August 5, 2025 • 71 mins
A conversation with Rob de Laet, project lead of Cooling the Climate and co-author of the book Cooling the Climate: How to Revive the Biosphere and Cool the Earth Within 20 Years. The science is pretty clear and getting clearer by the day: water cools the planet. The more living, healthy vegetation we have on this planet, predominately perennials and thus trees, agroforestry systems and healthy forests, the cooler the climate is and the less extreme weather events occur. Living plants literally make the Earth sweat and remove heat from the biosphere.
Humans have systematically devegetated the planet as Judith D. Schwartz likes to say, and the ongoing climate weirding suggests we may have gone too far. Now we're seeing real calculations: how many square kilometres do we need to regenerate to lower the global temperature by just one degree?
If this is all becoming increasingly evident, why isn’t it common knowledge yet, especially in the headquarters of banks, insurance companies, sovereign wealth funds, and governments?
This is the story of a successful entrepreneur getting drawn into water cycle restoration, planetary cooling and all the good stuff that comes with it. We share notes on why this movement, maybe the defining story of our time, hasn’t broken through yet and what we can do about it.
==========================
In Investing in Regenerative Agriculture and Food podcast show we talk to the pioneers in the regenerative food and agriculture space to learn more on how to put our money to work to regenerate soil, people, local communities and ecosystems while making an appropriate and fair return. Hosted by Koen van Seijen.
==========================
👩🏻💻 YOUR OUR WEBSITE
💪🏻 SUPPORT OUR WORK
- Join Gumroad
- Share it
- Give a 5-star rating
- Buy us a coffee… or a meal!
==========================
🎙 LISTEN TO OUR PODCAST AND SUBSCRIBE TO OUR CHANNEL OR WATCH IT ON 📽️ our YouTube channel
==========================
FOLLOW US!
==========================
The above references an opinion and is for information and educational purposes only. It is not intended to be investment advice. Seek a duly licensed professional for investment advice.
Thoughts? Ideas? Questions? Send us a message!
https://investinginregenerativeagriculture.com/course
Find out more about our Generation-Re investment syndicate:
https://gen-re.land/
Thank you to our Field Builders Circle for supporting us. Learn more here
Feedback, ideas, suggestions?
- Twitter @KoenvanSeijen
- Get in touch www.investinginregenerativeagriculture.com
Join our newsletter on www.eepurl.com/cxU33P!
Support the show
Thanks for listening and sharing!
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:03):
Welcome to another episode today with technologist
innovator, inventor and a leaderin the.
I have to do it again.
Technologist innovator,inventor and a leader Welcome,
paul.
Hi, good to be here and firstof all shout out to Henry
Dimbleby, who put us in touchand who already mentioned you in
our two hour long deep diveDon't go and listen to that now,
people, I'll put a link belowif you haven't and who already
(00:24):
mentioned you in ourtwo-hour-long deep dive Don't go
and listen to that now, people,I'll put a link below if you
haven't listened to that yet butmentioned digital twins and
said you really, really have totalk to Paul.
And here we are, and I agree.
We had a pre-call and I had thepleasure of doing a bit of
background research into yourfascinating journey into
(00:44):
technology and, of course, theworld of food and agriculture.
You're definitely not deep inthe world of regenerative food
and agriculture, but you'veplayed a large role in moving a
lot of food like more food thanmost people will ever touch in
their lives, and so I would loveto unpack that what you see,
what you have noticed, what arethe opportunities of digital
(01:04):
twins and other deeptechnologies that we need, that
are already there that we mightnot even know how to use yet, so
I'm really looking forward tothat.
This will be a wide-ranginginterview and conversation, but
first of all starting with apersonal question in this case,
how did you roll into the worldof technology case?
(01:25):
How did you roll into the worldof of technology?
I can see background behind youa lot of different boxes and
things, probably full of cablesand all kinds of um things to
build.
But what were you always likethat?
Or how did you roll into intothe world of tech?
Speaker 2 (01:36):
well, I suppose as a child, um, you know when, when
many of my contemporaries werereading kind of I don't know,
innie b Blyton or books likethat, you know, I was normally
reading instruction manuals forcars and things like that and
taking stuff apart and puttingit together again and maybe
making new things out of thoseparts.
(01:58):
So I was always fascinated byhow things work and how things
are built and by machines of allsorts, and so that love of how
the world works kind of drove meto studying physics at
university.
(02:18):
But even then I was fascinatedby the kind of the early days of
personal computers.
But even then I was fascinatedby the early days of personal
computers and I nearly droppedout of my physics degree to do a
first startup.
But I always kind of knew thatI wanted to work with computers
and technology, and probablywithin startups, and so that led
(02:43):
me into the computer industryand sure enough, about two years
in the opportunity came to do afirst startup and and I grabbed
it.
So it's, it's really it's beenthat way from the start and, and
I suppose over that journey thetechnologies may have changed
and evolved, but many of thesort of many of the themes have
(03:08):
continued, particularly the theexcitement of working at the
interface of software andhardware systems.
You know things, things thatmove under software control are
much more exciting than just ona screen.
Speaker 1 (03:31):
And was it always the case for you like that
interface or that connectionbetween the hardware, physical
world and the software world, orhas that been gradually over
time, like your first startup orthe first companies you worked
at and with?
Was that predominantly thesoftware side or also the stuff
you could touch the hardwareside?
Speaker 2 (03:49):
No, I think earlier on the startups that I worked in
were more software focused, butI suppose then the physical
world started to come in interms of sort of the more
real-time applications.
So so software that interactswith the real world in terms of
(04:13):
control, and then that led on tothe whole area of what I would
now call smart machines.
Others might call them robots,but I think increasingly that
term robots is too narrow toaround is in the area of smart
machines and relatedtechnologies to that, including
(04:51):
what the modeling technologiesyou know, like digital twins
that are hugely important in thegestation cycle, if you like,
of those smart machines and Idon't know if you are, if you
were or are a foodie at all.
Speaker 1 (05:07):
But let's say, when did the food world um enter your
, your professional life, notjust your kitchen um in in your
journey and, uh, there was quitea profound uh one, I think, in
terms of scale, in terms oftechnology, in terms of moving
parts, in terms of especially,let's say, moving things that
can can go bad quite quickly ifthey are in the wrong
(05:29):
temperature or are put on top ofeach other in the wrong order.
It's, it's very different frommoving.
I mean, probably also, clothesare tricky and a lot of other
materials are tricky, but food,specifically fresh produce, even
more makes it a veryinteresting challenge.
How did did that happen?
How did you roll into the worldof food?
Or maybe it was always a dreamto be working on food and
(05:50):
agriculture?
Speaker 2 (05:52):
So I suppose the place at which I or the point at
which I got involved in movingfood was when I joined Cardo and
that kind of happened byaccident.
Somebody gave my name to arecruitment consultant and they
persuaded me to have a chat withsomebody and they then in turn
(06:13):
persuaded me to go and have alook at their first generation
warehouse in Hatfield, and whenI stepped into that kind of
Aladdin's cave of technology Iwas completely blown away.
It was their first generationwarehouse in Hatfield, so it was
conveyor-based and a long wayfrom where the technology got to
during my time there.
Speaker 1 (06:33):
What was impressive when you stepped in, Just for
people to understand was it thenoise, was it the smell, was it
the size?
Just for you coming from whatyou were doing before.
Speaker 2 (06:42):
So I'd had some involvement, certainly with
robots before that, but nothinglike this.
This was like a roller coasteror a theme park of food.
It was, you know, boxes ofgroceries flying around over
many kilometers of conveyor andbeing, if you like, controlled
(07:04):
like you trains on a track,taking the right kind of turning
left and right at junctions andeventually ending up in the
back of vans.
And so it was all aboutmachines for orchestrating, if
you like, that picking andpacking of grocery, controlled
by some very sophisticatedsoftware.
(07:28):
And I was just blown away, youknow, I just thought this is
like the ultimate kind of trainset that I've stepped into and
it was mesmerizing, you know.
And I think a few months later Ijoined to do a one-year
consultancy project.
Uh, I ended up then joiningpermanently and I ended up
(07:48):
staying 15 years, which, uh,given the journey I had been on
with startups, I was very, youknow, surprised by, in a sense,
I it was never my game plan todo that, but it was a kind of a,
you know it was a love affairwith the technology, but also
with the culture.
You know of the company, whichwas incredibly inventive and
(08:11):
creative, and you know, verylike you know, what you
experience in startups and I hadexperienced in other startups,
even though by then, by 2006,when I joined, it was a, you
(08:33):
know, still quite a big company.
And so I think you know it wasthis blend of technology and
invention that attracted me.
(08:54):
And you know, during thatperiod I became CTO, I grew a
huge technology division andthen handed that over in the UK
Aria you know it's the sort ofthe sharp end of the innovation
spike or the innovation factoryat Ocado and we were looking at
all sorts of future potentialtechnologies in that division
(09:17):
that might be relevant andapplications that might be
relevant to the future of thecompany.
And then, in 2020, I left andthe end of 2020 and decided that
what I wanted to do was Rightin COVID when.
Well, yes, it was in the midstof COVID, which was an easy
Interesting times for onlineretail.
Speaker 1 (09:36):
Yeah, well, yes.
Speaker 2 (09:38):
I mean, it was an accelerant, certainly.
You know, arguably, onlinegrocery delivery became sort of
the fifth emergency service.
You know it was something thata lot of people relied on, but
also all sorts of kind ofautomated processes became
important because of trying tokeep people out of areas that
(10:02):
were dangerous, of areas thatwere dangerous.
And indeed, you know, there wasa desire for all sorts of other
smart machines to help dealwith the pandemic.
But, you know, one of thechallenges with smart machines
is that they have a differentgestation cycle.
To what does that mean?
Well as in you know, if youthink about how a new piece of
(10:25):
technology is ideated and thendesigned and created and tested
and prototyped and eventuallyput into production, you know
that's what I would refer to,its kind of gestation cycle, as
in.
You know, like you know, agestation cycle of an animal or
a human.
You know a gestation cycle ofan animal or a human, but you
(10:49):
know software and hardware has agestation or a life cycle and a
gestation cycle.
So it's, you know, and digitaltechnologies.
You know software and AI and soforth.
It's very different and thedigital and the physical worlds
are different and the reallyexciting things happen when the
two come together.
Speaker 1 (11:10):
It's such an interesting point, because I
think many people underestimateand I will put a video below of
what you also did at OcadoBecause retail online is
incredibly difficult, incrediblysmall margins, which is you
don't have the margins that alot of other technology
companies have.
If you're mostly in thesoftware side, it's a different
(11:31):
world, which means you need tobe super efficient and cutting
edge and, at scale, you need tomove a lot of things very, very,
very smartly, and that led toan incredible cycle of
innovation from you and othersas well.
How do you push the boundaries?
How do you do this smarter,faster, which 10 years ago, when
we thought, or 20 years maybe,when Okado started online
(11:53):
supermarket or online groceriesor retail in general, was more
almost like picking from thesupermarket itself, like the
same layout, the same, like itwas basically just moving it
online, which, of course, is notideal, just as we're still
stuck with the same typewriterletters on our laptop, which is
not ideal.
So, just to the immenseinnovation that has happened
there by merging or by usingboth the hardware, the robotics
(12:16):
which has changed dramatically,I think, over the last years and
the incredible sort of directorsoftware, we now have to do the
dances, and COVID has pushedthat and pushed it to the limits
, probably in certain extents.
But now, what do you see now,with the possibilities we have
(12:38):
on all the suite of technologiesit's not one, it's not just
digital twins, it's not justliving labs, it's not just the
sw, it's not just living labs,it's not just the swarms
robotics you've built.
If we, I don't know, lay of theland is maybe the wrong intro
into we're now talking inNovember 2024, it seems like
everything is accelerating, butwhat do you see?
Where those physical andnon-physical will touch, what
(13:01):
are exciting pieces for you?
