roles. You have a teaching
role, so you need to educate undergrad and postgrad students. You have a research role you need to advance human knowledge and the third role is an
administrative role so when you start you have small roles and then when you you know pro in
your career and you step up the ladder then you get bigger and bigger roles that's very interesting
especially that you've also worked in parallel between, I guess,
starting from 2008 when you started at Sun Microsystems here in Australia. Over there,
you started your own lab called, I guess, the Sun Labs, where working there, you've created this
tool, which I found fascinating, called Parfait. Parfait is a French word for perfect.
What the tool does is it scans static code,
a massive amount of code to sort of find bugs and vulnerabilities.
Then in 2010, that Sun Microsystems was acquired by Oracle Labs,
and that's when you started working for Oracle Labs.
You're very well known, if someone looks you up for in 2016,
you've created Souffle, which is a programming language.
My understanding of it, which is very limited,
is that it's also focused on bug finding, compiling, and so on,
but it does other things too.
And the very interesting thing is that what you've done the technology that you've
created is still being used right now by very big companies such as directly or indirectly such as
aws even the technology that we have on our iphones right now is thanks to you and thanks to the work
that that you've done so for example grammar tech which is a private big tech company in the US,
has contracts with the DOD and even DARPA.
From what I understand, from what I've seen online,
they also use what you've created.
You've also recently published a paper on DAG verification,
which is to make sure that there's,
to prove mathematically that the protocol is safe.
And it's something that you've presented
at a NASA symposium early this year.
You've co-authored this paper.
Now, if we go back, travel back to 2016, 2017,
that's when you met Michael.
And I guess the reason why I brought this up and I found it fascinating is all of this
translated into what you've done, what you're doing now on smart contracts and the blockchain.
So you've started with debugging smart contracts and you do have a tool for that that you've created too
then with what you've said in previous interviews sort of michael got you into the the blockchain
world and here you are now in 2022 you took leave from the university of sydney and you started
working full-time as a chief research officer for Sonic, which we're very proud and happy to have
you, to have someone your caliber amongst us. But I just wanted to give this introduction because a
lot of time people don't see the very, the achievements that you've done and the side to you
that a lot of times are not covered in previous interviews. So did I miss anything?
Did I misrepresent anything?
I think it's, yeah.
So I think this is roughly correct.
So I think that the main reason why I came into blockchain
was my security work starting with Sun Microsystems
and Oracle Labs.
And there we developed this new programming language,
which is a logic programming language and it allows us to
express security properties and programs and can deduce
potential problems in programs.
And we had this project 2016-17 to really apply this technology for
smart contracts. That was something very new, something amazing in those days. And we wanted
to see whether it's working. I was very hesitant. I was more in the old world of Java and C++. But my students, there was a bunch of very talented students in those days,
really pushed me into this space. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Can you hear let us know if you can hear us?
Oh, you can.
Awesome.
All right.
Sorry, we just had technical difficulties.
I'm not sure what happened with my other microphone.
I was using my phone, but it did come with a plan B, so it's okay.
So sorry to interrupt you, Dr. Schultz. So you were talking about the team that you currently have, and if you can please continue.
Yes, I have been very lucky working with very talented and bright students in the past and postdocs, and we were able to hire them. And so we have now this amazing team of very well-educated people and people who used to work in big tech like Amazon, Google, Microsoft.
Amazon, Google, Microsoft.
And that really gives us an edge in the sense
that they are knowing how to cut software,
but also able to think about what they're doing
and do research on this topic.
Awesome, thank you so much.
The next question that I have, and this is, so that's regarding the next gen roadmap. And you've mentioned that Sonic is already planning the next gen, this could be a new chain or could be a major upgrade in the next two to three years. lifecycle of the blockchain tech shortened and is the roadmap sort of adding to the nascent
capabilities of Sonic? So things that are missing right now, such as maybe built-in quantum resistant
encryption or some sort of an AI blockchain fusion, or is it to keep Sonic on the edge,
sort of pioneering the future? Because you've mentioned before that
competition are eventually going to catch up, whether it's now or whether it's later.
So what are the reasons why we're doing this, although Sonic just launched six months ago?
Well, it's an arms race. So we have major advances in the virtual machine and database so we
are really here leading. Soon we are productizing a new consensus protocol, a
DEC based one, which also makes our chain faster but the competition is not sleeping. So we need to prepare for the next big push.
