Latest / The Future Herd / 31: Dark Data and Edge Computing Are Reshaping Canadian Farm Strategy

Transcript

31: Dark Data and Edge Computing Are Reshaping Canadian Farm Strategy · The Future Herd

0:01
Jesse Hirsh: Hi, I'm Jesse Hirsch.
0:03
Welcome to the Future Herd.
0:15
Today's episode is exceptional.
0:17
And, uh, granted, part of my job as host is to use this part of the
0:22
intro, to hype it up and tell you why listening to this is worth your time.
0:28
But in this case, I cannot overestimate the, the knowledge, the insights,
0:34
the dynamism, the brilliance of the people participating in this podcast.
0:39
Talking about a really contentious subject.
0:42
Artificial intelligence, which is a catchall, is a whole bunch of technologies
0:47
and tools fit under that spectrum.
0:50
It's something people have a lot of anxiety about, a lot of curiosity,
0:54
we hope, but also sometimes fear.
0:58
So today we really tried to not only break down what AI is, uh, how AI models are
1:05
trained, what edge computing is, but the business, uh, case, the, the, the larger
1:12
ecosystem and the way in which agriculture is particularly poised to be a beneficiary
1:19
of the AI revolution, especially the ag agentic revolution, IE the rise of agents.
1:26
Now, if all of this is a kind of buzzword bingo for you, which as a regular
1:31
listener, I would hope that it isn't, uh, this nonetheless is an advanced
1:37
class where we've taken the three, uh, top minds around AI and agriculture in
1:44
Canada and have tasked them with coming together to, to answer not just the basic
1:49
questions, but some of the more advanced when it comes to literacy, when it comes
1:54
to data governance, data sharing, and some of the challenges to sovereignty,
2:00
uh, not just around agriculture, but around the use of AI and society.
2:04
So this is, I think, one of those episodes that demonstrates why the future heard
2:10
exists as a cross-sectoral collaboration that brings together different actors,
2:15
different voices in, in a way that isn't happening anywhere else in
2:21
society or, or maybe around the world.
2:24
So again, I hope you enjoy it.
2:26
Certainly this is the kind of work that inspires me and it really gets into the
2:31
substance as to why we should be curious about AI rather than just close-minded.
2:37
Alright, let's enjoy the chat.
2:40
I.
2:42
Welcome all of you back to the Future Herd for what is on the technical level,
2:47
a panel episode, but in reality is an all-star episode as we've been able to
2:53
gather some of the greatest, uh, brightest minds in the agricultural sector, but
3:00
also in the AI sector in that, uh, the three of you, the five of us, are a
3:07
bunch of heavyweights when it comes to dealing with these particular issues.
3:11
And of course, I, uh, much prefer in the age of Google, in the age of
3:16
LLMs to forego any introductions.
3:19
Our listeners, as they listen, can do their due diligence
3:23
and find out who you are.
3:25
I really wanna take the most of the time we have together and, and really
3:30
try to get a sense of the advanced applications of the technology
3:37
in the broader public context.
3:39
And I say that because I personally feel torn every day, that on the one
3:46
hand I am experiencing such incredible technology that is allowing my brain
3:52
to go places I never thought possible.
3:55
And yet I'm surrounded by a public that seems afraid, seems angry, seems
4:01
opposed to the tools that I'm using.
4:05
So I kind of wanted to start with a general question of, you know, I
4:09
find the three of you individually and Sectorally doing amazing
4:14
work with amazing technology.
4:17
How are you wrestling with the public debate, whether in the abstract or
4:21
whether in your family, whether in your community, whether people are like,
4:25
what are you doing with that AI stuff?
4:28
How do you translate it in a way that, uh, cuts through the
4:32
noise, cuts through the hype?
4:34
And I'm saying this as an intro question.
4:37
' cause unfortunately, I think even within agriculture there are the sceptics.
4:42
There are the doomers, there are the people who aren't really ready
4:46
to open their mind to this stuff.
4:48
And I think the three of you are exactly the right leaders
4:51
to, to help make that happen.
4:53
So, Mohamed, I'm gonna throw to you if only because I suspect that you
4:58
have experienced what I'm describing, uh, uh, given the travels that
5:02
you engage in, how do you frame this technology to the public?
5:07
How do you counter some of the disinformation and some
5:10
of the fear that is out there?
5:14
Mohamad Yaghi: So first off, thanks for inviting me onto this Allstar panel.
5:18
I wouldn't say I'm an Allstar myself, but no, of course.
5:21
My two colleagues here definitely have, um, you know, a, a storied
5:26
history, uh, within the sector.
5:27
So great to be on with both of them.
5:30
Um, when it comes to the, the conversation about artificial intelligence and
5:34
within the ag sector, uh, I was joking to my wife last week that, um, I feel
5:39
like I have a house in Toronto, but I'm always outside across the country.
5:43
Um, so I've been able to see like on the ground what's happening and, you know,
5:47
the application of these tools, person.
5:51
Um, and know, fortunately a lot of many provinces, um, and I think when
5:57
it comes to the application of it, I think for a long time it's always
6:02
been about what could be possible.
6:04
And I think this, the conversation is shifting towards what needs to happen
6:09
in order to shift how we frame it into a return on investment conversation.
6:14
And the reason for that is, is that when we tend to ask whether artificial
6:18
intelligence can save money, uh, I think the more honest question is
6:22
whether it helps producers make better decisions before capital is committed.
6:26
I think that's gonna be the huge breakthrough for the sector and what
6:29
we need because when we just look at, you know, the sector as itself, it
6:34
contributes over $149 billion to GDP.
6:38
Um, and that's not a marginal gain and artificial intelligence
6:41
can compound it over every acre.
6:44
But the proof of profitability is not just productivity metrics, it's
6:48
also looking at the labour shortages we see in the sector as well.
6:53
Um, I think a recent study by the Canadian Agricultural Human Resources
6:57
Council, um, they have way too long of a name, but I love them.
7:01
Uh, con I think 28,000 unfilled positions signified a $3.5 billion economic loss.
7:09
So ultimately the conversation I'm having with producers, some who are
7:13
using Claude to build their own apps, all the way to producers who are starting
7:17
to, you know, tinker around with it, it's all about return on investment.
7:21
It doesn't matter what this technology is unless it can
7:25
bring that to them at the moment.
7:26
And return on investment isn't just about profitability, it's also about,
7:30
um, you know, how do they manage their farms and can we manage that data on
7:34
premise within Canada and keep our data sovereign as well, and not try
7:38
to take Donald's points away from him.
7:40
But, uh, yeah.
7:43
Donald Killorn | PEIFA: Good setup though,
7:45
Mohamad Yaghi: I try my
7:46
Donald Killorn | PEIFA: and shout out Kark.
7:47
Shout out Former future heard guest.
7:48
Jennifer Wright, the executive director of Kark.
7:51
They are doing a great job.
7:53
Mohamad Yaghi: Yeah.
7:54
Jesse Hirsh: I mean, build upon that Donald, he clearly set you up.
7:57
I mean, to what extent is that ROI self-evident.