Of course, we're interested infood and ag, but I'm also just
having you here to to see, okay,what should we know in food and
agriculture?
What is what is exciting to you?
What is interesting, what ismeaningful, what is not
interesting, what should weignore in your world?
Because you, you look so muchdeeper and further than I think
many of us do in the food and agspace when it comes to
technology gosh.
Speaker 2 (13:22):
Gosh, that's a good question or big question.
I think the first thing to sayis you know there is an
incredible frenzy and focus, youknow, on AI at the moment, and
arguably not AI in its inclusive, its wider sense, but in large
(13:43):
language models and generativeAI, and you know they're
incredibly important andpowerful technologies.
I was on the AI Council in theUK and helping to, you know, be
part of a group advisinggovernment, and so you know I'm
(14:03):
a fan of, of ai in all of itsflavors and it is a.
It's a much bigger toolkit, ifyou like, than just large
language models and generativeai.
But those are what is in asentence at the moment and
that's what's kind of captivated, if you like, uh, people's
attention, including, you know,uh, just, you know, normal
(14:25):
people for whom ai wasn't a bigpart of their world, and for
some it still isn't, but formany, uh, many more now it is
because it's become much moreaccessible.
Um, so ai has had its kind ofchat gpt moment.
Some other technologies haven'tyet.
But the point is, one of theother flavours of AI is embodied
(14:49):
AI, which is when you put thatkind of intelligence, if we want
to call it that, inside amachine, inside a smart machine,
and that's the smarts, if youlike, like in the smart machine.
But but it's a symbioticrelationship because the smart
machine also can be the arms andlegs of ai.
(15:11):
It can allow ai to escape, ifyou like, from inside uh, you
know, uh, your desktop orwhatever, and uh, or a server,
and get out into the physicalworld and and and do that with
scalability, scalability andagency.
So smart machines, uh, as theygo around, you know, often
(15:33):
they're controlled ororchestrated by sort of
centralized kind of ai systems,but, and sometimes they're
autonomous.
So sometimes you know, likeautonomous vehicles or drones,
you know they, they're able toperform without, you know,
minute to minute control.
They probably still need somedegree of orchestration because
(15:53):
they need to know what you wantthem to do.
But it's a blend of the two.
And so now we've got AI, andnow we've got smart machines,
and now we've got smart machines.
And then the third member ofthe Holy Trinity, if you like,
is what I would call syntheticenvironments.
So synthetic environments is afamily, another family like AI
(16:16):
and, in fact, another familylike smart machines, because
there are many different kindsof smart machines.
It's probably worth going onsynthetic environments just to
point out that you know, youoften say that you know, you
often say to people, you know.
So what robots do you have inyour home?
Well, I've done that.
And they go I don't have arobot at home.
I said, well, I think you do.
And they say what I said well,do you have a washing machine?
(16:39):
They say, oh, yes, I've got awashing machine, but that's as
robots anymore, you know.
So you know, a drone is a robot.
A 3D printer is a robot or asmart machine, an autonomous
vehicle.
In fact, a normal connected caris a very sophisticated smart
machine with hundreds of CPUs init.
You know controlling, you knoweverything that's happening.
(17:02):
So it's, you know.
There are lots of things out outthere, lots of smart machines
out there, hiding in plain sight, if you like.
You just don't necessarilythink of them as that, you know.
But so there is a whole familyof smart machines.
There's a whole family of AItechnologies, and then there's a
family of modeling technologiescalled synthetic environments,
and in that family it's got fivemembers.
(17:23):
It's simulations, emulations,visualizations, digital shadows
and digital twins, and onceagain they are very different,
they're complementary, but theyform a family and once again
there is an evolutionary cyclein there in the sense that you
(17:44):
know everyone a lot of peopleare talking about digital twins
now and it's in the zeitgeistand you know it's got people's
attention and that's great.
I'm a huge fan of digital twins,but unfortunately the term has
now been so hijacked Poor Henryhas had to listen to me bend his
ear about this because you, Isaid you know, unfortunately.
(18:04):
Well, a lot of people talkabout digital twins.
You know they don't meandigital twins.
Sometimes what they're talkingabout is just a static sort of
3d diagram or render of aphysical thing in the real world
.
That's a.
That's a powerful you knowthing.
It may be very useful.
It's not a digital you knowthing.
(18:24):
It may be very useful.
It's not a digital twin, youknow, and you know the key thing
is a digital twin has to havethat modeling element, you know.
It has to be able to predictthe future.
It's not just a data store,it's not just a sort of, you
know, a collection of data aboutsome physical asset, the
physical twin, collection ofdata about some physical asset,
(18:47):
the physical twin.
But then the other key thing isit needs to be connected to its
physical twin in bothdirections.
So if we think of an example ofa weather forecasting system,
that's a digital shadow.
It takes a lot of data in fromweather stations around the
world, it models it, it, itpredicts the weather in the
future, and then you know, we,we get it on the, you know on
(19:12):
the news, so to speak.
The.
That's a digital shadow becausethe connection is in one
direction.
What makes a true digital twin,in my terms, and most of the
terms, is that then, when youtake those insights and you use
it to control the real world,you know, then you've created a
digital twin.
So so, anyway it the point isthere is a family there of
(19:37):
technologies.
What they do is they allow youto under to understand how
complex systems work they.
They allow you to optimizethose systems.
They allow you to sort of delveback into the past.
They let you predict the future.
They let you understand whatthe implications are of the
(19:58):
interventions you want to makeon the physical world, whether
they're the intended or theunintended consequences, so
incredibly important in manyareas, including when you want
to innovate and model newtechnologies before they exist.
So if you're inventing a newkind of robot or a new kind of
(20:19):
smart machine, you know if youcan do the modeling in silico,
as it's called, ie in a digitalmodel, it's much cheaper and
faster and less risky thanbuilding prototypes in the
physical world.
But eventually, once you'vesort of hammered out the design
and you're confident that it'sgoing to work in this digital
(20:39):
model, then you start buildingthe physical prototypes and
putting them through their paces, and that's when you start
moving on to things like livinglabs as being environments in
which you can do that testing.
So there is a whole ecosystem,if you like, that supports this
blend of technologies.
But the really exciting alchemyhappens when you start cooking
(21:01):
with all of them together.
Where you start cooking withall of them together.
And, in fact, coming on to thesubject of food, my view on
transformative innovation andinvention is that it's very much
like great cooking.
You have to think about theutensils, so the technologies,
the building blocks, the talentand so forth.
(21:25):
You have to think about theutensils, so the tools, the
frameworks that allow you toassemble and manipulate those
raw ingredients.
But the recipes are incrediblyimportant, so the culture, the
ways of working, the creativity,the ways of solving problems,
(21:49):
so that it's that blend of theingredients, the utensils and
the recipes, just as it is incooking, that makes it
transformative.
And you need, if you reallywant to get the most out of it,
you need to work on all three ofthose together.
Speaker 1 (22:03):
So it's a sort of a, it's an amalgam, and yeah, and
when you sit with somebody likeHenry Dimbleby or with maybe
looking at a farm or at a foodoperator, of course coming from
the extreme high tech side ofthe food, like moving the food
as most efficiently as possibleto to our plates in the end
consumer Like what do you seewhen you look at the, when you
(22:25):
imagine I don't know how muchtime you spend on farms, but is
it I'm not saying laughable oryou would imagine what we could
do here, or does it get veryinteresting for you?
Or are you thinking, oh my God,this is also.
This really seems like we're in1950 in some cases.
Or what do you feel or sensewhen you're interacting with the
food and agriculture system, ifthat happens a lot, or if you
(22:49):
imagine an average farm in theUK or an average farm in your
area.
Speaker 2 (22:54):
So my first journey into agriculture and farming was
long before any of what I'vejust talked about happened.
My father, who had been acorporate lawyer all his life,
retired and suddenly announcedbecause he'd been very much
(23:20):
involved in the kind of thebirth of the Green Party in the
UK and was hugely interested inthe concept of organic food and
self-sufficiency, which was akind of a concept that had come
out then, and a famous bookcalled Of the Same Name by John
Seymour, you know, which he andmy mother read, and they were
(23:41):
captivated by that.
And they, they.
I was just about to go off touniversity and they said, oh,
we've decided we're going to goand buy, you know, a farm in
Wales and that's what they did.
So, just when most people, youknow, retiring at the age of 70,
you know, might have put theirfeet up and maybe done a bit of
(24:01):
gardening, you know, on theallotment, my father and my
mother and indeed all of us indifferent ways, headed off, you
know, to this hill farm in Walesand it was an extraordinary
eye-opener about just how, howchallenging, you know, farming
(24:22):
is, especially with animals, andyou know it's a 24 by 7, you
know job, and in all weathersand with all the
unpredictability that weatherand conditions bring, and I
think we get, we have beengetting better at that, but
still we've been getting better.
(24:43):
But you know I can tell you, ona on a hill farm in Wales, it's
pretty rugged.
On a hill farm in Wales it'spretty rugged, you know you do
get some sunshine but it's, youknow, snow, wind and hail is all
too common at that altitude andyou end up, you know, you, you,
you.
So it's, it's very challenging,but it was also an amazing
sandpit for creativity.
(25:04):
So you know, I used to spend mytime building machines, you
know, with my eldest brother,you know, welding stuff together
, making crop sprayers andmaking new machines for doing
things in the farmer.
But we were also doing thingslike building a biogas generator
to turn, you know, the pig shitinto gas, and anyway.
(25:25):
So it was a sandpit, you know,for, for, for starting to
explore different ways to, youknow, in a, in a kind of a very
kind of simplistic way, automateaspects of that but also and
unoptimize it, although weweren't building, you know,
digital twins of it, but the sothat was my first taste, if you
(25:46):
like, was my first taste, if youlike, and I came away from that
and it's still.
I look back on that as some ofthe sort of the four or five
most precious years of my life,you know, in terms of what I was
able to experience there, butalso with a huge degree of
admiration, if you like, forpeople who do that, you know,
day in, day out, you know forthe whole of their lives.
(26:08):
So then obviously, you know,got into food in a different way
.
But then when Henry, duringCOVID, asked me to join the
advisory board of the NationalFood Strategy, you know,
particularly in terms ofoffering kind of technology
advice, you know, I jumped at itbecause I just thought this is,
(26:30):
it's such an important.
It's important at the best oftimes, but it was incredibly
important, you know, during thepandemic.
So, but also not just the foodsystem as it is, but the fact
that we need to be, you know,finding ways, you know, for it
to be more sustainable, both forus as humans in terms of
healthy eating, but also interms of the impact on the
(26:53):
planet.
And obviously, if anybodylistening hasn't read the
National Food Strategy, I wouldcommend it to them.
It's amazing, but one of thethings in there, which I think
was chapter two, was aboutsystems thinking and was about
and one of the recommendationswas around, if you like the need
to create a data ecosystem forthe for the food system, and
(27:18):
perhaps come back to the secondone.
But but the first one isimportant because systems
thinking was incrediblyimportant.
You know, in terms of Ocado,what we did there, because you
need to optimize it in anend-to-end system, but it's also
incredibly important in allsorts of systems.
The food system is a verycomplex system of systems, but
(27:43):
it sits within a wider set ofsystems, whether that's energy
and transport and logistics andhealthcare and so on, and
international trade.
So our everyday lives are anamalgam of many different kinds
(28:05):
of systems and if we want tounderstand them better, you know
we model data and modeling areabsolutely key.
You know how we collect thedata, that that we need to build
the models, but then you knowbuilding the models and then
using those to answer questions,but also to do things like you
(28:29):
know, optimizing the behavior ofthose smart machines we talked
about earlier, using those toanswer questions but also to do
things like you know, optimisingthe behaviour of those smart
machines we talked about earlier.
And then, coming back to yourquestion, so if we when I look
at now, at agriculture, orindeed many other systems, from,
you know, transport to energy,to defence, to healthcare, you
(28:50):
know there are extraordinaryopportunities, you know, not
just for AI, but for smartmachines and for modelling and
for, you know, remote sensingand the internet of things, and
for living labs, labs, you know,and so it.