And unfortunately, technology takes time to develop.
So what we are thinking of doing and then doing, there is always a time span.
And what we have noticed for us, it's about two to three years.
What we have noticed for us, it's about two to three years.
And this doesn't mean that our current sonic technology will stagnate.
So we will still further improve the current tech stack.
As we speak, consensus will come out.
Refactoring will come out.
Some additional features like lightweight client and fasting will come out in in
the next few releases but beside this we also need to plan for the next big thing
and that will require and refactoring and new software architecture of our
client whether it will be a new chain or whether it's just an upgrade
of the current blockchain, that is really up to the business part of the company. This is not our
decision. We are responsible for the technology and we need to plan what will happen in the next
two three years now rather than in two three years because then it might be too late
two, three years now rather than in two, three years because then it might be too late.
it's good to know that uh sonic is always on the sort of leading edge and it's not that you guys
are not taking a step back and saying okay cool yeah we're uh we're the fastest blockchain so now
let's uh hang back and enjoy the fruits of our work but instead you're also planning in advance
which is uh which is great.
Now, the next question that I have is pretty interesting, and that's regarding L2s.
In general, I've heard your views and Andre's views and even Matt's views on L2s that they
might not be needed or that Sonic sort of, as it is, is enough, is sufficient,
and there's no need to have L2s.
However, with recent developments,
with the recent legislative changes
where now institutions are sort of jumping on,
they're looking for permissioned change
for their own environment where they can control
things like maybe gas token, they could have their own KYC for the wallets
that they have on there.
And the question that I have is,
what technical tools or features are currently missing
to make the chain more developer friendly on that front?
So the reason why the these robin hood
stripe paypal from my understanding are going to arbitram is because it makes it easier for
developers to develop their own it's an l3 on top of the l2 uh arbitram um so they have an an a tech
stack to do that called orbit it's a like a sort of modular approach to doing it what do we
have what are the tools that are present in sonic right now that are developer friendly to serve
this purpose or are there any future upgrades or sort of roadmap to close those gaps and attract
these kind of builders and these kind of institutions. So
currently, I mean our view on L2s is firm so we still don't believe in them. We
think that current state-of-the-art chains are fast enough and we feel that
L2s are really just a band-aid for the shortcomings of Ethereum in terms of performance but as you said
before L3's there might be a space for permissioned blockchains especially for
institutional needs and there we are thinking about it so we have an active
discussion in the company should we build a permission
blockchain on top of sonic from a technical viewpoint we have not enabled block blobs at
the moment so blobs is sort of a special format for blocks to help l2s running on L1s. Sonic was never intended to facilitate L2s so we didn't
enable it. But to make these permission chains possible on Sonic we would need to build you know support for this and then also we would need to build a
you know roll ups or whatever set k technology based permission chain on top it's really a
business decision it's not the question of whether it's possible or not that a sonic
as a company sonic labs as a company wants to go into this
permissioned space and that really has a lot of problems as well. We would
need compliance questions to be answered and it's really a business decision and
I really leave it to Michael and Andre so this kind of question is better to
ask Michael and Andre because they're they
they give the direction of the business and if needed Sonic will definitely be able to
have these kind of features to host L3s but it's really more business decision now whether we want
to go into this space or not. And that makes sense.
It's good to know that from your perspective, just from a technical perspective,
it's something that Sonic can definitely cater for,
but it's more of a high-level business decision.
On the same vein, so just for the same question,
what do you think only from a technical perspective right now might be missing from Sonic that could attract more developers, but not just developers developing, let's say, DeFi apps, but more institutions coming in?
What do you think at the moment, just even if we put the business aside from a technical perspective, what is it that could draw more institutions in? Well, if we talk about an L3, so what's missing is really from a technological
viewpoint, we need a block processing unit, which then can do the storage of the transactions onto Sonic and the tight integration of it and
that may take half a year or a year of development doing so we would need to
set up a team and with that we would have you know you mentioned orbit before
we would have been similar technology but we would tightly integrate it with the sonic l1
to not make it just an arbitrary l3 which can run on any ethereum compatible chain we would really
tightly integrate it into the l1 so that even the l3 can shine with super fast performance and
high transactions per second and low time to finality.
