Ep4: Who Controls Your AI? Europe's Sovereignty Crisis and the Race to Build Independent Tools

Jonny and Tom dig into the uncomfortable reality of Europe's dependence on US-controlled AI — sparked by Anthropic's Fable 5 being banned outside America days after launch. They explore what the dystopian narrative Europe 2031 predicts for the continent, debate London's surprising rise as an AI talent hub, wrestle with what it means to build a business in the age of AI, and Tom reveals his open-source design tooling project, Kazam.

0:00 — Introduction & the content pipeline
Jonny and Tom kick off the episode and share their very low-tech approach to collecting topics throughout the week — a Slack thread — which immediately surfaced the biggest AI story of the moment.

1:27 — Fable 5, banned in minutes: the sovereignty wake-up call
Anthropic's most powerful model yet was blocked for all non-US citizens within four days of launch. What does it mean when critical infrastructure can be switched off by a foreign government overnight?

4:42 — Europe 2031: a cautionary tale
Jonny breaks down the 18,000-word narrative podcast europe2031.ai — a near-future story of Europe's failure to build sovereign AI capability, its dependence on open models that eventually disappear, and the grim choice between American and Chinese technology.

9:53 — What does this actually mean for you?
Tom pushes for practical takeaways. If AI vanished tomorrow it'd probably be fine — but the longer dependency grows, the harder it becomes to function without it. From job market effects to information control, the risks are subtle but compounding.

13:21 — Practical steps: diversify your AI diet
Rather than avoiding AI, Jonny argues for being intentional about where you get it from. Use open-weight models, try Mistral, explore OpenRouter — don't build critical dependency on US big tech whose incentives may not align with yours.

15:36 — The acceleration problem and European incentives
Tom reflects on how the structural incentives driving US and Chinese AI development will naturally widen the gap with Europe — even without any bad intentions from the US side.

17:18 — London rising: DeepMind, Cursor, and the talent story
Cursor just announced a 200-person London office. Tom traces how Demis Hassabis's insistence on keeping a London team when Google acquired DeepMind seeded a generation of AI talent — now paying dividends for the whole city.

19:14 — Nuance: AI isn't just bad or just good
Jonny argues for holding the tension — concentration of power is genuinely worrying, but the potential benefits to humanity (drug discovery, scientific breakthroughs) are also real. The podcast's purpose is navigation, not evangelism.

22:51 — Building a business in the AI era
The guys get into the meta-question: what is the right business model for an AI consultancy? Training vs. capability delivery, day rates vs. outcome-based pricing, what to give away free vs. what to keep — and why neither of them really has the answer yet.

35:24 — Tom's open-source project: Kazam
Tom reveals what he's been building — a lightweight open-source design framework for creating portable, lo-fi design micro-tools as single HTML files. No npm, no server, no dependencies. Inspired by a workshop where 60 designers built 60 tools in 3 hours.

Links
europe2031.ai
Jacob Heftmann
grillitype.com
Our 1 page HTML tool example: https://dither-shape-p.ooda.run/

Find us at nearfuture.works

Creators and Guests

Host

Jonny Burch

Co-founder of Near Future

Host

Tom Harman

Co-founder of Near Future

What is Near Future Podcast?

A regular podcast covering design and AI from the founders of Near Future, a boutique AI consultancy focused on teams that care about craft. We cover both what we're seeing on the ground and industry trends, ways of working and occasional guests from the design world.

Jonny: Hi, everyone, and welcome to the
next episode of the Near Future Podcast.

This is Jonny and Tom.

We're checking in after another week
of AI action in the world and for us.

So we're gonna get straight
into it once again.

And, one thing that we have tried to
do since our last recording is become

a little bit more professional and
build a bit of a content pipeline or

topic pipeline for our own podcast,
so we're not scrambling around

trying to find topics to talk about.

So Tom, do you wanna talk
about the, our strategy so far?

Yeah.

Tom: basic.

imagine a Slack channel

Jonny: Zero AI involved in it.