That's part one of the reasonswhy one of the first things, uh,
(29:14):
one of the organizations that I, I'm a government organizations
, I'm I'm still involved in,it's called the robotics growth
partnership, which advisesgovernment on smart machine, um,
uh, and robotic uh technologies, and uh, back in 2022, we
authored something called thecyber physical infrastructure
vision, which is really abouthow we can use that blend of
(29:38):
technologies that we've beentalking about, but at a national
scale, or, as I like to put it,how do we build a better lego
set for the UK or, ultimately,for the planet?
You know, because we need.
We need to change the way inwhich we build things, whether
that be infrastructure or publicservices, or products, all
(29:59):
kinds of products and services,because we need to do it in ways
that are, you know, moresustainable, more efficient and
more affordable.
So that's one of the thingsthat's really excited me in
recent years is that idea of howwe use all of these different
(30:23):
kind of technologies, but at anational scale and ultimately a
planetary scale.
And that includes, if we wantto zoom back to agriculture, you
know the opportunity, you knowto use automation and
(30:45):
optimization to improve yields,to improve sustainability and,
if you like, find new ways inwhich to turn photons from the
sun into calories, which is thegame that we're really playing
(31:05):
when we grow food but to findnew ways of doing that that are
not only healthier in terms ofthe products but also kinder to
the environment.
And one of the challenges weface with that is when you want
to compare conventionalagriculture with alternative
(31:29):
forms of agriculture, likevertical farming and algae and
lots of other things like that.
You know the problem is thatit's not a like for like, and
conventional agriculture, asmany of your listeners will know
, you know has a lot of hiddenexternalities.
So you know whether it's, youknow, pollution or soil erosion
(31:54):
or effects on flooding, orchemicals, pesticides you know
all of those things have animpact, if you like, on the
environment.
But when you are working out,if you like, the net cost, if
you like, not just to theconsumer but to society, they
often don't get taken intoaccount and therefore we need to
(32:15):
find ways to understand thoseexternalities and once again,
the National Food Strategytalked about that in detail.
And one of the ways we can dothat is through modeling.
You know, we can produce, youknow, potentially digital models
of a theoretical, if you like,farm or field or whatever it is
part of the outcome, and modelthose externalities and
(32:39):
understand, or start making astep towards understanding, what
their true cost is, if you like, in a whole of life sense, so
that when you, you know, you,you buy an apple, you know, yes,
you're buying an apple, for youknow whatever it costs, but
what does it cost, uh, if youlike, uh, the planet to to grow
(33:01):
that apple, and what would thatbe like if it was grown in a
different way?
Um, and obviously the whole pushtowards sustainable agriculture
is trying to get to somewherein between whereby we can have,
you know, the benefits ofconventional agriculture, but
without some of the currentnegative, unintended
consequences and because itfeels like there's such a
(33:25):
powerful tool set like thesethree families and within them,
um, like an immensely powerfultool set, um readily available
or coming available or, in manycases, um.
Speaker 1 (33:37):
But it also feels it's mostly ignored by food and
ag.
I mean, we see a lot ofinvestments going into ag tech
or we saw and it's cooling downa bit, um, robotics have a name
but it always feels verysuperficial in general, like it
doesn't feel part of a muchbigger vision and it sort of
seems I don't know why.
That is why food andagriculture hasn't embraced or
(34:01):
hasn't at least interacted withlike the simplest forms of just
modeling one or two farms andseeing what are the
externalities, what are the netnegative, what are the the net
negative and the net positivesthat it's producing, etc.
Like how do we, if we wouldchange this in management, what
would happen without changing it?
One of the many um issues withfarming is is you have a harvest
(34:22):
every year, or even less if youdo perennial, and and you just
the iteration cycle is so longand so slow that it's very, very
difficult to imagine or to havea culture of innovation.
A lot of this suite oftechnologies could fix part of
that, because you could iteratevery quickly or you could model
and understand what potentiallywill happen.
(34:44):
And yet I don't see peopletalking about, okay, what would
an agriculture system in the UK?
I mean, you've done some ofthat, I think, with the food
system, but what could it looklike?
Or what somebody on the podcastjust a few hours ago said.
We lack the imagination.
We cannot even imagine what afood system looks like and we're
just stuck in.
Okay, this just gets worse overtime.
Let's just put a few Band-Aidsleft and right.
(35:05):
What do you feel?
What's the blockage there?
Is it a mindset?
Is it age of farmers?
Is it we just don't care toomuch about?
Speaker 2 (35:17):
the countryside, like what's the?
Because the tools are there.
It seemed to be there and onlygetting better every minute.
So it it's.
It's a lot of different things.
So the first thing I would sayis look, the, the technology
does uh, you know is there butin many cases needs to mature
further.
You know, smart machines aregetting smarter, they're getting
more capable, you know,particularly on the back of ai,
(35:38):
not just in terms of theintelligence within them, but
the, the power of ai in thedesign process, if you like, of
the of the smart machine.
Speaker 1 (35:47):
So I really even before they're built, even
before they're built.
Speaker 2 (35:50):
So in the kind of stage of what you would call
computational design, theability for AIs to come up with
new, new solutions, if you like,for making smart machines
smarter and better, often byconnecting AIs to simulation
models, because when you putthose two together, all sorts of
(36:11):
things happen, because thesimulation allows the AI to
explore the real world.
You know the physical, itunderstands things like the laws
of physics and how things move,and gravity and the constraints
that come with that.
So it's really important thatyou do that.
Speaker 1 (36:27):
Yeah, gravity is good .
You know it's kind of difficult, it wastes a lot of energy when
you fight it.
Speaker 2 (36:29):
so you know, as poor old you do, that, yeah, gravity
is good.
You know it's kind of difficult, it wastes a lot of energy when
you fight it.
So you know, as poor old dronesknow.
But you know the point is youneed that blend of not just AI
going off, you know, generativeAI coming up with all sorts of
weird and wonderful solutionswhich actually wouldn't work in
physical.
(36:50):
But when you put it connectedand this is what has been
happening in, for instance, inarchitecture for some time I
mean, you've seen extraordinarybuildings that have been built
where you know the design oftenhas come from an AI but
connected to an engineeringmodel that actually puts
(37:14):
constraints, if you like, on thecreative juices of the AI to
make sure that it has structuralintegrity and could be built
and could be maintained and allthe other things.
So it's that combination.
Well, so there is work to bedone in making smart machines
smarter and more capable, andthat's something that you know I
(37:35):
and others are very muchinvolved in and we're about to
publish the latest version of astrategy for smart machines that
sort of takes us out to 2035.
So it's it's kind of work inprogress, but, but that's part
of it is the evolution of smartmachines, but the other part is,
(37:59):
as your other podcast wassaying, it is the sense of the
possible.
It is our ability to imaginefutures, if you like, where
these technologies have beendeployed and and how.
What would it look like, whatwould we want from it, what
problems could it solve, whatproblems might it create?
(38:20):
You know, and that, um, forthose of us who live and breathe
these technologies, it's easierto do some of that, but none of
us can do it perfectly.
but the um, uh, uh, but for youknow, for people who don't use
these technologies it's uh, it'svery uh, it's much more, it's
(38:41):
much more difficult, and that'swhy we need to find ways to
share um, uh, use cases, toshare videos, to share models of
it, so that people can immersethemselves in those kind of
futures and get an idea of, oh,look, what they did, you know,
over here on this farm or inthis you know industry, or
(39:05):
whatever, and how might mybusiness be different or similar
to that?
How might my business bedifferent or similar to that?
And once again, so modellinghas a role to play in helping
explore, helping peopleunderstand and have that sense
of the possible.
I'm going to have to stop andthrow a cat out of the room
(39:25):
because it's got in here andit's being incredibly noisy, so
I'm just going to throw this catout the side.
Sorry about that.
The joys of animals you know,there we go.
(39:46):
We've had many differentonesvening in different episodes
.
My cats like being involved inonline recordings, so where were
we?
Speaker 1 (40:00):
Yes, the imagination piece, the imagination.
Speaker 2 (40:03):
The sense of the possible is very important and
it could help with that.
Speaker 1 (40:07):
That's what you're saying.
We can absolutely help with thatShow and even immerse ourselves
, which is the word you use likewhat, and I I often feel we
miss that, like if we look at alandscape, we look at a most
landscapes are very degraded andand are not full of life,
definitely, but somehow, becausemaybe we've lived there for a
long time or we we're used tothe landscape without a lot of
(40:29):
trees or without a lot ofinsects, or somehow over time,
it slowly went down and, likeour baseline, just have been,
has been slipping away and we'relike this is normal and and
somehow I that it's it's notnormal, or at least not how it
could be in 10 years or 15 yearsor whatever cycles we we're
going through.
And so I think there's, apartfrom managing what we currently
(40:50):
have or managing it a waysmarter, there's a huge
potential of OK, what could itlook like?
What are the differentscenarios, what are different
options?
With climate modeling, withwhat species actually could live
here in 10 years, 15 years?
What management we need tomanage complex systems because a
(41:12):
farm, we've made them simplerover time, but we probably need
to complexify them over time,which brings a lot of challenges
and we have a smaller workforceand we need more machinery and
smart machinery to do that likeI don't think many people can
even imagine what a farm lookslike in 15 years.
We sort of think, okay, maybe asmarter tractor and that's sort
of hit, but of course it's notgoing to look like that well,
things like autonomous tractorsand and you know, drones for
(41:36):
microdosing well, yes, but it's.
Speaker 2 (41:39):
I was going to say that, yes, it's about you know
the machinery to work onconventional uh farms and fields
and so forth, you know.
But the sad reality is that, youknow, climate change means that
areas that used to be farmed inthat way, you know, many of
(42:00):
them, are becoming, you know,infertile.
And it's not about you know,putting an autonomous tractor on
there.
Actually, just the soil qualityand the amount of water that
you can, you can uh bring, youknow, and, and, and, and, the,
the, the kind, the kind of umcrops, if you like, you'd need
(42:24):
that would withstand, you know,the, perhaps the increased heat,
all of that um means that we'regoing to have to find new ways
also to grow things in areasthat would be called extreme
environments, if you like, orelse they're just going to
(42:46):
become deserts.
So I think there's a whole areaof exploration for this kind of
technology.
That is about how do you createways of growing food in
controlled conditions but in alandscape that is perhaps now no
longer amenable to conventionalagriculture, and people have
(43:10):
been starting to do that, youknow, using solar power and, and
you know, effectively, hugegreenhouses in, you know,
deserts and places like that andit's, you know, and how do you
recycle the water, and and soyou know that is a whole branch
of exploration but that's goingto be very important.
So it's it's not just about,you know, bringing technology to
(43:32):
bear, you know on, you know onfarming as we knew it, you know
it's also going to be about, uh,about that.
But then there's there's anotherstrand which is back to the
whole thing about how do we turnphotons into calories in in
other ways, um, which, um,somebody called I Boyd, who was
(43:52):
also mentioned by Henry, I think, on that podcast that he did.
A good friend of ours, you know, has done a huge amount of
thinking and in this area andtalks very eloquently about it,
and we are going to have to findalternative ways to do that.
(44:14):
You know which, obviously somepeople have been exploring with
lab based meat and other peoplehave been exploring with.
You know, can we turn thingslike algae, you know, into
palatable food?
You know there are manydifferent ways and other people
are trying to grow stuff inbioreactors, you know,
(44:34):
particularly because some ofthose technologies are going to
be important if we want to togrow food, you know, on other
planets, whether that's the moonor mars or whatever.
So so it isn't just about theearth.
Now, you know, people arethinking very much ahead to ways
to do that kind of foodcreation, if you like, in very
(44:56):
hostile conditions.
So it's a multi.
What I'm trying to say is it'sa multi-pronged or
multi-strategy approach to howwe might solve this very complex
problem of sustainable, healthyfood production.