Speaking of the fast change, so again transaction per second finality and which
Sonic is very well known for. I've got a question at the moment. I've heard it mentioned in a few instances, and I think one
of them was actually by Andre. Is it true that Sonic speed at the moment is breaking current
apps? So apps that are, for example, maybe expanding into Sonic. They already have an app,
they're expanding into Sonic, and Sonic is so fast that
it's breaking their app. And the actual question is, now with the next upgrade that's coming,
which is CS 2.0, makes Sonic twice faster, do you see this as a hindrance to attracting apps on other chains coming to Sonic?
I wouldn't say this is a hindrance but it's very sort of interesting that we see that
the chain itself is not the problem it's now the D apps which need to catch up and they need to use better technology in form of maybe more beefier servers and they need to
adjust the graphical user interfaces because they're more tuned to change like aetherium
and so there is a little bit of work required from the dApp side but what I noticed and this is I met a person running a DAX in Vienna. There was a sonic
Summit in in Vienna recently and I met a person there and he was saying
It's so wonderful, but we have this problem that the graph queries. So it's sort of a
so it's sort of a standard technology to utilize data management for DApps that
standard technology to utilize
these graph queries the subgraph queries are too slow and what can they do they
can't catch up I mean with the speed of the chain and so it seems like this is
sort of a standard problem and so craft technology needs to improve to keep
up with modern chains like sonic awesome thank you for that I just want to acknowledge that one of the
most popular questions that was asked is a bridge to Bitcoin and I understand from the offline conversation that we've had that this is
doable but it's more of a business decision to be made so I'm not going to go there because
I guess I'm asking the question to the technical person when this should be a business decision.
Now in the last interview
that I've watched with Blockmates, which it was actually a really good interview,
you've spoken about the utility of a chain and you said something that it
should be beyond DeFi or just tokenomics or people pumping a token, you were looking more of the future of
L1s. And one of the things that you've mentioned was real world asset, which is a sort of an
acronym or an expression that's thrown around now with people not even knowing what it is.
But you've mentioned something specific, which is land titles, registration management.
And you've mentioned how you could get the custodians, so maybe the government to trust
the blockchain and the blockchain technology, A a and maybe a specific blockchain technology
how do you think sonic could position itself moving forward to be the chain that
Governments let's say adopt in in this this area
well from a business viewpoint and and
Compliance viewpoint this this is really a can of worms.
So you have to go really full out.
You need to align yourself with government.
And so, for example, in Australia, there are very special research projects just doing this.
They're called CRCs.
And the idea here is that you cozy up with government
you cozy up with decision makers there but that's only one side of the coin and and you need
compliance certification so you need to go through all these processes so this is the technical
organizational part of it and I feel that's really the big challenge there but and I really hope that blockchains go beyond DeFi and
web-vi at the moment because the technology is wonderful I mean human
humans need this kind of technology in future so the idea there is that in the
past people mainly trusted institutions like you go to an Australian
bank like Commonwealth Bank or a National Australian Bank and you trust the bank that they
look after your money but it's a trust you give the bank but the bank cannot prove it except
historically because they can say over the last few decades we had millions
of customers and they didn't lose their money or only a few percentage has issues and so through
historical data you you can basically through history you can build trust but you have no
active way of building trust in these institutions and now with blockchains we
have a technology which is basically a trust machine you put energy in and trust
comes out trust graded through mathematics through cryptography and
the idea now is how can we utilize this trust machine beyond d5 beyond finance and there are
so many applications in law going from the execution of bills when people die execution
of simple legal transactions and all this can be could be done in on blockchain it's just it hasn't
materialized and it would be very interesting as a question to the community what can be done so
that we can widen our applications and improve the utility of the technology this is more the general question but more specifically
what can be done from the technological viewpoint is that the software building
a blockchain is unfortunately buggy whatever you do in terms of programming
is buggy unless you follow very strict protocol to produce software and these
kind of ideas are not new ideas I mean there were always mission critical
systems in the past using software like flight controllers if you fly an
airplane you you don't want buggy software on your flight controller. So way back people understood that software needs to be proven to be correct and not just functionally correct.
So that when the task is to compute 2 plus 2 that you always get 4 as a result.