Tom: zero AI so far.

imagine a Slack channel, and then
we have a thread in there that just

captures up we find across the week
that we might wanna talk about or

might be interesting to talk about.

but I think the…

maybe the irony or the, the interesting
learning we had is we started this

a week ago, the first thing that
enters that channel is the news

of Anthropic releasing Fable 5.

And then within days it was, it was banned
in the UK, where we're recording this.

banned in, I guess, every
country outside the US.

so, what quickly turned from a potentially
interesting topic to talk about, sort

of evolved into questions about European
sovereignty and, these questions of e-

where, where are AI tools built and where
they, who controls who has access to them.

may actually lead us onto one of the
topics we wanted to talk about today.

but that, that is a quick overview
of our, amazingly high tech content

pipeline topic generator thing.

and maybe in the future that
will, that will lead to something

that we can actually use AI to,

Jonny: Yeah

Tom: know, pull together more
of a cohesive approach to, but

right now it's, bag of things.

So anyway, on the topic of topics to
talk about, how about, as a theme, Jonny?

And I know you shared with me earlier
in the week about Europe 2031, which

is, would love, love for you to tell
us a bit about and then, perhaps we can

talk through some reflections on that

Jonny: Yeah.

Yeah, and I think, I think,
sovereignty is an enormous

question at the moment right now.

Sovereignty has not been something
that a lot of people, I think, have

wanted to necessarily, talk about,
or it's been a very difficult topic.

And I feel like over the last Maybe
six months, but especially this week,

it's become something else entirely.

We're seeing an enormous amount of
power accrue to very few people around

this new technology that's e-emerging.

in fact, the other thing that happened
this week is xAI, SpaceX went public.

Elon Musk became the first trillionaire
in the world, and straight after that,

he bought Cursor, who I think we're gonna
talk about potentially in, later on.

But Cursor, which is one of the most
popular, IDEs, and they're building

models, and now they're building
models using all of Elon's compute,

and actually even more of the ecosystem
around how software is built and

where it lives and what it does,
accrues to one man, literally one man.

So, that's fine as long as that person is
benevolent and, exists in an ecosystem and

a democracy that is fair, and in a country
that wants to do business with the rest

of the world and believes in, I suppose,
like the modern form of, capitalism and

globalism in the way that we've sort
of built over the last 20, 30 years.

But we also have a pr- an American
president who is erratic and an

American government that, is-- doesn't
feel like that at the moment as well.

So that has sort of been the backdrop.

And then this fa- so that resulted in
Fable, which is if you aren't as AI,

obsessed as us, is-- was Anthropic,

who built Claude, Claude's latest model,
which is extraordinarily powerful, and

within four days it was banned by the
US government for all non-US citizens.

So they basically turned
it off for everyone.

Not, not even the, the Americans
get it yet, 'cause they haven't

worked out how to even block it
for some people and not others.

Even internal employees of
Anthropic aren't allowed to use

it if they're not US citizens.

So, c- completely mad, and it will
probably end up we'll get access again

somehow and, and they'll work it out.

But it signals this enormous risk that
we are subject to by being receivers

of technology rather than creators of
technology in this space specifically.

So Europe 2031, which is like a
two-hour podcast or two-hour narrative

to listen to and 18,000 words, so
enormous sort of, sort of mini story,

basically a mini novel, novella, tells
this story of AI from about this time

last year or February, February 2025.

It follows the existing players, and
then it switches into the future, and

those existing players disappear, and
we start following these fictional

companies and fictional people as
they navigate Europe's place in this

AI-driven world over the next 10 years.

And, it paints a pretty dystopian future.

I, you know, I think some criticism of
the story has been it is a little bit

hand-wringy and, and extreme possibly,
but it's all based on real things

that have happened and could happen.

So I listened to it, and then
you listened to it, Tom, and,

There's, I think there's, there's a
few, there's a couple of things that

I took from, from this narrative.

and I would recommend anyone listening
who's interested in this stuff, and

especially Europeans, do listen to,
listen to or read it themselves.

europe2031.ai

I think is the, is the URL.

it's completely free.