And to do that, you know, forthe whole world, you know.
(45:17):
So we've got to make sure thatthe ways in which we would do it
don't, if you like, aresustainable and don't just
increase the problems that wehave that have led to climate
change.
Speaker 1 (45:30):
And to ask a few questions.
We always like to ask whatwould be your main message to
the investment world, thefinancial world either people
investing their own money orother people's money about these
technologies or this family ofdifferent technologies, but also
the work you've been doing.
(45:50):
What would be be?
Of course, they let's say we dothis on stage and they listen
to this, but they also forgetand people don't remember a lot
of things.
If there's one seed, one thingyou want them to remember and to
really to guide their workthrough, let's say what would be
the one thing you want them totake away?
Speaker 2 (46:12):
well, I'm not sure I'm going to stick to one, but
I'll stick to a few.
But I think the first thing Iwould say is that, in my
experience, everybody focuseswell, not everybody, but many
people focus on what I wouldcall the outcomes, the verticals
, you know, whether it'stechnologies or solutions or
sectors or whatever, and ofcourse, those are incredibly
(46:33):
important because that's wherethe results come.
But I'm much more interested inwhat I would call the
horizontals, the enablingtechnologies, the enabling
solutions, the, the things thatlet you scale, that let you be
more efficient, that let youoptimize, and we've been talking
about some of those.
(46:54):
So the modeling, you know, theAI, smart machines they are all
about things like scalabilityand efficiency and
reproducibility and resilience,you know, and that blend of
horizontal technologies thatcuts across and supports the
(47:15):
verticals often doesn't getenough investment and it doesn't
get enough, if you like,interest, I would say, you know,
from governments.
And there's an S at the end ofthat, you know, because it's a
recurrent problem.
Because it's a recurrentproblem Because it takes it's
like slow cooking, it takeslonger often to build those
(47:36):
horizontals and everybody wantsthe fast food, the kind of the
outcomes you know, and I thinkwe've.
That's got to change, because wedo need to build things better.
You know in it, wherever youlook, you look, you know whether
it's services, whether it'sproducts.
We need to build better.
(47:56):
We need to build formaintainability, we need to
build for recycling and reuse.
We need to change the materialsthat we use so that they are
more sustainable, and so, again,it's back to that systems
approach.
We need to look at it from asystems approach, but we need to
(48:18):
change the Lego.
We need to change the recipesthat we employ.
You know, in how we go aboutbuilding what we call the built
environment, which is basicallyeverything that is man-made
person-made, if you like.
So it's everything that isn'tthe natural environment.
Speaker 1 (48:37):
So you know we've built a lot of it now, and even
a big chunk of that is man-madeor managed.
Speaker 2 (48:43):
Oh well, exactly, or man-affected, you know, as in
you know.
So the natural environment andthe built environment have, you
know, crossed over, often in notsuch positive ways.
So the first thing I would sayis focus on the horizontals.
(49:03):
But also one of the thingsabout the horizontals is they
lead to intersectionality.
They often cut across.
They're the things that connectone industry maybe with another
.
So solutions in one industry orone sector that you can then
steal and adapt to solveproblems in another and I'm an
(49:25):
intersectionality nut.
I think it's incrediblyimportant.
I think, if we stay, if we, ifwe stay, you know, I used to say
to people when I was at Ocadoyou know, working for me, you
know, don't go out and hang outwith people and build networks
with people who work in onlinegrocery.
Go off and talk to people whobuild lawnmowers or nuclear
(49:46):
reactors or whatever, becauseyou know what you may learn in
those other areas.
You know you may then be thoseother areas you know you may
then be able to bring back andit may be incredibly powerful.
But it's also likely to be asource of of novel thought and
creation, because you know, ifyou're in an echo chamber of
people doing the same things asyou.
It's probably just, you know,the same old things will go
(50:08):
round and round.
So intersectionality is veryimportant and it's very
important in innovation andinvention.
So it's very important to helptackle some of this growing list
of exponential challenges thatwe've made for ourselves as a
species.
So it's you know, we need toget as many novel ideas into the
(50:31):
solutions as possible, andthose horizontal technologies
are part of the kind of the web,if you like, that connects all
those different verticals in thedifferent areas.
So that's the next thing Iwould say is invest in the kind
of the tools that let us dothings faster and more scalable,
(50:53):
but also let us join up thesolutions, uh, from different,
so that we're not buildingeverything from scratch, because
obviously, the more that we canbuild shared solutions across
different systems, you knowthat's good for everybody.
It's it takes cost out.
Uh, it means we don't have allsorts of different spares.
It means we don't.
It means we have greaterinteroperability.
(51:14):
You know this.
This happened in the pandemicwhen, you know, governments were
asking, you know, people in thesmart machine world, oh, we
want smart machines to lookafter and clean hospital red
zones.
You know what have you got.
And the answer is well, theanswer is twofold, one or
(51:34):
threefold.
First one, the gestation cycle,means that I talked about
earlier, means that you knowwe're not going to just be able
to create them overnight.
Secondly, the supply chain tobuild them, you know, is kind of
shut down, as it was, you know,during the pandemic.
And then the third thing is thatyou know most of these machines
(51:56):
are not built withinteroperability or reuse in
other areas in mind.
But there was a classic exampleof where you might want to take
things from one place where youhad them and move them over
here.
But the answer is you know, youneed to design for that, you
need to think about thatbeforehand and therefore the
more in which there arestandards and the kinds of
standards that actually we takefor granted in the software
(52:19):
world.
So open source and cloud hastransformed how we do software.
Well, we need the equivalentfor hardware.
Speaker 1 (52:27):
We need that same sort of, those same kind of
shared solutions for hardwareand what would you do if you
would be on the other side ofthe table, as as an investor?
Let's say, you wake up tomorrowmorning and you are in charge
of a significant amount of money.
We usually say, let's say, abillion, in this case pound
sterling, and I'm not lookingfor dollar amounts or exact
(52:47):
amounts, looking for what, ourbig buckets, what are things you
would focus on for a verylong-term potential investment
strategy.
But if you had to put that kindof money to work which is a bit
of a weird question, Iunderstand, but it's just.
I like to ask it because I'mwhen resources are almost
unconstrained, like what would,what do people prioritize and
(53:07):
what do they focus on?
Speaker 2 (53:09):
Well, I would definitely focus on exactly what
I've just been saying, which isthose enabling technologies and
solutions to accelerate manydifferent kinds of outcomes.
I like to think of them aswormholes.
They allow you to perhapsout-compete other players who
(53:37):
are doing things in atraditional way.
If you can find a wormhole thatacts as a shortcut to that and
the most powerful one I've evercome across is modeling, but,
you know, and, and inconjunction with AI now, as
we've been talking about, so Ithink, um, that would definitely
affect my investment.
I'm I think the the um thingsthat make you go faster and
(54:02):
accelerate are, you know, areincredibly powerful.
Secondly, um and this is verymuch affected, if you like, the
portfolio of startups I workwith.
Yes, climate technology in allits shapes and forms is
important.
One of the challenges is that alot of typical investment
(54:27):
sources don't really like tobuild or invest in hardware,
don't really like to build orinvest in hardware.
They prefer digital becauseit's cheap to build and to
distribute and it's quicker tobuild and less risky.
But, unfortunately, most of thethings that we need to deal
(54:48):
with the challenges in thephysical world that we created,
which is not just climate change, it's, you know, how do we have
affordable and sustainablehealth care around the world?
How do we, how do we feedeverybody around the world?
You know there are, and how dowe have, you know, better
resilience to future pandemics,because we're definitely going
to have them.
So, therefore, you know thatthat that's all going to involve
(55:11):
physical interventions in ourphysical world, and so we're
going to need hardware, and sowe've got to get better at
building it, like we've beentalking about, but we've also
got to encourage new forms ofinvestment in it, because if we
want these technologies andthere are some very exciting
(55:31):
climate technologies emergingwe're going to have to find a
way to fund not only theirinitial research but their scale
up and their adoption.
You know, and that that isoften where they hit trouble,
because and that is where nationstates, I think, need to step
in, you know, and that is wherenation states, I think, need to
(55:54):
step in, you know, and sovereignwealth funds and so forth,
because we it's no longer achallenge, I would argue for
traditional investment.
This is the future of ourspecies and this planet that
we're playing with here, andtherefore, you know there is
really nothing more important.
You know that we could investin, you know, than the solutions
(56:17):
to these kinds of exponentialchallenges, and we need to get
really serious about that.
If it's now 15 seconds tomidnight or whatever it is, and
you only have to listen, youknow.
You listen to the speeches atconferences like COP, and I
think I stopped listeningprobably a couple of them ago,
(56:38):
because one of my takeaways fromthat was there's barely any
talk about better ways ofcollecting data, you know, more
efficiently, but there's no talkabout better tools.
There's no talk about, you know,how do we create the sort of,
you know, the tools, both tounderstand the problems but also
(57:01):
to solve it.
You know there's lots of talkabout how technology will do it,
but we've got to startinnovating, if you like, on
behalf of our planet rather thanjust, if you like, the vested
interests of individualcompanies and shareholders and
even nation states.
(57:21):
So how we do that in a joinedup way, you know, is something
that we haven't got a solutionto at the moment.
You know we struggle to do that, you know, with healthcare, in
terms of the WHO, but we're notreally driving planetary scale
innovation and problem solving,you know, in a joined up way,
(57:45):
and we're going to have tobecause it doesn't, as we know,
respect borders, you know, andso, and I'm not sure, yeah, if a
billion would be enough tounlock it.
A billion is definitely notenough there, but maybe what one
can start doing with that istrying to nurture some of those
really key emergent climatetechnologies, but also maybe do
(58:13):
more of the modeling that allowsyou to demonstrate, if you like
, what those different solutionsbe.
At me, what you know, andthat's that's kind of what you
know an area that's alwaysinterested me is the whole idea
of building earth twins.
How do we build digital twinsat a planetary scale of both our
man-made systems and ournatural environment?
(58:34):
How do we collect the datascalably from, you know, on land
, in and under the sea, in theair?
You know how do we use spaceobservation and satellites to do
that?
But building, we need all thatto feed the earth twins.
But you know we need all thatto feed the earth twins.
But you know that's somethingthat a growing number of people
(58:56):
are starting to talk about andthere are some interesting
projects there, but it'scritically important because
we've left it so late tointervene.
We can't afford to get it wrong, and one of the things we've
been talking about is modelsallow you to explore those
futures.
As long as they're accurate andyou can run them at many times
(59:18):
faster than real time, so thatyou can you know you can
accelerate that exploration intothe future.
It's not like you have to waitten years for a model to model
ten years, you know you can runit many times faster than real
time, and and so it's a.
So I would definitely beputting some money into that,
(59:39):
not just because I think it'simportant, but because it's
going to be needed, so it's it'salso commercially, you know,
important and it's somethingwhere it could have, you know,
all manner of different impacts,both commercial and
environmental, and also theNational Food Strategy has got a
(01:00:00):
jolly good list of things toinvest in.
You know, in my experience, mostinnovation, as it's called and
I don't really like the wordinnovation, because I'm much
more interested in invention,which is more radical, but
(01:00:24):
because innovation tends to beincremental, most innovation is
focused on playing what I wouldcall the current games a little
bit better, and an example ofthat might be the fascination at
the moment with humanoid robots, which are incredibly complex,
(01:00:46):
costly machines, basically totry and mimic what humans do
using human tools and humanprocesses, with all the
limitations that come with those, and you know.
That's about finding a betterway to play the current game you
know of of human createdproducts and services.
I think the future lies in infinding new, completely new.
(01:01:10):
We need to change the game.
You know.
We need to to as I've beentalking about in this podcast
radically change the way that webuild solutions um, uh and uh,
if you know, because we need todo it faster, but also because
we need to do it moresustainably, but to do it with a
systems approach.
(01:01:31):
So don't't just find ways toautomate, if you like, what
we've done before, but thinkabout it, you know, from scratch
and think about it at a systemslevel.