You also want certain requirements on performance. example a flight controller that's a
good example you want when you have a sensor reading that the sensor reading
ends up as an activation of a flap in a certain time period and if you don't do
this the plane will crash and here in finance or in in our
blockchain world we have something else it's called time to finality we want
that then a transaction is submitted into the system that it settles in a
certain time period and from these regulation viewpoints that maybe you
really want to have enforced this and say it must be
settled in less than a second for example or you need to prove that when
you send a transaction into the system it always ends up onto the ledger so
there are certain questions I guess legislators would like to seen answers that software is compliant it's
not a new question it has been around for a long time but for the blockchain
world it might be fairly new as a challenge and so just a follow-up for you
when we say compliance in this particular context in the context of
RWAs, my understanding is we don't know what compliance means at the moment right now.
That hasn't been defined and so just before I give you the microphone back, the other
question that I have so that's from a technical perspective.
Actually, no, I'm not going to ask you about the business perspective
because I just realized that you're not the right person to ask.
But yeah, this is just for compliance-wise.
So the intersection between compliance and technology,
my understanding is at the moment it's not well-defined for us
to know how to be compliant.
Is that correct?
Yes, I would say so so for example in in certain industries they have certified
processes which you need to follow to ensure that you're compliant at the
moment we don't have these compliance procedures in blockchains and we are very far
away from producing blockchain software having certain norms and regulations
just a quick question so to i've got a massive list here but just to take a a um to deviate
from these questions what what's the one thing that you're excited for
tech wise for sonic that's coming up either in the near future or the long-term future
so i mean it's a whole range of things we are working on so we have new technologies like um
we had we talked about next gen so my team team reported that they have started on a new database.
So we have the current database is S5, so the fifth iteration, which then really got into the product.
And so now we are working on S6, a new database, which then further pushes the limits,
what's possible in terms of block
processing then they have been talking about you know potential new virtual
machines we are talking about consensus so these are technologies we which we
actively work on and lightweight clients and there we look maybe into
potential use of zero knowledge proofs so that that's also
super exciting getting into a really theoretical field of computer science and applying it
into the um you know l1 space but um that's one thing that's really technologies which we
sort of need to be stay competitive and and really deliver a good product
for the end users. The other aspect that we have put a lot of effort into
is proving the correctness of our software so it's really exciting using the latest formal methods of what's possible
in software. For example, we productized Sonic Gateway. It's the bridge between Ethereum first proven financial bridge using a formal semi-automated proof system.
So to our understanding, this is the first bridge which was mathematically proven to be correct,
at least the concept, the model.
And what's showing is that the equilibrium of balances is always kept, so
money is never lost. So either it has this fail, our bridge has this fail safety, so this means
when the bridge and sonic goes down, then the end user still can retrieve the balances, and so we,
this formal system,
proved that the balance is always intact.
Money is never lost.
And that's sort of really exciting,
using these really state-of-the-art software tools,
improving tools, to show the correctness of software.
Sorry, just to follow up on the bridge,
because that's something I actually didn't know.
Is that, I don't know how to ask this question,
but the bridge, currently the Sonic Ethereum bridge
is the technology that you're talking about.
I'm not sure how the framework or the technology work.
Could the same thing be applied to other blockchains?
And it's exactly the same thing where applied to other blockchains and it's exactly the same
thing where it's still the only mathematically proven bridge in the space that's that's uh that's
safe well it's it's from a conceptual viewpoint it's safe i mean there could be still implementation
errors and system problems which make it unsafe so it's not 100 safety but
at least conceptually we know that the financial bridge is working and it's quite specific this
proof system because it's specific for our fail safe. Other bridges don't have the fail safety, so they can't really
guarantee that money is never lost because they don't have this fail safety. So it's
quite specific to us, I would say. But this is only one aspect of proving correctness.
We have been applying formal methods to virtual machines we had a recent article on this as we
speak we even build now symbolic tools to check the correctness of the specification of the virtual
machine we want to also apply this to databases and we have been doing this for the consensus protocols with the conjunction of
in the French National Research Organization and logicians here at the University of Sydney we have
been working on proving the safety of consensus and we had a recent peer-reviewed publication on this and so we really try to push hard
besides new technologies also these formal methods and correctness
you know
technologies so that we can show not just by testing and auditing that our software is really working
and this is towards making sure
that we will long-term be compliant.