8:02
Donald Killorn | PEIFA: yeah, last week, maybe two weeks ago now, uh, my team was
8:07
on a roof in Eastern PEI, uh, installing a gateway and, uh, a wifi gateway and,
8:13
and basically creating the mesh network to bring the realtime data, uh, into
8:18
our AI platform that we need to mix with the meteorological and soil data to,
8:23
to try and do what mo's talking about.
8:25
And simultaneous to that, I was at the House of Commons, uh, presenting
8:29
to the standing committee on our agriculture compute strategy.
8:32
Uh, it's become pretty clear to us pretty quickly that precision agriculture
8:37
is becoming about what data needs to hit the, the main models and what
8:41
data can be used inside the field, boundaries in an inference model,
8:45
and matching hardware to software and, and where we can use inference.
8:50
Feels like the cutting edge this weekend.
8:52
Um, and agentic memory was sort of the cutting edge last weekend.
8:57
And so we're, that's kind of where we're at, uh, in the curve.
9:01
And, um, so yeah, if we try to move every single data piece around Prince Edward
9:05
Island, or pardon me around Canada, uh, we will need about 200 gigawatt hours of
9:10
electricity to run AI on Canadian farms.
9:13
But if we allow the system to learn and, and decide for itself what
9:18
data is required, uh, we believe we can get that number down to
9:22
about a gigawatt hour and a half.
9:24
Uh, so, uh, significant, uh, an order of magnitude less.
9:28
And those are the types of decisions and the types of strategies that
9:31
we need to be thinking about.
9:32
Um, it's difficult to find megawatt hours.
9:36
Um, we have to, you know, distribute our compute need across the country,
9:42
uh, but to most point ensure that, uh, agricultural data is treated with the
9:48
utmost, uh, security and, and sovereignty.
9:51
And, uh, the interesting thing, the most interesting thing about
9:54
working on that bottleneck while my team was sort of filling in the
9:57
blanks behind me was that the AI strategy came out that day at lunch.
10:02
And, uh, last week the food security strategy came out.
10:05
And the place where that overlaps is a tremendous space for research
10:10
and development and implementation and productivity gain in Canada.
10:15
Jesse Hirsh: Well, and I, I definitely want to come back.
10:18
Donald Killorn | PEIFA: Please.
10:18
Let's do it.
10:19
Mohamad Yaghi: jump in.
10:20
I wanna jump in.
10:20
I wanna jump in.
10:21
Sorry, Jesse.
10:22
Uh, Donald, I've, you know, I thought, I think like the, the way we're
10:26
positioning this discussion, I think is really important, and I appreciate
10:30
the efforts that you and your team are making, and I think when it looks, when
10:33
I look at the different federations of ag across the country, I think, you
10:38
know, you have a really interesting strategy, and I'm really curious when
10:41
it comes to that national sovereignty in terms of, you know, keeping data
10:45
within Canada or even the modelling capabilities for our producers.
10:49
How are you dealing with the OEMs, like the, um, you know, the equipment
10:53
manufacturers like Deer, case, new Holland, um, you know, like, are you, are
10:58
you speaking to them right now because they are some of the biggest providers of
11:02
that data, but I don't know how much of that information is within Canadian soil.
11:07
Um, and also when it comes to that utilisation of data, how is
11:11
it being used for our farmers?
11:13
I'm just maybe really curious, just because I was at a conference last
11:17
week speaking to other manufacturers, and that conversation was one of
11:21
the, I guess it was a, um, I'm trying to find the right word.
11:25
It was, um, it, it basically, it was something of like a elephant in the
11:31
room that nobody really was addressing.
11:33
Donald Killorn | PEIFA: Certainly one of, um, the best perks of getting on
11:39
this data governance train early a few years ago was being at the North
11:43
American European Union Conference in Cuomo, um, last summer, uh, in Italy.
11:48
And I was,
11:49
Mohamad Yaghi: pretty, that's a nice
11:50
Donald Killorn | PEIFA: it was, it was great.
11:51
And, and, and thank you to the Canadian Federation of Agriculture and, and
11:54
the PEI Federation of Agriculture for making that happen for me.
11:58
Um, and so I got to present and, and had a a, a lunch and learn,
12:02
and, and I was grateful to be joined by the, um, technical, the, the
12:07
technology lead for the Italian Equipment Manufacturing Association.
12:12
And he gave the room, myself included, a full briefing on the European regulations
12:17
that, um, compel these companies to, uh, provide, uh, true interoperability,
12:22
uh, in, in their platforms.
12:24
And so that's been, um, tackling, they're, they're tackling that through
12:29
an industry association, the A EF, uh, agricultural equipment fe uh, federation,
12:35
and they're developing a software tool called AG in, and they're ostensibly
12:40
doing that in North America and Europe.
12:43
Uh, now we've gone through the whole, um, the whole process to
12:50
achieve API compatibility, basically receive the agon connector.
12:55
Um, we've done everything on our part and sent them the check, and
13:00
we've sort of gotten, it's sort of, we're a little bit concerned now,
13:03
it seems like, you know, the wizard behind the curtain is, you know, not
13:08
necessarily what we thought it would be.
13:11
Um, know, and, and so it's an API integration that's necessary
13:17
to do what we're trying to do.
13:19
We believe the marketplace will allow that.
13:21
We, we, it seems like those companies are adopting a, um, a, a strategy
13:27
that allows 'em to be compliant to the European regulations in
13:30
the North American market as well.
13:32
And, uh, we believe we're creating an option for farmers to work with those
13:40
companies, but not be, um, locked into their, um, know, proprietary
13:46
data, data platforms, and that we will have interoperability in
13:50
Canadian agriculture between equipment manufacturers and that, um, farmers
13:55
will have a great deal of control over how that data is collected and used.
13:59
Jesse Hirsh: Well, and this is a good opportunity.
14:02
Jenn: is a
14:02
Jesse Hirsh: Let, let me just bring, let me bring Mosen into the conversation here.
14:05
'cause this is a good opportunity to both tie in the research angle in terms of,
14:10
I, I, I think Mosen, you're experiencing a, a different, uh, reaction from some
14:16
of the researchers that you're sharing your research with and you're sharing
14:19
your work from, to the extent that I think you're, you're, you're finding
14:22
more and more enthusiasm for, for the opportunities these tools provide.
14:27
But I'd love for you to also chime in on this data piece.
14:30
'cause obviously part of the process of research is data
14:34
collection and, and the, the kind of governance of the data collected.
14:39
So I'd, I'd love from the research perspective for you to wade
14:42
into this conversation before.
14:43
Jen.
14:44
We, we, we throw it to you.
14:48
Mohsen YN: That's good.
14:48
So basically, you know, just, I've actually enjoyed, you know, what,
14:53
uh, MOED and Donald talk about it.
14:55
But, uh, in terms of my experience, I have a kind of like day-to-day
15:00
experience with farmers, growers, industry partners and researchers, and
15:04
all of them have different ideas in terms of adapting new AI technologies.
15:09
Like for farmers, they are usually suspicious about is
15:13
it like the AI technologies?
15:15
Is it going to be a good, um, platform for decision making or how they can trust ai?
15:22
Is it really good for them on, even if they are going to, like, if AI in the near
15:27
future going to replace farmers or some of the jobs that they currently expert on it,
15:33
like these kind of things, uh, are like, are totally valid, are totally valid,
15:38
to be honest, and I really appreciate what farmers are bringing to me, but the
15:43
way that I'm usually addressing these kind of challenges is not by answering
15:48
directly the question that they ask, but providing them necessary information.