Speaker 1 (01:01:44):
And so, once again, I would want to bring that to
everything in that innovationportfolio thinking, you know, in
terms of systems rather thanjust a set of individual
investments and it's a greatpoint to wrap up, and I want to
be conscious of your time, butstill ask one, one final
(01:02:04):
question on the magic wand whichwe always ask.
It might be actually the, theearth, uh, the earth twin or
earth twins, but it might besomething completely different.
If you so, we take away your,your investment portfolio, but
you do have the power to changeone thing, one thing, overnight.
We've had countless differentexamples, from extremely
(01:02:25):
practical changing the commonagriculture policy in europe to
the global mindset shift andeverything in between.
So definitely feel free to gowild if you want to, but if you
can change one thing overnight,what would that be?
Speaker 2 (01:02:43):
Okay, well, the.
I love that film.
I mean, mean, we've beentalking.
We touched for a brief momenton the idea.
Welcome to another episode today with technologist
innovator, inventor and a leaderin the.
I have to do it again.
Technologist innovator,inventor and a leader Welcome,
paul.
Hi, good to be here and firstof all shout out to Henry
Dimbleby, who put us in touchand who already mentioned you in
our two hour long deep diveDon't go and listen to that now,
people, I'll put a link belowif you haven't and who already
(00:24):
mentioned you in ourtwo-hour-long deep dive Don't go
and listen to that now, people,I'll put a link below if you
haven't listened to that yet butmentioned digital twins and
said you really, really have totalk to Paul.
And here we are, and I agree.
We had a pre-call and I had thepleasure of doing a bit of
background research into yourfascinating journey into
(00:44):
technology and, of course, theworld of food and agriculture.
You're definitely not deep inthe world of regenerative food
and agriculture, but you'veplayed a large role in moving a
lot of food like more food thanmost people will ever touch in
their lives, and so I would loveto unpack that what you see,
what you have noticed, what arethe opportunities of digital
(01:04):
twins and other deeptechnologies that we need, that
are already there that we mightnot even know how to use yet, so
I'm really looking forward tothat.
This will be a wide-ranginginterview and conversation, but
first of all starting with apersonal question in this case,
how did you roll into the worldof technology case?
(01:25):
How did you roll into the worldof of technology?
I can see background behind youa lot of different boxes and
things, probably full of cablesand all kinds of um things to
build.
But what were you always likethat?
Or how did you roll into intothe world of tech?
Speaker 2 (01:36):
well, I suppose as a child, um, you know when, when
many of my contemporaries werereading kind of I don't know,
innie b Blyton or books likethat, you know, I was normally
reading instruction manuals forcars and things like that and
taking stuff apart and puttingit together again and maybe
making new things out of thoseparts.
(01:58):
So I was always fascinated byhow things work and how things
are built and by machines of allsorts, and so that love of how
the world works kind of drove meto studying physics at
university.
(02:18):
But even then I was fascinatedby the kind of the early days of
personal computers.
But even then I was fascinatedby the early days of personal
computers and I nearly droppedout of my physics degree to do a
first startup.
But I always kind of knew thatI wanted to work with computers
and technology, and probablywithin startups, and so that led
(02:43):
me into the computer industryand sure enough, about two years
in the opportunity came to do afirst startup and and I grabbed
it.
So it's, it's really it's beenthat way from the start and, and
I suppose over that journey thetechnologies may have changed
and evolved, but many of thesort of many of the themes have
(03:08):
continued, particularly the theexcitement of working at the
interface of software andhardware systems.
You know things, things thatmove under software control are
much more exciting than just ona screen.
Speaker 1 (03:31):
And was it always the case for you like that
interface or that connectionbetween the hardware, physical
world and the software world, orhas that been gradually over
time, like your first startup orthe first companies you worked
at and with?
Was that predominantly thesoftware side or also the stuff
you could touch the hardwareside?
Speaker 2 (03:49):
No, I think earlier on the startups that I worked in
were more software focused, butI suppose then the physical
world started to come in interms of sort of the more
real-time applications.
So so software that interactswith the real world in terms of
(04:13):
control, and then that led on tothe whole area of what I would
now call smart machines.
Others might call them robots,but I think increasingly that
term robots is too narrow toaround is in the area of smart
machines and relatedtechnologies to that, including
(04:51):
what the modeling technologiesyou know, like digital twins
that are hugely important in thegestation cycle, if you like,
of those smart machines and Idon't know if you are, if you
were or are a foodie at all.
Speaker 1 (05:07):
But let's say, when did the food world um enter your
, your professional life, notjust your kitchen um in in your
journey and, uh, there was quitea profound uh one, I think, in
terms of scale, in terms oftechnology, in terms of moving
parts, in terms of especially,let's say, moving things that
can can go bad quite quickly ifthey are in the wrong
(05:29):
temperature or are put on top ofeach other in the wrong order.
It's, it's very different frommoving.
I mean, probably also, clothesare tricky and a lot of other
materials are tricky, but food,specifically fresh produce, even
more makes it a veryinteresting challenge.
How did did that happen?
How did you roll into the worldof food?
Or maybe it was always a dreamto be working on food and
(05:50):
agriculture?
Speaker 2 (05:52):
So I suppose the place at which I or the point at
which I got involved in movingfood was when I joined Cardo and
that kind of happened byaccident.
Somebody gave my name to arecruitment consultant and they
persuaded me to have a chat withsomebody and they then in turn
(06:13):
persuaded me to go and have alook at their first generation
warehouse in Hatfield, and whenI stepped into that kind of
Aladdin's cave of technology Iwas completely blown away.
It was their first generationwarehouse in Hatfield, so it was
conveyor-based and a long wayfrom where the technology got to
during my time there.
Speaker 1 (06:33):
What was impressive when you stepped in, Just for
people to understand was it thenoise, was it the smell, was it
the size?
Just for you coming from whatyou were doing before.
Speaker 2 (06:42):
So I'd had some involvement, certainly with
robots before that, but nothinglike this.
This was like a roller coasteror a theme park of food.
It was, you know, boxes ofgroceries flying around over
many kilometers of conveyor andbeing, if you like, controlled
(07:04):
like you trains on a track,taking the right kind of turning
left and right at junctions andeventually ending up in the
back of vans.
And so it was all aboutmachines for orchestrating, if
you like, that picking andpacking of grocery, controlled
by some very sophisticatedsoftware.
(07:28):
And I was just blown away, youknow, I just thought this is
like the ultimate kind of trainset that I've stepped into and
it was mesmerizing, you know.
And I think a few months later Ijoined to do a one-year
consultancy project.
Uh, I ended up then joiningpermanently and I ended up
(07:48):
staying 15 years, which, uh,given the journey I had been on
with startups, I was very, youknow, surprised by, in a sense,
I it was never my game plan todo that, but it was a kind of a,
you know it was a love affairwith the technology, but also
with the culture.
You know of the company, whichwas incredibly inventive and
(08:11):
creative, and you know, verylike you know, what you
experience in startups and I hadexperienced in other startups,
even though by then, by 2006,when I joined, it was a, you
(08:33):
know, still quite a big company.
And so I think you know it wasthis blend of technology and
invention that attracted me.
(08:54):
And you know, during thatperiod I became CTO, I grew a
huge technology division andthen handed that over in the UK
Aria you know it's the sort ofthe sharp end of the innovation
spike or the innovation factoryat Ocado and we were looking at
all sorts of future potentialtechnologies in that division
(09:17):
that might be relevant andapplications that might be
relevant to the future of thecompany.
And then, in 2020, I left andthe end of 2020 and decided that
what I wanted to do was Rightin COVID when.
Well, yes, it was in the midstof COVID, which was an easy
Interesting times for onlineretail.
Speaker 1 (09:36):
Yeah, well, yes.
Speaker 2 (09:38):
I mean, it was an accelerant, certainly.
You know, arguably, onlinegrocery delivery became sort of
the fifth emergency service.
You know it was something thata lot of people relied on, but
also all sorts of kind ofautomated processes became
important because of trying tokeep people out of areas that
(10:02):
were dangerous, of areas thatwere dangerous.
And indeed, you know, there wasa desire for all sorts of other
smart machines to help dealwith the pandemic.
But, you know, one of thechallenges with smart machines
is that they have a differentgestation cycle.
To what does that mean?
Well as in you know, if youthink about how a new piece of
(10:25):
technology is ideated and thendesigned and created and tested
and prototyped and eventuallyput into production, you know
that's what I would refer to,its kind of gestation cycle, as
in.
You know, like you know, agestation cycle of an animal or
a human.
You know a gestation cycle ofan animal or a human, but you
(10:49):
know software and hardware has agestation or a life cycle and a
gestation cycle.
So it's, you know, and digitaltechnologies.
You know software and AI and soforth.
It's very different and thedigital and the physical worlds
are different and the reallyexciting things happen when the
two come together.
Speaker 1 (11:10):
It's such an interesting point, because I
think many people underestimateand I will put a video below of
what you also did at OcadoBecause retail online is
incredibly difficult, incrediblysmall margins, which is you
don't have the margins that alot of other technology
companies have.
If you're mostly in thesoftware side, it's a different
(11:31):
world, which means you need tobe super efficient and cutting
edge and, at scale, you need tomove a lot of things very, very,
very smartly, and that led toan incredible cycle of
innovation from you and othersas well.
How do you push the boundaries?
How do you do this smarter,faster, which 10 years ago, when
we thought, or 20 years maybe,when Okado started online
(11:53):
supermarket or online groceriesor retail in general, was more
almost like picking from thesupermarket itself, like the
same layout, the same, like itwas basically just moving it
online, which, of course, is notideal, just as we're still
stuck with the same typewriterletters on our laptop, which is
not ideal.
So, just to the immenseinnovation that has happened
there by merging or by usingboth the hardware, the robotics
(12:16):
which has changed dramatically,I think, over the last years and
the incredible sort of directorsoftware, we now have to do the
dances, and COVID has pushedthat and pushed it to the limits
, probably in certain extents.
But now, what do you see now,with the possibilities we have
(12:38):
on all the suite of technologiesit's not one, it's not just
digital twins, it's not justliving labs, it's not just the
sw, it's not just living labs,it's not just the swarms
robotics you've built.
If we, I don't know, lay of theland is maybe the wrong intro
into we're now talking inNovember 2024, it seems like
everything is accelerating, butwhat do you see?
Where those physical andnon-physical will touch, what
(13:01):
are exciting pieces for you?
Of course, we're interested infood and ag, but I'm also just
having you here to to see, okay,what should we know in food and
agriculture?
What is what is exciting to you?
What is interesting, what ismeaningful, what is not
interesting, what should weignore in your world?
Because you, you look so muchdeeper and further than I think
many of us do in the food and agspace when it comes to
technology gosh.
Speaker 2 (13:22):
Gosh, that's a good question or big question.
I think the first thing to sayis you know there is an
incredible frenzy and focus, youknow, on AI at the moment, and
arguably not AI in its inclusive, its wider sense, but in large
(13:43):
language models and generativeAI, and you know they're
incredibly important andpowerful technologies.
I was on the AI Council in theUK and helping to, you know, be
part of a group advisinggovernment, and so you know I'm
(14:03):
a fan of, of ai in all of itsflavors and it is a.
It's a much bigger toolkit, ifyou like, than just large
language models and generativeai.
But those are what is in asentence at the moment and
that's what's kind of captivated, if you like, uh, people's
attention, including, you know,uh, just, you know, normal
(14:25):
people for whom ai wasn't a bigpart of their world, and for
some it still isn't, but formany, uh, many more now it is
because it's become much moreaccessible.
Um, so ai has had its kind ofchat gpt moment.
Some other technologies haven'tyet.
But the point is, one of theother flavours of AI is embodied
(14:49):
AI, which is when you put thatkind of intelligence, if we want
to call it that, inside amachine, inside a smart machine,
and that's the smarts, if youlike, like in the smart machine.