Also, I've got so many follow-up questions on this,
but then I wanna do the rest justice
and also keeping in mind the time.
I'm only gonna ask two more questions, or maybe three.
One of the community questions I thought was quite interesting
was to do with latency and the node.
So it says, will Sonic pursue geolocation-aware nodes placement
or a smaller set, highly trusted, low-latency validators?
And the question is particularly targeted towards
for Web3 gaming and to reduce latency.
Yeah, so it's this question about time to finality.
It's...
And the geolocation of the nodes,
if something that plays a role, I don't know.
It could be.
So, I mean, it's all about, I mean, this is a game for how long does it take to send messages
around the globe.
And it really is a question where validators are located. located so it's really how quickly can transactions be submitted and end up onto
the ledger it's this time span and at the moment we are in the sub-second
finality with a distributed set of validators and the question is can this be lowered and the question is
yes I mean the answer is yes we can lower it we can improve the protocols
and the consensus machineries to make it lower much lower but it it requires but the new consensus is coming
out so it makes this already faster these the machinery but in in future we
can it really envision making it much lower but this is really now more a question into the next
generation technology so we would need protocol changes for say let's say we
want to reduce it from sub seconds to hundred milliseconds so in order of
magnitude faster and then we need really a complete new protocol and maybe a new way of
thinking about blockchains to make it say one to two magnitudes faster because we get here now
really close to the limits of what the internet can deliver so to answer this, either the internet gets 10 to 100 times faster, or we need to improve
our consensus protocol by sending less information across the globe.
At the second last question, this actually tailgates really nicely into what you've mentioned
previously, which is the next tech stack that you're working on.
And you've asked the community multiple times
for import for some sort of,
because you're going through requirements
gathering exercise, I guess.
And it's really admirable that you and your team
are reaching out to the community
rather than just doing your own thing and asking,
well, what are the things that we asking, well, what are the things
that we can change and what are the things that you want to see in the future? On that note,
if I flip the script and ask you this question, when taking into account the future of blockchains
and where you see, or specifically the future of, let's say,
L1s and where you see us heading in five to 10 years? Because I don't think that we're going to
stay the way that we are, where we're so fragmented and all over the place. And, you know, everyone,
there's a lot of L1s. How would you see, if you were writing your own requirements, if you were
the customer, what are the things that you would write in those requirements to align with the future of the blockchain?
So, I mean, there are so many aspects to blockchain technology.
We talked about DApp technology and we talked about L1 blockchain technology we talk about a little bit
about Solidity and AI programming blockchains so there are a lot of
different parts of what needs to change in future to make it really fast and and better for people to use so I see in these three spaces different
directions so for the blockchain itself it's really a game of the throughput so
current technology was aimed for hundred million transactions per day we succeed that limit so we are much better than
what we aimed for but the question is a visa can do maybe on a day 2 billion and if we are generous
maybe we can say we have maybe 10 billion of financial transactions per day and can we build
technology which can cover all financial transactions on the globe in a single
day so that kind of question would be very fascinating can be built the
internet of finances in future and that that will be sort of interesting from a technological
viewpoint the other question is latency or time to finality so how long does it
take to settle a transaction how long does it take from submitting a
transaction so that it comes shows up on the ledger if we have such vast volumes
we most likely need complete new different ways of
thinking about what is a ledger that just might be partitioned into
geographic regions and there might be and also applications so certain
applications may not talk to each other so for example gaming may not talk to
DeFi and vice versa and so if we have sort of
small independent groups or partitions of the blockchain that may help the
speed and and the latency this is more the the blockchain technology the
question now is can the existing technology just be improved or do we need complete new technology there are also
big pushes into from from the outside like quantum computers we don't know whether this will work or
not quantum computers have been very promising for decades now but there are some physical limitations
to build quantum computers at the moment so it's all about noise and can be
really built quantum computers at large scale and at the moment we can't but
let's assume we can then blockchains need a new cryptographic library being sure that with quantum
computers existing yet that they cannot be hacked basically the blockchain so
quantum computers have the potential that they can hack cryptographic codes
from an implementer viewpoint it could be just saying or we swap a
cryptographic library but but there are other implications and that needs to be
started so that's this there's another big push from the outside it's AI I
would not see AI directly in the blockchain technology but I see a big use case for programming the apps and their AI will be most likely really very
dominant in the next few years the reason is that people want to go to the market as quickly as possible they want to explain to the
computer in natural language what the D app should do and then hopefully the AI
is able to translate the requirements the natural language requirements to
Solidity code or any other kind of programming language which then gets executed on the
blockchain but the problem here is that AI is a very brittle so they are not
very solid or robust technologies and so what we would need is most likely a
interesting interaction between what the user wants and explains the computer how it should
work but the user may want to also express guarantees about the DApp so
some invariance some conditions and then maybe with mathematical tools, we can then show that it's like a testing.