15:53
Like who said AI is going to be a good decision making?
15:58
Who said AI is going to replace your job?
16:01
I recently developed an app that is not going to make a decision for you.
16:05
It's going to help you to find relevant information quickly.
16:10
That's it.
16:11
It's your decision, your act, your action, whatever you wanted to do it, do it.
16:18
But if you wanted to find more information, instead of searching over the
16:22
internet for an hour or two hours or even a day to find something interesting, you
16:27
can use this app and find it in a second.
16:30
That's it.
16:32
So this is one of the, like, you know, one of the things that I'm
16:35
usually addressing, and I believe trust is not building in one night.
16:40
It's building gradually.
16:43
And especially, especially for farmers and growers, this is something
16:47
that they wanted to focus on.
16:48
And I, I've, we've talked about like in the last time that we discussed
16:53
that I'm so happy that I was invited by farmers to talk about being GPT,
16:58
like something totally related about ai instead of talking about variety
17:03
and some of the like, you know, new varieties that we are releasing.
17:06
Of course I'm going to talk about them like in a couple of like meetings with
17:09
them, but that I have in the future.
17:11
But that my eyes on that time was how farmers are enthusiastic to adopt
17:19
these new technologies once they are certain that these are not going to
17:24
interrupt their day-to-day activity.
17:28
So this is something that I founded for the farmers, for researchers there are
17:32
still uncertain about like, you know, how valid the results that they are
17:36
getting from it or how much they can replicate it, how much they can adapt it
17:41
in the actual like research or even is it really good to use AI for the research
17:47
or not, or how ethical is this, right?
17:50
So I'm right now a part of the committee for, uh, AI Strategy Committee University,
17:56
and I'm seeing this kind of concern.
17:58
And again, all those concerns are valid.
18:01
We need to come up with a good strategy, how to adopt ai.
18:05
And I don't want it to just totally and blindly advocate for the use of AI and
18:10
say that AI is perfect for everything.
18:12
No, it, maybe it's not right.
18:14
Let's adopt it.
18:16
Let's check it and see if it's good.
18:18
We can use it if not.
18:20
Find a new way.
18:21
Right?
18:22
So, and then, uh, last things I wanted to talk about is how much data we are able to
18:28
share and we should share in the AI basis.
18:30
And this is something that, in the conference that I had last week,
18:34
I discussed with a couple of my colleagues, you know, in Europe, and
18:37
they said we wanted to adopt this new technologies, but we are afraid to
18:41
what extent we can share our dray data, what we, to what extent we can share
18:45
our information over the internet.
18:47
And there are a couple of base, like at the university, we have our own
18:53
networking, our own domain that we can use to build like, you know, these kind of
18:58
ais and we can like, you know, develop and train our AI models based on whatever data
19:03
that we have and only share the results with everyone who wanted to use it.
19:09
So there is no need that everyone has an access to the fundamental
19:13
data set that we are having.
19:15
We can build data, we can build, sorry, ai, open all the data that we have and
19:20
share the final result with everyone so they can use these models, right?
19:26
this is one of the good strategies that we can do it.
19:29
Another one is type of the data that we are sharing.
19:32
It's most of the data that we have currently and we are not using, and I
19:36
call it dark and I call this data as dark data, is those data that we are not paying
19:40
attention to them and we don't mind if we wanted to share them publicly or not.
19:44
Like let's say in my breeding programme, I'm selecting, I'm working with
19:49
200,000 individual plants, right?
19:52
And I'm evaluating all of them, but at the end of the day, I'm selecting 7,000
19:56
of them and I'm cheering for one of them.
19:58
But how about the rest?
20:00
the rest of the data?
20:02
What, what I'm doing?
20:03
I'm just saving them, storing them as a archive.
20:07
archiving them.
20:07
So, okay.
20:08
If I wanted to build an AI or in what, like, you know, decision, if I wanted
20:14
to include them in decision making process, do I really mind about it?
20:17
No, I don't mind it.
20:19
I wanted to help on that case.
20:20
So there I, I'm sure that there are so many researchers like me, they wanted to
20:24
share such this kind of information, but at some point they need to have, like, we
20:29
need to have a good discussion around it.
20:31
We need to come up with a good strategy and also we need to have
20:35
a good transfer learning in terms of the knowledge that we have.
20:39
Jesse Hirsh: Good on Jen.
20:41
Donald Killorn | PEIFA: is funny.
20:41
You, you guys are funny.
20:43
Yeah.
20:45
You guys are funny.
20:47
Jesse Hirsh: Keep it civil,
20:48
Donald Killorn | PEIFA: anyway, no, the, no, I mean the, you know, academia,
20:52
there's a lot of like, you know, caution I understand that, you know, there's
20:57
also a lot of fervour behind the scenes about, about trying to position Canadian
21:05
agriculture and, and acade academic data management as part of the next policy
21:10
framework in agriculture in Canada.
21:12
And, um, you know, caution is important and that's sort of
21:16
always been our thing, is that data governance is gonna be, become the
21:19
most important thing to profitability and business risk management.
21:22
And, uh,
21:23
Jenn: And, uh,
21:23
Donald Killorn | PEIFA: and here we are.
21:24
And so, you know, we've got two years to figure out the next policy framework.
21:29
Uh, nobody can,
21:30
Jenn: can can
21:31
Donald Killorn | PEIFA: up with
21:32
Jenn: up
21:32
Donald Killorn | PEIFA: effective business risk management, um,
21:35
Jenn: um,
21:36
Donald Killorn | PEIFA: strategies, you know, both sides are
21:38
unhappy with the current state.
21:40
And,
21:41
Jenn: And,
21:41
Donald Killorn | PEIFA: you know, the on the key is, is in better data
21:44
processing and analysis and, uh,
21:47
Jenn: and, uh,
21:47
Donald Killorn | PEIFA: of,
21:48
Jenn: the use of, um,
21:50
Donald Killorn | PEIFA: these new software tools to, to
21:52
deliver that on a massive scale.
21:53
And, and a great example is
21:55
Jenn: is
21:56
Donald Killorn | PEIFA: credits.
21:56
I mean,
21:57
Jenn: I
21:57
Donald Killorn | PEIFA: long as I've been in Canadian agriculture for
21:59
five years, I've been waiting for environment in climate change Canada
22:02
to release the protocol for making regulated soil carbon credits in Canada.
22:06
They seem to be completely unable to.
22:09
And our agentic AI system has developed a, a very clear pathway that not only
22:15
satisfies Vera, but also keeps us from having to deal with the licencing
22:19
of third party models and, um,
22:21
Jenn: and,
22:22
Donald Killorn | PEIFA: and are able to deal with the standards and, and
22:24
parse them and, and deliver real,
22:26
Jenn: real,
22:27
Donald Killorn | PEIFA: data requirements to actually deliver
22:29
environmental goods and services credits to, to the marketplace.
22:34
Jenn: So if I could, I just, I'm gonna interrupt here because the initial
22:38
question that Jesse asked was, are we at risk of leaving some farmers behind?
22:42
He, he said it much more eloquently, uh, when it comes to ai and then
22:47
we dove into an entire discussion about data and decision making.