But but it's a symbioticrelationship because the smart
machine also can be the arms andlegs of ai.
(15:11):
It can allow ai to escape, ifyou like, from inside uh, you
know, uh, your desktop orwhatever, and uh, or a server,
and get out into the physicalworld and and and do that with
scalability, scalability andagency.
So smart machines, uh, as theygo around, you know, often
(15:33):
they're controlled ororchestrated by sort of
centralized kind of ai systems,but, and sometimes they're
autonomous.
So sometimes you know, likeautonomous vehicles or drones,
you know they, they're able toperform without, you know,
minute to minute control.
They probably still need somedegree of orchestration because
(15:53):
they need to know what you wantthem to do.
But it's a blend of the two.
And so now we've got AI, andnow we've got smart machines,
and now we've got smart machines.
And then the third member ofthe Holy Trinity, if you like,
is what I would call syntheticenvironments.
So synthetic environments is afamily, another family like AI
(16:16):
and, in fact, another familylike smart machines, because
there are many different kindsof smart machines.
It's probably worth going onsynthetic environments just to
point out that you know, youoften say that you know, you
often say to people, you know.
So what robots do you have inyour home?
Well, I've done that.
And they go I don't have arobot at home.
I said, well, I think you do.
And they say what I said well,do you have a washing machine?
(16:39):
They say, oh, yes, I've got awashing machine, but that's as
robots anymore, you know.
So you know, a drone is a robot.
A 3D printer is a robot or asmart machine, an autonomous
vehicle.
In fact, a normal connected caris a very sophisticated smart
machine with hundreds of CPUs init.
You know controlling, you knoweverything that's happening.
(17:02):
So it's, you know.
There are lots of things out outthere, lots of smart machines
out there, hiding in plain sight, if you like.
You just don't necessarilythink of them as that, you know.
But so there is a whole familyof smart machines.
There's a whole family of AItechnologies, and then there's a
family of modeling technologiescalled synthetic environments,
and in that family it's got fivemembers.
(17:23):
It's simulations, emulations,visualizations, digital shadows
and digital twins, and onceagain they are very different,
they're complementary, but theyform a family and once again
there is an evolutionary cyclein there in the sense that you
(17:44):
know everyone a lot of peopleare talking about digital twins
now and it's in the zeitgeistand you know it's got people's
attention and that's great.
I'm a huge fan of digital twins,but unfortunately the term has
now been so hijacked Poor Henryhas had to listen to me bend his
ear about this because you, Isaid you know, unfortunately.
(18:04):
Well, a lot of people talkabout digital twins.
You know they don't meandigital twins.
Sometimes what they're talkingabout is just a static sort of
3d diagram or render of aphysical thing in the real world
.
That's a.
That's a powerful you knowthing.
It may be very useful.
It's not a digital you knowthing.
(18:24):
It may be very useful.
It's not a digital twin, youknow, and you know the key thing
is a digital twin has to havethat modeling element, you know.
It has to be able to predictthe future.
It's not just a data store,it's not just a sort of, you
know, a collection of data aboutsome physical asset, the
physical twin, collection ofdata about some physical asset,
(18:47):
the physical twin.
But then the other key thing isit needs to be connected to its
physical twin in bothdirections.
So if we think of an example ofa weather forecasting system,
that's a digital shadow.
It takes a lot of data in fromweather stations around the
world, it models it, it, itpredicts the weather in the
future, and then you know, we,we get it on the, you know on
(19:12):
the news, so to speak.
The.
That's a digital shadow becausethe connection is in one
direction.
What makes a true digital twin,in my terms, and most of the
terms, is that then, when youtake those insights and you use
it to control the real world,you know, then you've created a
digital twin.
So so, anyway it the point isthere is a family there of
(19:37):
technologies.
What they do is they allow youto under to understand how
complex systems work they.
They allow you to optimizethose systems.
They allow you to sort of delveback into the past.
They let you predict the future.
They let you understand whatthe implications are of the
(19:58):
interventions you want to makeon the physical world, whether
they're the intended or theunintended consequences, so
incredibly important in manyareas, including when you want
to innovate and model newtechnologies before they exist.
So if you're inventing a newkind of robot or a new kind of
(20:19):
smart machine, you know if youcan do the modeling in silico,
as it's called, ie in a digitalmodel, it's much cheaper and
faster and less risky thanbuilding prototypes in the
physical world.
But eventually, once you'vesort of hammered out the design
and you're confident that it'sgoing to work in this digital
(20:39):
model, then you start buildingthe physical prototypes and
putting them through their paces, and that's when you start
moving on to things like livinglabs as being environments in
which you can do that testing.
So there is a whole ecosystem,if you like, that supports this
blend of technologies.
But the really exciting alchemyhappens when you start cooking
(21:01):
with all of them together.
Where you start cooking withall of them together.
And, in fact, coming on to thesubject of food, my view on
transformative innovation andinvention is that it's very much
like great cooking.
You have to think about theutensils, so the technologies,
the building blocks, the talentand so forth.
(21:25):
You have to think about theutensils, so the tools, the
frameworks that allow you toassemble and manipulate those
raw ingredients.
But the recipes are incrediblyimportant, so the culture, the
ways of working, the creativity,the ways of solving problems,
(21:49):
so that it's that blend of theingredients, the utensils and
the recipes, just as it is incooking, that makes it
transformative.
And you need, if you reallywant to get the most out of it,
you need to work on all three ofthose together.
Speaker 1 (22:03):
So it's a sort of a, it's an amalgam, and yeah, and
when you sit with somebody likeHenry Dimbleby or with maybe
looking at a farm or at a foodoperator, of course coming from
the extreme high tech side ofthe food, like moving the food
as most efficiently as possibleto to our plates in the end
consumer Like what do you seewhen you look at the, when you
(22:25):
imagine I don't know how muchtime you spend on farms, but is
it I'm not saying laughable oryou would imagine what we could
do here, or does it get veryinteresting for you?
Or are you thinking, oh my God,this is also.
This really seems like we're in1950 in some cases.
Or what do you feel or sensewhen you're interacting with the
food and agriculture system, ifthat happens a lot, or if you
(22:49):
imagine an average farm in theUK or an average farm in your
area.
Speaker 2 (22:54):
So my first journey into agriculture and farming was
long before any of what I'vejust talked about happened.
My father, who had been acorporate lawyer all his life,
retired and suddenly announcedbecause he'd been very much
(23:20):
involved in the kind of thebirth of the Green Party in the
UK and was hugely interested inthe concept of organic food and
self-sufficiency, which was akind of a concept that had come
out then, and a famous bookcalled Of the Same Name by John
Seymour, you know, which he andmy mother read, and they were
(23:41):
captivated by that.
And they, they.
I was just about to go off touniversity and they said, oh,
we've decided we're going to goand buy, you know, a farm in
Wales and that's what they did.
So, just when most people, youknow, retiring at the age of 70,
you know, might have put theirfeet up and maybe done a bit of
(24:01):
gardening, you know, on theallotment, my father and my
mother and indeed all of us indifferent ways, headed off, you
know, to this hill farm in Walesand it was an extraordinary
eye-opener about just how, howchallenging, you know, farming
(24:22):
is, especially with animals, andyou know it's a 24 by 7, you
know job, and in all weathersand with all the
unpredictability that weatherand conditions bring, and I
think we get, we have beengetting better at that, but
still we've been getting better.
(24:43):
But you know I can tell you, ona on a hill farm in Wales, it's
pretty rugged.
On a hill farm in Wales it'spretty rugged, you know you do
get some sunshine but it's, youknow, snow, wind and hail is all
too common at that altitude andyou end up, you know, you, you,
you.
So it's, it's very challenging,but it was also an amazing
sandpit for creativity.
(25:04):
So you know, I used to spend mytime building machines, you
know, with my eldest brother,you know, welding stuff together
, making crop sprayers andmaking new machines for doing
things in the farmer.
But we were also doing thingslike building a biogas generator
to turn, you know, the pig shitinto gas, and anyway.
(25:25):
So it was a sandpit, you know,for, for, for starting to
explore different ways to, youknow, in a, in a kind of a very
kind of simplistic way, automateaspects of that but also and
unoptimize it, although weweren't building, you know,
digital twins of it, but the sothat was my first taste, if you
(25:46):
like, was my first taste, if youlike, and I came away from that
and it's still.
I look back on that as some ofthe sort of the four or five
most precious years of my life,you know, in terms of what I was
able to experience there, butalso with a huge degree of
admiration, if you like, forpeople who do that, you know,
day in, day out, you know forthe whole of their lives.
(26:08):
So then obviously, you know,got into food in a different way
.
But then when Henry, duringCOVID, asked me to join the
advisory board of the NationalFood Strategy, you know,
particularly in terms ofoffering kind of technology
advice, you know, I jumped at itbecause I just thought this is,
(26:30):
it's such an important.
It's important at the best oftimes, but it was incredibly
important, you know, during thepandemic.
So, but also not just the foodsystem as it is, but the fact
that we need to be, you know,finding ways, you know, for it
to be more sustainable, both forus as humans in terms of
healthy eating, but also interms of the impact on the
(26:53):
planet.
And obviously, if anybodylistening hasn't read the
National Food Strategy, I wouldcommend it to them.
It's amazing, but one of thethings in there, which I think
was chapter two, was aboutsystems thinking and was about
and one of the recommendationswas around, if you like the need
to create a data ecosystem forthe for the food system, and
(27:18):
perhaps come back to the secondone.
But but the first one isimportant because systems
thinking was incrediblyimportant.
You know, in terms of Ocado,what we did there, because you
need to optimize it in anend-to-end system, but it's also
incredibly important in allsorts of systems.
The food system is a verycomplex system of systems, but
(27:43):
it sits within a wider set ofsystems, whether that's energy
and transport and logistics andhealthcare and so on, and
international trade.
So our everyday lives are anamalgam of many different kinds
(28:05):
of systems and if we want tounderstand them better, you know
we model data and modeling areabsolutely key.
You know how we collect thedata, that that we need to build
the models, but then you knowbuilding the models and then
using those to answer questions,but also to do things like you
(28:29):
know, optimizing the behavior ofthose smart machines we talked
about earlier, using those toanswer questions but also to do
things like you know, optimisingthe behaviour of those smart
machines we talked about earlier.
And then, coming back to yourquestion, so if we when I look
at now, at agriculture, orindeed many other systems, from,
you know, transport to energy,to defence, to healthcare, you
(28:50):
know there are extraordinaryopportunities, you know, not
just for AI, but for smartmachines and for modelling and
for, you know, remote sensingand the internet of things, and
for living labs, labs, you know,and so it.
That's part one of the reasonswhy one of the first things, uh,
(29:14):
one of the organizations that I, I'm a government organizations
, I'm I'm still involved in,it's called the robotics growth
partnership, which advisesgovernment on smart machine, um,
uh, and robotic uh technologies, and uh, back in 2022, we
authored something called thecyber physical infrastructure
vision, which is really abouthow we can use that blend of
(29:38):
technologies that we've beentalking about, but at a national
scale, or, as I like to put it,how do we build a better lego
set for the UK or, ultimately,for the planet?
You know, because we need.
We need to change the way inwhich we build things, whether
that be infrastructure or publicservices, or products, all
(29:59):
kinds of products and services,because we need to do it in ways
that are, you know, moresustainable, more efficient and
more affordable.
So that's one of the thingsthat's really excited me in
recent years is that idea of howwe use all of these different
(30:23):
kind of technologies, but at anational scale and ultimately a
planetary scale.
And that includes, if we wantto zoom back to agriculture, you
know the opportunity, you knowto use automation and
(30:45):
optimization to improve yields,to improve sustainability and,
if you like, find new ways inwhich to turn photons from the
sun into calories, which is thegame that we're really playing
(31:05):
when we grow food but to findnew ways of doing that that are
not only healthier in terms ofthe products but also kinder to
the environment.