You do some calculation given as a math exercise in a math class,
and then you have a solution, but then you want to apply a test
that the solution is really correct.
And similar to programming, we want to know that when an AI produces something
that what the AI produced really delivered what we wanted initially and so there are big questions
related to AI I see AI really in that space can really dominate the apps and let the apps flourish and that's really needed we
need more D apps in future otherwise why do we build these big blockchain
technologies and so here we we see this big gain here and yeah so I mean
different parts of this system requires different new directions we are mainly in the
blockchain front here so we are mainly pushing what's possible with the
blockchains but the programmability of blockchains that that's clearly another
big big challenge and we need to adjust about here. And also in the company, we have activities to use AI
to produce smart contracts.
Zach has been doing amazing work recently
and showcased it recently in Korea.
So that's also very wonderful what's happening there.
So that's also very wonderful what's happening there.
You see the AI just sitting on the application layer,
not on the infrastructure, not on the chain layer.
Is there anything that the chain could do moving forward to,
I don't know, cater or maybe to encourage AI developers coming in?
I don't know, cater or maybe to encourage AI developers coming in?
Yes, I mean, we would need to grow our ecosystem
and make it more AI friendly in terms of, you know,
DApp programming environments, what Zag has started
environment, what ZAC has started and we will continue to do.
and we will continue to do.
And that's definitely a focus.
In future, I would say we, I mean, it's unlikely that in the blockchain space itself,
what I mean here is a consensus database, virtual machine, that AI will take a role. Yes, like it or not, there is AI in
there, but it's mainly from the programming viewpoint. So our guys use Copilot. They don't
write code with Copilot for a virtual machine or database. That would be very dangerous because
or a virtual machine or database, that would be very dangerous because AIs have unfortunately this problem of hallucinations.
So they just pretend to know, but they don't know. And that's a big issue.
And so in that space, we can't allow flaky code. It needs to be top code and very solid code.
and very solid code.
But we use it, for example, to write unit tests
to help programmers with repetitive tasks.
So they use Copilot,
but that's the only extent how we use AI at the moment.
Yeah, that makes sense.
The hallucination issue is still there.
It doesn't matter how many new models are released.
You still notice it.
Just being very respectful of your time,'m going to take more of your time. One really last
question. It has nothing to do with the technical aspect, just for us as a community to get to know
you a little bit more. For me, I like to know people's hobbies and things they do outside of work. So I've noticed you're active, you like walking, you like cycling.
I think it's a nice way for people that can't see you.
I can see you now, but people who can't see you to connect with you.
What are the things that you enjoy doing outside of the blockchain space,
blockchain space outside of work, outside of the professional or academic settings?
outside of work, outside of the professional or academic settings?
Well, first of all, when I worked for university, there was very little spare time for me.
But I'm getting older and I'm more aware that you really have to have a healthy lifestyle. And so I started as a member
of the local river canoeing club here in Merrickville.
So you can see me on the border occasionally,
or once a week, and I really enjoy this.
So being out on the border, but similarly, I like hiking.
I like going up to the Blue Mountains. It's a
wonderful national park, very close to Sydney, with amazing cliffs and views. And this is how I
spend my time with my kids and my wife. Thank you very much, Bernard. Thank you very much for your
time today. Really appreciate it. Thank you very much for everyone who joined us today.
Hopefully this is not the last time that we do this.
I'm pretty sure with future upgrades that are coming up,
I'm more than happy to continue doing it.
Thanks all.
And oh yeah, sorry about the technical difficulty
that we've had before, but glad that that was resolved.
All right, everyone have a good day.