22:51
And what I find super curious is when I talk to farmers about ai, they get
22:56
nervous If I talk to farmers about data and decision making, they know
23:00
exactly what they need, what they're nervous about, and the kinds of
23:05
data they need to make decisions.
23:07
And so I'm wondering if we're just having parallel conversations
23:11
it's just the words we're using that are making people nervous.
23:14
And I'm curious to think if you think if we changed our dialogue or how we
23:19
are defining some of the words, if not really leaving farmers behind when it
23:23
comes to ai, and in fact, they could be further ahead than most people.
23:27
Mohsen YN: That's, this is something, uh, sorry.
23:29
Sorry, mom.
23:29
I just wanted to, uh, yeah, sorry for that.
23:32
So this is Jen.
23:33
Mohamad Yaghi: No.
23:34
Mohsen YN: Jen.
23:34
This is exactly something that I'm really advocating for that, and I just wanted
23:38
to make sure that farmers and growers are very well, like, you know, understanding
23:43
what's the use of ai, how they can use the AI for the, for their betterments of their
23:47
lives, not just replacing their job or even like threatening any of their lives.
23:54
And this is something that I, I I, I, I've mentioned at the beginning of
23:57
my talk that like, you know, instead of make them how AI can be a good
24:04
decision maker, how AI can be a good, like automating something, or that we
24:09
can tell them how AI can help them.
24:12
instead of just as I said, you know, instead of finding information for days,
24:16
they can find any information on a second.
24:18
Instead of just calling me every time or calling any of the persons every time
24:22
I, maybe I'm not as a person, I'm not, I may not be available on that time.
24:27
They can ask an AI and getting help like, you know, from ai.
24:32
These are the come of some of some of the things that they can use it.
24:35
And once they are started to use AI in this way, they are getting adapted.
24:40
And they are, I'm sure, and this is exactly the same thing that happened
24:44
with the Ontario bean growers.
24:45
They are very well, like, you know, I'm so happy and I'm so proud that
24:49
I'm working with this community.
24:51
They are very well, they are awesome Ontario bean growers.
24:54
So they adapted this new technologies and gradually right
24:57
now they are advocating for ai.
25:00
So this is something that we built based on not a very good trust at the
25:07
beginning, but we started to use it and test it to see how it goes, right?
25:12
And it goes perfectly right now.
25:14
And we are both are in a agreement that we should advocate for AI for
25:20
the next, like, you know, future, uh, whatever, like projects or subjects.
25:25
So this is something that I'm trying to, uh, do it, uh, when, when it comes
25:30
to the farmers and growers, not only for decision on how I wanted to use
25:34
the data or how I wanted to keep the data or managing data, but also try
25:38
to make AI simple and understandable and try to assure everyone that AI is
25:44
here to be a good tool for all of you.
25:49
Mohamad Yaghi: So I, I'll jump in.
25:51
And, Jen, I think to, to answer your question, I think the, the term
25:56
artificial intelligence is a cliche.
25:59
And the reason I say that is it has been, been mentioned so many times
26:05
in so many capacities that it's lost.
26:07
Its like real definition.
26:09
And if I were to put a definition for artificial intelligence, I
26:12
would say it is advanced statistical pattern recognition, at scale.
26:18
And there are multiple ways of achieving this through tools, like large language
26:22
models, through neural networks, through, uh, a plethora of different tools.
26:27
So when it comes to the conversation of ai, because at a really high level, it's
26:32
been talked about in really broad terms, it seems as if, you know, farmers have to
26:38
walk through this new paradigm when they are trying to operationalize this, these
26:42
tools on their farms, for the most part, in my opinion, they are using elements
26:47
of artificial intelligence, albeit, okay, they're not using, they're not
26:51
maybe building their own neural networks and processing billions of images at a
26:56
time that, but they are using, again, introductory tools at the very least in
27:01
order to evaluate different decisions.
27:04
Now, whether artificial intelligence or the data they have on their farm
27:08
will enable them to make real time decisions, that's a, that's a journey
27:14
every farmer is at a different stage on.
27:17
So when it comes to using artificial intelligence at scale, it really
27:22
depends on the tool that they have.
27:24
And the thing I would, I recommend to producers is don't
27:28
buy something off the shelf.
27:30
If you think that you're going, because you think you're missing a gap on
27:34
your farm, use your, the strength that you have on your operations
27:39
to fully utilise these tools.
27:41
So for instance, if you, as a producer, again, let's take Jesse for instance.
27:46
Jesse is an operator and he has a really keen ability to maybe forecast
27:50
yields, and he has a data set that helps him every year build on that
27:55
expertise as opposed to introducing a new tool to enter your operation.
28:00
And I think that's the, the, the evolution, a lot of producers at
28:03
a are at a different stage yet.
28:05
And I think that's part of the conversation as to why AI is a bit
28:09
convoluted these days is because we just put any tool that we're
28:13
seeing, uh, you know, mention all the news into this bucket.
28:17
And it's, that's why it's become a bit of a cliche and it's sometimes
28:19
why some producers are a bit apprehensive to have that conversation.
28:23
But when I talk to a lot of producers, even using your mobile
28:26
device is a form of using ai.
28:28
If we're gonna be nitty gritty about those details.
28:32
Donald Killorn | PEIFA: mobile devices are great Edge computers.
28:34
That's, that's a big, uh, like, that's a big, um, rallying statement in our
28:38
shop these days is that every farmer's already holding an edge computer.
28:43
Um, you know, the exciting thing is agriculture has a, so much to
28:50
make up or, or sort of has missed kind of Web 2.0 and, and a lot of
28:56
cloud-based computing opportunities.
28:58
And so managing a trillion dollars worth of capital.
29:02
Um, and they're doing it with, systems that haven't been fully modernised,
29:09
uh, compared to other serious capital buckets and, and supply chains.
29:14
And so if you boil it down to it, I still think that this moment is about
29:20
to Muhammad's point about advanced statistical probabilities and at
29:25
scale, agriculture's a perfect example.
29:28
And bringing the ability to collect, uh, analyse and manage data effectively
29:34
and scale, um, scale up our ability to learn from that full dataset through
29:42
a, a sovereign compute strategy that, that keeps, you know, effective gates.
29:47
That's how we fix business risk management.
29:50
It's business risk management is slow backwards looking.
29:54
It doesn't incentivize resilience.
29:56
And in order to do those things with a national, uh, business risk management
30:02
programme for Canadian farmers, we have to use cutting edge technology to
30:06
collect, manage and, and analyse data and, uh, deliver insights both to the
30:12
farmer and uh, to the lenders and the insurers and, um, and the marketplace.
30:18
Jesse Hirsh: Well, and, and Donald, let, let me get you to ho
30:20
Mohamad Yaghi: in super quick.
30:21
Jesse Hirsh: ahead.
30:24
Mohamad Yaghi: Go.
30:24
Go ahead, Jesse.
30:24
Go ahead.
30:25
Jesse Hirsh: Well, I, I wanted Donald, you, you evoked something really powerful
30:29
earlier when you talked about, you know, the way in which your agents were
30:33
outrunning, uh, environment Canada's carbon credit system, to, to what
30:38
extent is the agentic era accelerating everything we're talking about?