And one of the challenges weface with that is when you want
to compare conventionalagriculture with alternative
(31:29):
forms of agriculture, likevertical farming and algae and
lots of other things like that.
You know the problem is thatit's not a like for like, and
conventional agriculture, asmany of your listeners will know
, you know has a lot of hiddenexternalities.
So you know whether it's, youknow, pollution or soil erosion
(31:54):
or effects on flooding, orchemicals, pesticides you know
all of those things have animpact, if you like, on the
environment.
But when you are working out,if you like, the net cost, if
you like, not just to theconsumer but to society, they
often don't get taken intoaccount and therefore we need to
(32:15):
find ways to understand thoseexternalities and once again,
the National Food Strategytalked about that in detail.
And one of the ways we can dothat is through modeling.
You know, we can produce, youknow, potentially digital models
of a theoretical, if you like,farm or field or whatever it is
part of the outcome, and modelthose externalities and
(32:39):
understand, or start making astep towards understanding, what
their true cost is, if you like, in a whole of life sense, so
that when you, you know, you,you buy an apple, you know, yes,
you're buying an apple, for youknow whatever it costs, but
what does it cost, uh, if youlike, uh, the planet to to grow
(33:01):
that apple, and what would thatbe like if it was grown in a
different way?
Um, and obviously the whole pushtowards sustainable agriculture
is trying to get to somewherein between whereby we can have,
you know, the benefits ofconventional agriculture, but
without some of the currentnegative, unintended
consequences and because itfeels like there's such a
(33:25):
powerful tool set like thesethree families and within them,
um, like an immensely powerfultool set, um readily available
or coming available or, in manycases, um.
Speaker 1 (33:37):
But it also feels it's mostly ignored by food and
ag.
I mean, we see a lot ofinvestments going into ag tech
or we saw and it's cooling downa bit, um, robotics have a name
but it always feels verysuperficial in general, like it
doesn't feel part of a muchbigger vision and it sort of
seems I don't know why.
That is why food andagriculture hasn't embraced or
(34:01):
hasn't at least interacted withlike the simplest forms of just
modeling one or two farms andseeing what are the
externalities, what are the netnegative, what are the the net
negative and the net positivesthat it's producing, etc.
Like how do we, if we wouldchange this in management, what
would happen without changing it?
One of the many um issues withfarming is is you have a harvest
(34:22):
every year, or even less if youdo perennial, and and you just
the iteration cycle is so longand so slow that it's very, very
difficult to imagine or to havea culture of innovation.
A lot of this suite oftechnologies could fix part of
that, because you could iteratevery quickly or you could model
and understand what potentiallywill happen.
(34:44):
And yet I don't see peopletalking about, okay, what would
an agriculture system in the UK?
I mean, you've done some ofthat, I think, with the food
system, but what could it looklike?
Or what somebody on the podcastjust a few hours ago said.
We lack the imagination.
We cannot even imagine what afood system looks like and we're
just stuck in.
Okay, this just gets worse overtime.
Let's just put a few Band-Aidsleft and right.
(35:05):
What do you feel?
What's the blockage there?
Is it a mindset?
Is it age of farmers?
Is it we just don't care toomuch about?
Speaker 2 (35:17):
the countryside, like what's the?
Because the tools are there.
It seemed to be there and onlygetting better every minute.
So it it's.
It's a lot of different things.
So the first thing I would sayis look, the, the technology
does uh, you know is there butin many cases needs to mature
further.
You know, smart machines aregetting smarter, they're getting
more capable, you know,particularly on the back of ai,
(35:38):
not just in terms of theintelligence within them, but
the, the power of ai in thedesign process, if you like, of
the of the smart machine.
Speaker 1 (35:47):
So I really even before they're built, even
before they're built.
Speaker 2 (35:50):
So in the kind of stage of what you would call
computational design, theability for AIs to come up with
new, new solutions, if you like,for making smart machines
smarter and better, often byconnecting AIs to simulation
models, because when you putthose two together, all sorts of
(36:11):
things happen, because thesimulation allows the AI to
explore the real world.
You know the physical, itunderstands things like the laws
of physics and how things move,and gravity and the constraints
that come with that.
So it's really important thatyou do that.
Speaker 1 (36:27):
Yeah, gravity is good .
You know it's kind of difficult, it wastes a lot of energy when
you fight it.
Speaker 2 (36:29):
so you know, as poor old you do, that, yeah, gravity
is good.
You know it's kind of difficult, it wastes a lot of energy when
you fight it.
So you know, as poor old dronesknow.
But you know the point is youneed that blend of not just AI
going off, you know, generativeAI coming up with all sorts of
weird and wonderful solutionswhich actually wouldn't work in
physical.
(36:50):
But when you put it connectedand this is what has been
happening in, for instance, inarchitecture for some time I
mean, you've seen extraordinarybuildings that have been built
where you know the design oftenhas come from an AI but
connected to an engineeringmodel that actually puts
(37:14):
constraints, if you like, on thecreative juices of the AI to
make sure that it has structuralintegrity and could be built
and could be maintained and allthe other things.
So it's that combination.
Well, so there is work to bedone in making smart machines
smarter and more capable, andthat's something that you know I
(37:35):
and others are very muchinvolved in and we're about to
publish the latest version of astrategy for smart machines that
sort of takes us out to 2035.
So it's it's kind of work inprogress, but, but that's part
of it is the evolution of smartmachines, but the other part is,
(37:59):
as your other podcast wassaying, it is the sense of the
possible.
It is our ability to imaginefutures, if you like, where
these technologies have beendeployed and and how.
What would it look like, whatwould we want from it, what
problems could it solve, whatproblems might it create?
(38:20):
You know, and that, um, forthose of us who live and breathe
these technologies, it's easierto do some of that, but none of
us can do it perfectly.
but the um, uh, uh, but for youknow, for people who don't use
these technologies it's uh, it'svery uh, it's much more, it's
(38:41):
much more difficult, and that'swhy we need to find ways to
share um, uh, use cases, toshare videos, to share models of
it, so that people can immersethemselves in those kind of
futures and get an idea of, oh,look, what they did, you know,
over here on this farm or inthis you know industry, or
(39:05):
whatever, and how might mybusiness be different or similar
to that?
How might my business bedifferent or similar to that?
And once again, so modellinghas a role to play in helping
explore, helping peopleunderstand and have that sense
of the possible.
I'm going to have to stop andthrow a cat out of the room
(39:25):
because it's got in here andit's being incredibly noisy, so
I'm just going to throw this catout the side.
Sorry about that.
The joys of animals you know,there we go.
(39:46):
We've had many differentonesvening in different episodes
.
My cats like being involved inonline recordings, so where were
we?
Speaker 1 (40:00):
Yes, the imagination piece, the imagination.
Speaker 2 (40:03):
The sense of the possible is very important and
it could help with that.
Speaker 1 (40:07):
That's what you're saying.
We can absolutely help with thatShow and even immerse ourselves
, which is the word you use likewhat, and I I often feel we
miss that, like if we look at alandscape, we look at a most
landscapes are very degraded andand are not full of life,
definitely, but somehow, becausemaybe we've lived there for a
long time or we we're used tothe landscape without a lot of
(40:29):
trees or without a lot ofinsects, or somehow over time,
it slowly went down and, likeour baseline, just have been,
has been slipping away and we'relike this is normal and and
somehow I that it's it's notnormal, or at least not how it
could be in 10 years or 15 yearsor whatever cycles we we're
going through.
And so I think there's, apartfrom managing what we currently
(40:50):
have or managing it a waysmarter, there's a huge
potential of OK, what could itlook like?
What are the differentscenarios, what are different
options?
With climate modeling, withwhat species actually could live
here in 10 years, 15 years?
What management we need tomanage complex systems because a
(41:12):
farm, we've made them simplerover time, but we probably need
to complexify them over time,which brings a lot of challenges
and we have a smaller workforceand we need more machinery and
smart machinery to do that likeI don't think many people can
even imagine what a farm lookslike in 15 years.
We sort of think, okay, maybe asmarter tractor and that's sort
of hit, but of course it's notgoing to look like that well,
things like autonomous tractorsand and you know, drones for
(41:36):
microdosing well, yes, but it's.
Speaker 2 (41:39):
I was going to say that, yes, it's about you know
the machinery to work onconventional uh farms and fields
and so forth, you know.
But the sad reality is that, youknow, climate change means that
areas that used to be farmed inthat way, you know, many of
(42:00):
them, are becoming, you know,infertile.
And it's not about you know,putting an autonomous tractor on
there.
Actually, just the soil qualityand the amount of water that
you can, you can uh bring, youknow, and, and, and, and, the,
the, the kind, the kind of umcrops, if you like, you'd need
(42:24):
that would withstand, you know,the, perhaps the increased heat,
all of that um means that we'regoing to have to find new ways
also to grow things in areasthat would be called extreme
environments, if you like, orelse they're just going to
(42:46):
become deserts.
So I think there's a whole areaof exploration for this kind of
technology.
That is about how do you createways of growing food in
controlled conditions but in alandscape that is perhaps now no
longer amenable to conventionalagriculture, and people have
(43:10):
been starting to do that, youknow, using solar power and, and
you know, effectively, hugegreenhouses in, you know,
deserts and places like that andit's, you know, and how do you
recycle the water, and and soyou know that is a whole branch
of exploration but that's goingto be very important.
So it's it's not just about,you know, bringing technology to
(43:32):
bear, you know on, you know onfarming as we knew it, you know
it's also going to be about, uh,about that.
But then there's there's anotherstrand which is back to the
whole thing about how do we turnphotons into calories in in
other ways, um, which, um,somebody called I Boyd, who was
(43:52):
also mentioned by Henry, I think, on that podcast that he did.
A good friend of ours, you know, has done a huge amount of
thinking and in this area andtalks very eloquently about it,
and we are going to have to findalternative ways to do that.
(44:14):
You know which, obviously somepeople have been exploring with
lab based meat and other peoplehave been exploring with.
You know, can we turn thingslike algae, you know, into
palatable food?
You know there are manydifferent ways and other people
are trying to grow stuff inbioreactors, you know,
(44:34):
particularly because some ofthose technologies are going to
be important if we want to togrow food, you know, on other
planets, whether that's the moonor mars or whatever.
So so it isn't just about theearth.
Now, you know, people arethinking very much ahead to ways
to do that kind of foodcreation, if you like, in very
(44:56):
hostile conditions.
So it's a multi.
What I'm trying to say is it'sa multi-pronged or
multi-strategy approach to howwe might solve this very complex
problem of sustainable, healthyfood production.
And to do that, you know, forthe whole world, you know.
(45:17):
So we've got to make sure thatthe ways in which we would do it
don't, if you like, aresustainable and don't just
increase the problems that wehave that have led to climate
change.
Speaker 1 (45:30):
And to ask a few questions.
We always like to ask whatwould be your main message to
the investment world, thefinancial world either people
investing their own money orother people's money about these
technologies or this family ofdifferent technologies, but also
the work you've been doing.
(45:50):
What would be be?
Of course, they let's say we dothis on stage and they listen
to this, but they also forgetand people don't remember a lot
of things.
If there's one seed, one thingyou want them to remember and to
really to guide their workthrough, let's say what would be
the one thing you want them totake away?
Speaker 2 (46:12):
well, I'm not sure I'm going to stick to one, but
I'll stick to a few.
But I think the first thing Iwould say is that, in my
experience, everybody focuseswell, not everybody, but many
people focus on what I wouldcall the outcomes, the verticals
, you know, whether it'stechnologies or solutions or
sectors or whatever, and ofcourse, those are incredibly
(46:33):
important because that's wherethe results come.
But I'm much more interested inwhat I would call the
horizontals, the enablingtechnologies, the enabling
solutions, the, the things thatlet you scale, that let you be
more efficient, that let youoptimize, and we've been talking
about some of those.
(46:54):
So the modeling, you know, theAI, smart machines they are all
about things like scalabilityand efficiency and
reproducibility and resilience,you know, and that blend of
horizontal technologies thatcuts across and supports the
(47:15):
verticals often doesn't getenough investment and it doesn't
get enough, if you like,interest, I would say, you know,
from governments.
And there's an S at the end ofthat, you know, because it's a
recurrent problem.
Because it's a recurrentproblem Because it takes it's
like slow cooking, it takeslonger often to build those
(47:36):
horizontals and everybody wantsthe fast food, the kind of the
outcomes you know, and I thinkwe've.
That's got to change, because wedo need to build things better.
You know in it, wherever youlook, you look, you know whether
it's services, whether it'sproducts.
We need to build better.
(47:56):
We need to build formaintainability, we need to
build for recycling and reuse.
We need to change the materialsthat we use so that they are
more sustainable, and so, again,it's back to that systems
approach.
We need to look at it from asystems approach, but we need to
(48:18):
change the Lego.
We need to change the recipesthat we employ.
You know, in how we go aboutbuilding what we call the built
environment, which is basicallyeverything that is man-made
person-made, if you like.
So it's everything that isn'tthe natural environment.
Speaker 1 (48:37):
So you know we've built a lot of it now, and even
a big chunk of that is man-madeor managed.
Speaker 2 (48:43):
Oh well, exactly, or man-affected, you know, as in
you know.
So the natural environment andthe built environment have, you
know, crossed over, often in notsuch positive ways.
So the first thing I would sayis focus on the horizontals.
(49:03):
But also one of the thingsabout the horizontals is they
lead to intersectionality.
They often cut across.
They're the things that connectone industry maybe with another
.
So solutions in one industry orone sector that you can then
steal and adapt to solveproblems in another and I'm an
(49:25):
intersectionality nut.
I think it's incrediblyimportant.
I think, if we stay, if we, ifwe stay, you know, I used to say
to people when I was at Ocadoyou know, working for me, you
know, don't go out and hang outwith people and build networks
with people who work in onlinegrocery.
Go off and talk to people whobuild lawnmowers or nuclear
(49:46):
reactors or whatever, becauseyou know what you may learn in
those other areas.
You know you may then be thoseother areas you know you may
then be able to bring back andit may be incredibly powerful.
But it's also likely to be asource of of novel thought and
creation, because you know, ifyou're in an echo chamber of
people doing the same things asyou.
It's probably just, you know,the same old things will go
(50:08):
round and round.
So intersectionality is veryimportant and it's very
important in innovation andinvention.
So it's very important to helptackle some of this growing list
of exponential challenges thatwe've made for ourselves as a
species.
So it's you know, we need toget as many novel ideas into the
(50:31):
solutions as possible, andthose horizontal technologies
are part of the kind of the web,if you like, that connects all
those different verticals in thedifferent areas.
So that's the next thing Iwould say is invest in the kind
of the tools that let us dothings faster and more scalable,
(50:53):
but also let us join up thesolutions, uh, from different,
so that we're not buildingeverything from scratch, because
obviously, the more that we canbuild shared solutions across
different systems, you knowthat's good for everybody.
It's it takes cost out.
Uh, it means we don't have allsorts of different spares.
It means we don't.
It means we have greaterinteroperability.
(51:14):
You know this.
This happened in the pandemicwhen, you know, governments were
asking, you know, people in thesmart machine world, oh, we
want smart machines to lookafter and clean hospital red
zones.
You know what have you got.
And the answer is well, theanswer is twofold, one or
(51:34):
threefold.
First one, the gestation cycle,means that I talked about
earlier, means that you knowwe're not going to just be able
to create them overnight.
Secondly, the supply chain tobuild them, you know, is kind of
shut down, as it was, you know,during the pandemic.
And then the third thing is thatyou know most of these machines
(51:56):
are not built withinteroperability or reuse in
other areas in mind.
But there was a classic exampleof where you might want to take
things from one place where youhad them and move them over
here.
But the answer is you know, youneed to design for that, you
need to think about thatbeforehand and therefore the
more in which there arestandards and the kinds of
standards that actually we takefor granted in the software
(52:19):
world.
So open source and cloud hastransformed how we do software.
Well, we need the equivalentfor hardware.
Speaker 1 (52:27):
We need that same sort of, those same kind of
shared solutions for hardwareand what would you do if you
would be on the other side ofthe table, as as an investor?
Let's say, you wake up tomorrowmorning and you are in charge
of a significant amount of money.
We usually say, let's say, abillion, in this case pound
sterling, and I'm not lookingfor dollar amounts or exact
(52:47):
amounts, looking for what, ourbig buckets, what are things you
would focus on for a verylong-term potential investment
strategy.
But if you had to put that kindof money to work which is a bit
of a weird question, Iunderstand, but it's just.
I like to ask it because I'mwhen resources are almost
unconstrained, like what would,what do people prioritize and
(53:07):
what do they focus on?
Speaker 2 (53:09):
Well, I would definitely focus on exactly what
I've just been saying, which isthose enabling technologies and
solutions to accelerate manydifferent kinds of outcomes.
I like to think of them aswormholes.
They allow you to perhapsout-compete other players who
(53:37):
are doing things in atraditional way.
If you can find a wormhole thatacts as a shortcut to that and
the most powerful one I've evercome across is modeling, but,
you know, and, and inconjunction with AI now, as
we've been talking about, so Ithink, um, that would definitely
affect my investment.
I'm I think the the um thingsthat make you go faster and
(54:02):
accelerate are, you know, areincredibly powerful.
Secondly, um and this is verymuch affected, if you like, the
portfolio of startups I workwith.
Yes, climate technology in allits shapes and forms is
important.
One of the challenges is that alot of typical investment
(54:27):
sources don't really like tobuild or invest in hardware,
don't really like to build orinvest in hardware.
They prefer digital becauseit's cheap to build and to
distribute and it's quicker tobuild and less risky.
But, unfortunately, most of thethings that we need to deal
(54:48):
with the challenges in thephysical world that we created,
which is not just climate change, it's, you know, how do we have
affordable and sustainablehealth care around the world?
How do we, how do we feedeverybody around the world?
You know there are, and how dowe have, you know, better
resilience to future pandemics,because we're definitely going
to have them.
So, therefore, you know thatthat that's all going to involve
(55:11):
physical interventions in ourphysical world, and so we're
going to need hardware, and sowe've got to get better at
building it, like we've beentalking about, but we've also
got to encourage new forms ofinvestment in it, because if we
want these technologies andthere are some very exciting
(55:31):
climate technologies emergingwe're going to have to find a
way to fund not only theirinitial research but their scale
up and their adoption.
You know, and that that isoften where they hit trouble,
because and that is where nationstates, I think, need to step
in, you know, and that is wherenation states, I think, need to
(55:54):
step in, you know, and sovereignwealth funds and so forth,
because we it's no longer achallenge, I would argue for
traditional investment.
This is the future of ourspecies and this planet that
we're playing with here, andtherefore, you know there is
really nothing more important.
You know that we could investin, you know, than the solutions
(56:17):
to these kinds of exponentialchallenges, and we need to get
really serious about that.
If it's now 15 seconds tomidnight or whatever it is, and
you only have to listen, youknow.
You listen to the speeches atconferences like COP, and I
think I stopped listeningprobably a couple of them ago,
(56:38):
because one of my takeaways fromthat was there's barely any
talk about better ways ofcollecting data, you know, more
efficiently, but there's no talkabout better tools.
There's no talk about, you know,how do we create the sort of,
you know, the tools, both tounderstand the problems but also
(57:01):
to solve it.
You know there's lots of talkabout how technology will do it,
but we've got to startinnovating, if you like, on
behalf of our planet rather thanjust, if you like, the vested
interests of individualcompanies and shareholders and
even nation states.
(57:21):
So how we do that in a joinedup way, you know, is something
that we haven't got a solutionto at the moment.
You know we struggle to do that, you know, with healthcare, in
terms of the WHO, but we're notreally driving planetary scale
innovation and problem solving,you know, in a joined up way,
(57:45):
and we're going to have tobecause it doesn't, as we know,
respect borders, you know, andso, and I'm not sure, yeah, if a
billion would be enough tounlock it.
A billion is definitely notenough there, but maybe what one
can start doing with that istrying to nurture some of those
really key emergent climatetechnologies, but also maybe do
(58:13):
more of the modeling that allowsyou to demonstrate, if you like
, what those different solutionsbe.
At me, what you know, andthat's that's kind of what you
know an area that's alwaysinterested me is the whole idea
of building earth twins.
How do we build digital twinsat a planetary scale of both our
man-made systems and ournatural environment?
(58:34):
How do we collect the datascalably from, you know, on land
, in and under the sea, in theair?
You know how do we use spaceobservation and satellites to do
that?
But building, we need all thatto feed the earth twins.
But you know we need all thatto feed the earth twins.
But you know that's somethingthat a growing number of people
(58:56):
are starting to talk about andthere are some interesting
projects there, but it'scritically important because
we've left it so late tointervene.
We can't afford to get it wrong, and one of the things we've
been talking about is modelsallow you to explore those
futures.
As long as they're accurate andyou can run them at many times
(59:18):
faster than real time, so thatyou can you know you can
accelerate that exploration intothe future.
It's not like you have to waitten years for a model to model
ten years, you know you can runit many times faster than real
time, and and so it's a.
So I would definitely beputting some money into that,
(59:39):
not just because I think it'simportant, but because it's
going to be needed, so it's it'salso commercially, you know,
important and it's somethingwhere it could have, you know,
all manner of different impacts,both commercial and
environmental, and also theNational Food Strategy has got a
(01:00:00):
jolly good list of things toinvest in.
You know, in my experience, mostinnovation, as it's called and
I don't really like the wordinnovation, because I'm much
more interested in invention,which is more radical, but
(01:00:24):
because innovation tends to beincremental, most innovation is
focused on playing what I wouldcall the current games a little
bit better, and an example ofthat might be the fascination at
the moment with humanoid robots, which are incredibly complex,
(01:00:46):
costly machines, basically totry and mimic what humans do
using human tools and humanprocesses, with all the
limitations that come with those, and you know.
That's about finding a betterway to play the current game you
know of of human createdproducts and services.
I think the future lies in infinding new, completely new.
(01:01:10):
We need to change the game.
You know.
We need to to as I've beentalking about in this podcast
radically change the way that webuild solutions um, uh and uh,
if you know, because we need todo it faster, but also because
we need to do it moresustainably, but to do it with a
systems approach.
(01:01:31):
So don't't just find ways toautomate, if you like, what
we've done before, but thinkabout it, you know, from scratch
and think about it at a systemslevel.
Speaker 1 (01:01:44):
And so, once again, I would want to bring that to
everything in that innovationportfolio thinking, you know, in
terms of systems rather thanjust a set of individual
investments and it's a greatpoint to wrap up, and I want to
be conscious of your time, butstill ask one, one final
(01:02:04):
question on the magic wand whichwe always ask.
It might be actually the, theearth, uh, the earth twin or
earth twins, but it might besomething completely different.
If you so, we take away your,your investment portfolio, but
you do have the power to changeone thing, one thing, overnight.
We've had countless differentexamples, from extremely
(01:02:25):
practical changing the commonagriculture policy in europe to
the global mindset shift andeverything in between.
So definitely feel free to gowild if you want to, but if you
can change one thing overnight,what would that be?
Speaker 2 (01:02:43):
Okay, well, the.
I love that film.
I mean, mean, we've beentalking.
We touched for a brief momenton the idea.
Popular Podcasts
Fudd Around And Find Out
UConn basketball star Azzi Fudd brings her championship swag to iHeart Women’s Sports with Fudd Around and Find Out, a weekly podcast that takes fans along for the ride as Azzi spends her final year of college trying to reclaim the National Championship and prepare to be a first round WNBA draft pick. Ever wonder what it’s like to be a world-class athlete in the public spotlight while still managing schoolwork, friendships and family time? It’s time to Fudd Around and Find Out!
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by audiochuck Media Company.
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.