30:43
Like, on the one hand, I think Mohamed, you know, quite correctly was
30:46
pointing out the win, which a range, an ecosystem of technologies are
30:51
being collapsed into a single phrase.
30:54
But the agents now are a, a, a, a driving part of that ecosystem.
31:00
It's a part you're playing with, I, I think, uh, in,
31:04
in a really interesting way.
31:06
So take what you just described and throw agents into the mix and
31:10
how do we ground that back into the AI for all kind of rallying cry.
31:18
Donald Killorn | PEIFA: Well, the models, you know, the, these models
31:22
and, and I agree with Mohammed that artificial intelligence is cliche.
31:26
I think that's very well put and that it encompasses many
31:29
different types of technology.
31:31
But the, um, these, a, these mo these age agentic approaches to
31:38
utilising the models is really about giving them managing context.
31:44
So you're, you're trying to create boundaries in which they're
31:49
responsible for in an effort to manage their context window and
31:53
produce results more efficiently.
31:56
And efficiency, you know, is why we need a sovereign compute
32:02
strategy for agriculture.
32:03
We need to under, you know, research, conduct, research and development on how
32:07
to identify and move the data around and ensure that we can maximise the benefit
32:12
of, of all agricultural data while, um, operating within the bounds of good
32:17
taste when it comes to data governance.
32:19
Um, so that ag agentic moment, we saw a lot of, saw a very rapid movement
32:26
between, um, Claude code and, and the ag agentic approach to utilising
32:34
those frontier large language models.
32:36
We're now at a point where the frontier models already are becoming,
32:41
uh, prohibitively expensive for the type of day-to-day modelling that
32:45
we were using them for in February.
32:47
And so we need, and not only that, but they're a huge risk to data governance,
32:53
uh, and, and data sovereignty and need to be utilised appropriately
32:59
and, and, and effectively, uh, to manage cost and, and manage, um, risk.
33:05
And so now the, you know, we're, we're, we're working with to, to Mohammed's
33:11
point, we, we match these frontier models with the, the research loops.
33:15
And, and really what, what's happening now is trying to get these systems
33:20
running, um, fully air gapped, you know, we've got this mocked up air gap
33:26
system using a series of Mac minis and these days at, at the Federation, it's
33:31
all about bare metal, you know, and, and getting the models on bare metal.
33:35
And there's no question that there will be a hardware device designed
33:39
that combines a raspberry pie and a wifi gateway with, with a, an
33:44
agricultural, um, inference model.
33:47
And that that will be sold to farmers in the next, you know, two to three years.
33:52
That that hardware, software, um, to facilitate the field boundary management
34:00
of data, um, is the, the whole, that's where the, that's where LLMs
34:06
are headed across the whole economy.
34:07
And, uh, you know, this is as an exciting a space for this technology
34:12
as any, uh, in the Canadian economy.
34:15
Mohsen YN: So, uh, ma Matt, I, I just wanted to, uh, ask you another question.
34:20
So would you please again define what is artificial intelligence?
34:24
What did you say?
34:25
Sorry, for what?
34:26
How did you define it?
34:27
Mohamad Yaghi: yeah.
34:28
The definition I've been using just as a, a, an approachable standard
34:32
is it is advanced statistical pattern recognition at scale,
34:38
Mohsen YN: Okay.
34:38
Mohamad Yaghi: um, leveraging different tools like large language models
34:42
or neural networks, uh, in order to uncover insights, forecast or,
34:48
um, tech patterns within datasets not discernible to the naked eye.
34:52
Um, again,
34:53
Mohsen YN: So.
34:54
Mohamad Yaghi: my, I'm just, I'm just riffing here.
34:56
Mohsen YN: mentioned, so you mentioned advances, statistical approach, but
34:59
my question is, do you think, uh, simple regression, linear regression
35:03
model is a part of AI or not?
35:07
Mohamad Yaghi: You know what's funny Mo then, um, the Ontario Securities
35:11
Financial Institution, whatever it's called, office fee, would consider that ai
35:16
Mohsen YN: it's not, it's not only Ontario, it's the science.
35:19
Mohamad Yaghi: It.
35:19
Yeah, and it
35:20
Mohsen YN: ai,
35:21
Mohamad Yaghi: the
35:21
Mohsen YN: one of the, one of the aspects, one of the subjects
35:24
of the AI is machine learning.
35:25
And on the machine learning we have supervised learning.
35:28
And on the supervised learning we have the regression models.
35:30
And on the regression we have the simple linear regression.
35:34
So AI is not only pattern cognition and AI is not like classification.
35:40
We have so many areas in ai, you, when, when we are talking about ai, it's, it's,
35:45
it's exactly look like that You talk about agriculture, but what's inside agriculture
35:50
is plant breeding, plant physiology, geneticists, you know, all those areas.
35:55
I, I, I, to be honest, I cannot say AI is the advance advanced statistical approach.
36:03
I can, I can refer to what is AI is based on the IBM that provided as artificial
36:09
intelligence, the technology that enable computers and machines to simulate human
36:13
learning, comprehension, problem solving, decision making, creativity and autonomy.
36:20
This is something that I can refer what is ai, but, and it comes to the
36:25
advance, I'm, I'm, I'm against of that.
36:28
I don't, I don't think that, you know, AI is the all about advance.
36:32
It's
36:32
Jesse Hirsh: So.
36:33
Mohsen YN: and ai, this is something AI established in 1915.
36:37
Mohamad Yaghi: don't want to split hairs on like a definition.
36:40
I think my intention with the term advanced is not as simple.
36:45
It's not, it's not, it's not binary, right?
36:48
It's just, it's, it's more of like, you're using sophisticated ways to get to,
36:53
let's say, um, an insight for instance.
36:56
Jesse Hirsh: let me.
36:56
Mohamad Yaghi: and it's like the language I'm using isn't
36:58
Donald Killorn | PEIFA: Co codifying, codifying language and, and,
37:01
and, and conducting modelling and being able to, um, I, I'm not,
37:06
I don't think it's particularly elegant, but, you know, neither
37:10
are, um, power lines, you know, so,
37:12
Jesse Hirsh: But let me, let me refocus.
37:14
Donald Killorn | PEIFA: but, but breaking language into code, the same way we
37:17
broke genetics into code, and we have economics, you know, analysis is, is
37:22
Jesse Hirsh: So.
37:23
Donald Killorn | PEIFA: you know, an advanced piece of technology.
37:26
There's no question about it.
37:27
Like it, it
37:27
Jesse Hirsh: We're, we're.
37:28
Donald Killorn | PEIFA: one of the main three languages, like
37:31
the three, three codes that humans use to create the associates.
37:34
Jesse Hirsh: So let me, let me bring this back.
37:36
Hold, hold, hold on.
37:38
You know, Mo Molsen, I think made a really important point, which is, if we
37:43
think of AI as a spectrum rather than a binary, then that spectrum on both
37:49
sides goes much farther than a lot of people really entertain or are aware of.
37:56
And, and one thing I I really want to bring the conversation back to,
38:00
because Donald evoked this a couple of times, you know, while really fleshing
38:05
out why sovereignty is such a crucial part, uh, of this larger conversation,
38:11
I know that the vast majority of people do not understand what, what's
38:15
involved in training a model, right?
38:18
They hear about these models, they hear about Mythos, they hear about Fable,
38:22
they hear about, you know, a, a, a deep seek, all these new models out there.
38:27
But we do have two people here on the call, maybe three, uh, who have
38:32
actually trained and built models.
38:34
So break that down again, uh, not just for Jen's benefit, but I think for a
38:39
lot of people's benefit when we start to imagine a, a sovereignty, a Canadian AI
38:45
ecosystem that runs in Canada, trains in Canada, infers in Canada, how difficult
38:53
or easy it is to train and build models.
38:57
Moen, I'll go to you first, but then Mohamed, obviously, I, I, I want you
39:00
to jump in when you, when you see fit.
39:04
Mohamad Yaghi: Sure.
39:05
Mohsen YN: That that's a, that's a great question, Jess.
39:07
The point is, uh, it, it totally depends on the use of the AI models
39:14
and after, like, you know, in what area, in what like, you know, aspects,
39:19
do you wanted to use AI models?
39:21
And then I can tell you how difficult it is for training the models.
39:25
I'm, I'm thinking as a, like, as a person who wanted to do the research, and I
39:30
can't give you a very immediate answers, but to be honest, like, you know, as
39:34
long as we have a broad spectrum of the data that cover all the instances that
39:40
we are facing and we have faced before, I think that would be a good, uh, like, you
39:46
know, a database for training the models.
39:50
Let's say if I wanted to make a decision about like plant breedings, I'm, these
39:55
days I'm trying to move toward the digital twin and developing digital
39:59
twins for plant breeding, instead of just wasting so much of energy and like, you
40:03
know, plants in the field, I can just simulated what will happen in terms of
40:07
the future climate change scenarios.
40:08
So in this case, what I fed my models was the 50 years of the databases that all
40:15
was grown in different areas, locations in North America, and I collected all
40:20
this information and train, and it still is not enough because I can't predict
40:25
what will happen in the future, really.
40:26
It's so hard to predict, right?
40:29
But this is one instances.
40:31
Another instances is I, I'm usually, I, I wanted to use AI to predict
40:35
the pattern on the maturity of the pattern of my dry beans, right?
40:39
So the maturity pattern are usually, usually consistent.
40:42
So if I have a couple of years data and I can include it to the training,
40:47
the models, that would be enough.
40:49
So the major, the point, the question is and how you wanted to use this model.
40:55
And then training is based on like, you know, training the amount of the data
40:59
that you need to train is based on what, where you wanted to use this AI model.
41:09
Mohamad Yaghi: So, Jesse, can you repeat your question more time?
41:11
I just wanna make sure that, uh,
41:13
Jesse Hirsh: So
41:14
Mohamad Yaghi: all the
41:14
Jesse Hirsh: well, what's involved in building a model?
41:16
I, I think when people hear the phrase AI sovereignty or Canada needs to
41:21
control its AI infrastructure, they don't totally understand the pieces.
41:26
And obviously the model, whether the r Tom's points specifically the focus or the
41:32
capabilities of the model is important.
41:35
So again, from a big picture, from a leadership perspective, what's
41:39
involved in training a model?
41:41
And is that, is that the kind of capacity we can and should
41:44
be developing across the sector?
41:49
Mohamad Yaghi: Okay, so I'm gonna cheat a bit, uh, and I'm
41:51
gonna take it further than that.
41:53
But again, to my two colleagues on the call include where
41:57
you think I'm messing up.
41:59
But essentially, I'm gonna keep it simple, but essentially, I think the first thing
42:02
when it comes to model development, and this is for you Jen, I mean, like
42:06
the way I think about it is like, let's define the question, um, what are we
42:10
building the model to predict or decide?
42:13
And that's where we would start things off from.
42:16
Um, for instance, like when we're thinking about agriculture in particular,
42:20
like will, like can we build a model to predict if the yield will drop
42:24
below a threshold, for instance, based on a lot of different elements.
42:28
That's maybe the question we want to answer.
42:31
After you have that ref question again to the best of your ability.
42:36
And again, that question can also change.
42:38
This is a dynamic process.
42:40
You would gather that data.
42:41
And I think when people say data is the, the new oil, like that's the
42:46
point that I think this is all about.
42:47
It's because when it comes to looking at forecasting or trying to uncover insights,
42:54
more data you have historical, or in terms of let's say when you're looking
42:59
at yields, do you have weather data?
43:01
Do you have market information?
43:02
Do you have, um, information by activities and the impact of those activities
43:07
like a farm activity, like cover crops, or let's say no-till for instance.
43:11
How is that going to impact the model ultimately so that you can answer
43:16
that initial question you have.
43:18
And then of course, once you have that foundation of data, and it can
43:22
come from different platforms and the reason why, and just to go little
43:26
sidebar for a second, the conversation about that ecosystem play where all
43:31
three of us or all five of us on this call can share information freely
43:36
amongst each other, which is what we call an API, it's basically a digital
43:40
handshake for lack of a better term.
43:42
If we're able to share that information amongst us all, that actually enables us
43:47
to answer the question more accurately.
43:50
In my opinion.
43:51
of course takes a lot of training.
43:54
It, it's not easy, it's gonna take time as well.
43:57
But once you have an ecosystem where you can share that data, you're actually able
44:01
to get more information that you can, you can incorporate within your model as well.
44:06
then when you, once you decide on the appropriate data set, once you clean
44:11
that data as well, because Jen, I can't, I can't emphasise enough how many
44:15
times I've seen glyphosate spelt in.
44:17
I've seen it maybe spelt in a hundred different ways, like
44:20
I've seen it with the U once.
44:21
So you have to go through that process and maybe make sure
44:23
that the data itself is clean.
44:26
then like, I guess like the last two steps is then you have to test it.
44:30
And then you have to measure the confidence you have in the results
44:33
you get from that information.
44:35
And once you test it, then you can, uh, deploy and monitor it as well just to,
44:40
again, when you use, when you deploy it, you you're never going to have
44:43
it percent effective from the get go.
44:45
You have to improve on it.
44:47
Now, the part that I want to cheat a bit about is the concern I have
44:52
within the industry these days.
44:54
Um, and what we're seeing is, and not in a Canadian context.
44:57
I think agriculture for the most part has been forward looking
45:01
when it comes to sharing data.
45:04
And all five of us on this call have, are again, are building protocols to enable
45:09
ourselves to share that information.
45:10
And I think that is amazing.
45:12
The trend I'm starting to see, and this is coming from the equipment manufacturers
45:18
again, like tractors and whatnot.
45:20
They're starting to charge their users money in order to not only collect that
45:27
data, but then even to share that data.
45:30
And the way I would it in an analogy is it's like you're going to a pizza
45:35
parlour, you're paying for the pizza, and then you're getting charged to
45:38
share those slices to your friends.
45:41
And that's not really building an ecosystem then.
45:43
And that's the big concern I have.
45:45
So when it comes to that foundational model building,
45:49
that's the way one were to do it.
45:51
And the way we're able to maybe cultivate that ecosystem so that we
45:56
can share better and more pivotal information amongst one another.
46:00
So I try to keep that short.
46:03
Jesse Hirsh: No, that, that was fantastic.
46:05
And Donald, it, it really, uh, allows me to bring you in as someone who I
46:10
know is following news, uh, in the AI world, although I suspect all
46:14
present are, and with the removal of the, uh, uh, mythos class models of
46:21
Fable and, um, the, uh, larger one that was per, uh, made available.
46:27
There's a lot of people talking about open source.
46:30
There's a lot of people talking about sovereignty.
46:33
To your earlier point, the price of accessing the frontier
46:36
models continues to go up.
46:39
Is there a critical mass now, not just of collaboration as I think Mohamed was
46:45
inferring, but of active data sharing, of creating, uh, an ecosystem of models
46:52
that are more accessible, more affordable, or, or am I just dreaming here?
46:57
Right.
46:58
I I'm curious how, if you could connect some of these dots for us,
47:01
uh, especially grounded in, in the work that you're actively doing.
47:10
Donald Killorn | PEIFA: Yeah, I mean, ev it, this is a great panel with, with
47:15
an NGO rep and an, and an industry rep and, and an academic rep. And, uh, all
47:20
those sectors along with government have to work together to build more
47:23
sustainable systems, whether it's the food system or another complex system.
47:28
Um, and that's what, what we'll find here.
47:31
Um, I, I think, I personally think the, the whole mythos rollout is,
47:36
um, marketing, you know, up to and including this, um, is impressive
47:42
and, and sort of aligns with this kind of massive um, growth strategy
47:50
around, um, you know, picks and shovels and, um, what you can do with them.
47:56
Um, so yeah, gold Rush is tough to manage, and, and you never know who's working with
48:00
who to make sure that this thing continues to drive new records in the stock market.
48:05
but there's no question that, our future in, in agricultural data,
48:14
uh, means ESP at the farmer level.
48:17
And motion.
48:18
Motion has said a few times about how fast agriculture is and the different
48:23
implications for data in agriculture.
48:25
And so I do wanna just say that, you know, from the, from the grassroots at the farm
48:30
field level, some data will leave the firm, the field, I should say, the field.
48:37
Some data will leave the field only if it empowers the rest of the model, if it
48:42
empowers the model to grow more stronger.
48:44
For the most part, uh, decisions will be inferred at the field boundary
48:48
based on, uh, lightweight models that can be housed in edge computing.
48:54
And, uh, I suspect again, that there'll be hardware development
48:57
around that for farmers.
48:58
But, um, we will not be moving all of the data off farm.
49:03
Uh, we'll move hashes to the blockchain, and we may move the data into the cloud
49:08
if that's what the farmer wants, and, and they're willing to pay for that.
49:11
But, um, the effort to, um, collect and manage and analyse and build
49:17
decision making capacity for farmers at the farm field level, um, we cannot
49:23
move all of the data outta the field.
49:25
Um, it doesn't make sense for data security, and it doesn't make sense for,
49:31
um, a practical, efficient use of, um, of large language modelling and, and, uh,
49:37
building, uh, decision making capacity.
49:40
Artificial intelligence, it's, it's, um, know, we can outsource intelligence,
49:45
but we can't outsource understanding.
49:46
And so we will deliver intelligence to the farmers, uh, in a way that
49:50
they can, uh, uh, utilise and then they will interpret it and, uh, and
49:55
implement it, uh, on their land.
50:00
Jesse Hirsh: And do you wanna just take a moment to unpack edge computing and 'cause
50:04
again, I think it's one of those phrases that those of us in the know understand,
50:08
but you've made a really good argument today as to why it is an essential
50:12
element of any agricultural AI strategy.
50:16
Donald Killorn | PEIFA: Yeah.
50:16
Yeah.
50:17
We, we need $150 million, uh, to do a, a few years of
50:22
research and development here.
50:23
And, you know, in around 75 or 80 million of that has to be in equipping
50:30
our farmers with edge computing.
50:32
And it should have been part of off calf, it should have been
50:34
part of the last policy framework.
50:37
Um, but we have to build an intelligence layer, uh, right on top of the field.
50:42
Um, and in order to do that, we need edge computing, which would be any form of
50:46
compute that's found sort of in the, in the three dimensional world, like in the,
50:52
in the space and time as we understand it.
50:55
Um, so that edge can be sensors, it can be cameras, uh, it can be GPS,
51:01
um, you know, that's life on the edge.
51:04
It's, it's the, um, it's your, uh, real, your connectivity to the real world.
51:12
Uh, as much as we want to use our 50 year bio geochemical model, and, uh, as much
51:18
as we want to use our climate modelling, uh, to help build decision making to e
51:23
everyone's point today, uh, we need to mix that with what's happening on the farm
51:27
right now, uh, with those outlier events and, uh, make sure that, um, the models
51:33
are ground truth effectively, constantly being updated as the climate changes.
51:38
And, uh, that they are equipped to not just deal with the status
51:42
quo or the average, but with, um, what's actually happening on farm.
51:46
And that happens through the, through the edge layer.
51:49
Jesse Hirsh: Right on.
51:50
And we are, uh, just about out of time.
51:53
So I'll, we'll take a moment to do kind of forward looking or speculative
51:57
thoughts on where this is heading.
51:59
And, you know, Mosen, I'm gonna, I'm gonna throw to you first only 'cause I just
52:03
read about, uh, you, uh, uh, alluding to being GPT expanding beyond Beans and
52:10
expanding into other research areas.
52:13
So perhaps use that as a way to talk about where, where you see the kind of near
52:17
term in terms of this stuff evolving.
52:21
Mohsen YN: Yeah.
52:22
That's great.
52:22
So, uh, so before I talk about like, you know, what would be next, uh, after
52:31
being GBT, I just want to say like, you know, uh, I totally agree with Donald
52:35
and Mohammad, what they said, what they mentioned about like, you know, data
52:39
management and also how to get this data because I think AI is been, has been here
52:45
for like more than seven years, but the major problem right now that we are facing
52:49
is how we can obtain the data and how we can share to what extent we are able to
52:54
share the data and to what extent we can ask farmers to produce this and collect
53:00
this amount of data and make it accessible to us for developing different models.
53:05
But anyway, so in terms of the being GPTI initially, like, you know, we initially
53:11
created being GPT specifically for beans.
53:13
But right now, after I had this, uh, good like discussion after the presentation
53:18
that they had last week for in, uh, international Legum conference, uh, I, I,
53:24
I found a very great interest on the use of binge GPT and they said they wanted
53:28
to like, you know, uh, I, I received a good like, you know, uh, interest in
53:34
terms of, uh, advancing binge GPT beyond beans, like, you know, on the lentil, on
53:41
other legumes and also on other crops.
53:45
Definitely possible, possible like, you know, to extend it to another crops.
53:48
So we wanted to do it.
53:50
And also another feature that we wanted to add is the powder recognition.
53:54
Definitely wanted to add it to the BGPT as the be vision that everyone
53:58
can just take the picture of the raw seed bin on ZI bin and then, you know,
54:03
post it on the BGPT and binge G PT can tell them that what with the quality.
54:07
But this is all about the BGPT.
54:09
But the future of AI I think's going to be more on, uh, um, quantitative ai.
54:19
So I think, you know, uh, I think it's going to be more on, oh, sorry.
54:25
I'm just saying, uh, it's going to be more on the quantum quantum approach and in the
54:31
next couple of years I think we are going to shift toward the quantum approaches.
54:36
And I'm seeing that a couple of interest around that they are sorting
54:40
to adopt it, the quantum approach.
54:42
And again, quantum approach is not something new.
54:45
It's, and it'll be new like in the couple of years everyone will
54:48
talk about the quantum approach.
54:50
But I think, you know, the future of AI is belongs to quantum approaches.
54:55
Yeah.
54:58
Jesse Hirsh: Mohamed.
54:59
Mohamad Yaghi: I'll let Donald go.
55:00
Jesse Hirsh: Okay, Donald.
55:01
Donald Killorn | PEIFA: Yeah.
55:01
And, uh, yeah, the, the Bitcoin owners are definitely focused
55:05
on a quantum approach for sure.
55:07
The, uh, and farmers will be interested to see quantum is, you know, it's
55:13
very, it's a huge part of if this thing is gonna get, get somewhere.
55:17
But anyway, um, yeah, that's very interesting.
55:20
I, I think that, um, you know, I haven't made any, any big, um, secret about
55:27
where I think we're headed, I guess in the, in the near term, like, uh, we
55:31
need a tiered sovereign architecture.
55:33
Um, we need the edge layer to be a secure public gateway
55:37
back and forth from the field.
55:40
Uh, we need the compute layer, uh, that has that frontier AI capability for when
55:45
we need advanced reasoning and, um, AI assisted automation and decision support.
55:51
Um, high performance retrieval operation, scalability.
55:55
And then the key is the sentinel layer that's sovereign data control.
55:59
That sentinel node that sort of is the traffic cop, uh, that allows
56:03
us to unlock, uh, the full benefits of the edge and the compute layers.
56:10
Um, that third piece, um, helps us use Frontier Cloud ai, but also ensures
56:18
fully sovereign local ai, uh, for those sensitive or regulated workloads.
56:24
Um, I think, you know, mixing all those together, um, and, and, and ensuring
56:30
that we're focused on, sharpening our pencils when it comes to, how
56:35
much compute we need and how, and therefore how much electricity we need.
56:39
'cause there's no question there's a bottleneck in electricity.
56:42
And so being as efficient as possible will help us execute on our, on
56:46
our sovereign compute requirements.
56:51
Mohamad Yaghi: Then I'll go, um, I guess wanna make you richer, Jesse.
56:56
That's, that's where I want to get to in the next five years.
56:58
And if we can use artificial intelligence to get you there, like the different
57:02
tools, uh, that's the way I wanna do it.
57:04
And I think, you know, I think again, the tools that are encapsulated within
57:09
artificial intelligence will move away from a tool that's an individual op
57:14
from individual operations to adopt.
57:17
Um, that, that, that are adopted in isolation actually to the connective
57:22
layer that connects equipment, agronomy, lending, uh, and markets
57:27
into a single intelligence system.
57:30
And it's through the use of agents as well.
57:32
We didn't get into too much of that today, but maybe we can assemble all the
57:35
Avengers again and, uh, talk more AI soon.
57:39
Jesse Hirsh: Yes, absolutely.
57:40
If, if in no small part because of the rapid rate of change that I think
57:45
has been one of the undercurrents, uh, to our conversation today.
57:49
And that is where I think the original challenge that, that Jen paraphrased
57:54
of making sure that no farmer is left behind is going to be an ongoing process.
57:59
Uh, and that's where your respective leadership is indispensable.
58:03
Jen, any final thoughts or comments from you before we conclude?
58:09
Jenn: Yeah, so many.
58:11
Uh, I have a lot more questions, so I would love to assemble
58:14
the Avengers once more.
58:16
Um, you guys have taught me a lot.
58:19
I have some deep thinking to do,
58:21
Jesse Hirsh: Right on.
58:22
Jenn: in touch for sure.
58:23
Jesse Hirsh: Right on.
58:23
Thank you very much.
58:24
Mohsen YN: Good.
58:26
Jesse Hirsh: And again, thank you.
58:28
Uh,
58:28
Jenn: again,
58:28
Jesse Hirsh: will have to on Mohammed's call, uh, uh,
58:31
reassemble, uh, some point soon.
58:34
I will work, uh, have my scheduling agents talk to your scheduling agents and
58:39
we will be able to work something out.
58:41
Thanks again, everybody.
58:43
Donald Killorn | PEIFA: Thank you, Jesse.
58:44
Thanks everyone.
58:45
Talk to you soon.
58:46
Mohsen YN: you.
58:46
Thank you.
58:47
Mohamad Yaghi: later.
58:47
Mohsen YN: Take care.
58:47
Bye bye.
58:50
Jesse Hirsh: Future heard assemble.
58:53
What a fantastic session that quite frankly exceeded the already
59:00
ambitious, uh, expectations I had when gathering together the
59:05
kind of bright minds that we did.
59:07
Now granted, this is the kind of episode that we could have gone on,
59:12
uh, for another hour, maybe even two.
59:15
I really do try to limit these episodes, uh, both for my own, uh, biological
59:21
needs, but also so that we can allow these conversations, uh, to, to uh,
59:27
cultivate with time to like fine wine, continue to ferment and allow
59:32
that knowledge, uh, to multiply.
59:35
' cause there were a lot of threads that came up today.
59:38
In particular, Donald kind of evoking how language is really at the core
59:44
of the current revolution in ai.
59:47
And it's our ability to use our vocabulary, in particular, an agricultural
59:53
vocabulary, a business vocabulary, a natural vocabulary, biological chemical.
1:00:02
All of this suggests that we're entering into a world where it's
1:00:06
not just about talent, it's about collaboration amongst the talented.
1:00:10
And granted, the agricultural sector has always valued collaboration.
1:00:16
But one of the very subtle through lines of today's conversation that Donald,
1:00:21
to his credit, kept emphasising is the value of cross-sectoral collaboration.
1:00:27
What happens when you get a researcher and a banker and an association rep in a
1:00:33
room with a hacker, and you start having conversations about what is possible, what
1:00:39
we should focus on, what is important, what is, uh, the source of curiosity
1:00:44
when it comes to not just surviving but thriving in the technological
1:00:49
revolution we find ourselves in?
1:00:52
I'm already starting to schedule, uh, uh, these three wise guys,
1:00:56
uh, into another session where we will go even further because the
1:01:01
AI revolution waits for nobody.
1:01:04
And the question is not whether we need to conform to the marketing
1:01:09
and hype of the hyperscalers.
1:01:11
Quite the opposite.
1:01:12
We can and will bend the technology to our needs, our culture, our values, and
1:01:19
that's where Muhammad Mosen and Donald are really, uh, exemplary of the kind
1:01:26
of leadership we have here in Canada and how that can be applied to technology
1:01:31
to make the technology work for us.
1:01:34
So thanks to them.
1:01:35
Thanks to Jen, thanks to you, uh, for listening, for watching, and for sharing
1:01:40
this with your friends who are wondering why AI should be taken seriously, why
1:01:46
we can be both critical of it, but also bending it to meet the desires and dreams
1:01:52
that we as a society, as a culture hold.
1:01:55
Uh, so thanks again.
1:01:56
Uh, we'll see you soon here on the Future Herd.