But It sort of paints this story of
skepticism around the technology and

waiting for the bubble to burst and
bureaucracy for sure across the EU

that leads to some investment, but
always too late and never quite enough.

and a big bet on being able to build
something within the EU despite all the

limitations of compute and all of these,
these other, geopolitical challenges.

And they basically never quite achieve it.

and they never are able to
compete on a technological

level, and there's no plan B.

So Europe ends up caving to or being
put in a, a nightmare scenario where

they have to choose between American
and Chinese technology, and take a side

in this, technology war effectively
that, that the two new powers are,

are waging against each other.

And it doesn't end up with
a good situation for Europe.

There's no growth, and there's no control.

It becomes effectively a
vassal state of, of the US.

So yay.

Fun, fun times.

it's not actually happened yet, and
there is, a chapter at the end which

effect- effectively describes some of the
things that they could have done in this

narrative to have avoided this situation.

But it's quite-- It's sobering
and necessarily so I think.

You know, if you live…

It's more about the EU than
the Europe including the UK.

So the UK is sort of left out
of it and gets one footnote,

which is they did better than us.

But, I think we're in the same situation.

Europe has w- exactly one frontier
lab of note, Mistral, and, we're not

currently developing any others or
they're so early that, that they're

not sort of worth, worth noting.

And to catch up on that technological
front is, very difficult.

One, one of the other things or
the, the sort of narrative points

that I really took away from it
was at some point open models lost.

So I think the Chinese were developing
open models in this story, and then

they decided they didn't want to.

The models got good enough that they
wanted to keep them to themselves and

not actually give them to their enemies.

And so therefore, as soon as Chinese
incentives switched away from open

weights and, and models that others
could use, Europe and presumably

all the other countries that were
relying on open models, stopped.

Their, their progress completely
stopped because they had no way

of actually developing that, that
technolo- technology themselves.

So,

I think it was a sort of big tacit
vote in favor of open models actually.

But even an open mo- even an, open
weight model or a company that is

dedicated to that, if they're not,

benevolent or they're not within
your control, then anything can

happen, and that can be taken away.

So yeah, that was a…

Talked for a long time,
I'm, I'm aware there.

Tom, was there anything I missed or
anything that you wanna react to?

Tom: yeah.

maybe just like a, a couple
of points just to, bring…

I, I think a lot of what you
mentioned there is about politics

and economics and, geopolitics even.

that might feel a bit distant for
people to wrap their head around.

What does this mean for somebody
who is using ChatGPT to help them

write notes up more neatly or or even
using Claude Code in their day job?

what are some concrete tactical takeaways
people, might be able to imagine how

this sort of theoretical story might
apply to them in a few years' time?

Jonny: Yeah, it's a good question.

I mean, if all your, if all your really--
If your only interface with AI today

is using it personally to get ideas
for presents for your cousin or write

you a new recipe for roast chicken,
then I mean, that w- that could go.

But I think actually the models that
exist today are fine for that kind

of thing anyway, so you probably
wouldn't ever lose access to that.

you might have to get it from
a different place or pay a

different amount of money for it.

But in a professional context, I
think more and more what we're seeing

is companies are leaning into this
and building AI into the core of how

businesses work and how they function.

And some of that then means they reduce
headcount, and those people in those

roles change their skill set, so they
can't actually ultimately do the jobs

they were doing previously before.

There's no training, for, graduates
and, and younger employees and,

and graduate employment's down.

So those kind of things then have second
and third order consequences around the

ability of companies and therefore society
to function without this technology.

So if that tech-- The more, we become
dependent on this technology, the harder

it would be to, to, exist without it.

And I think a lot of the
effects would-- are not obvious.

and actually maybe only
still emergent now.

You know, if, if AI was taken away
today, probably be fine, I would guess.

but that muscle memory disappears
over time, so the longer we leave

it, the more challenging it would be.

I d-- I also-- You
know, this narrative is,

I don't think there's any situation where
AI goes away probably, I would guess.

but

For example, your information
diet being fed to you by people

that are not, don't have your best
interest at heart, is a real risk.

AI being used to influence elections
and, and, all sorts of things like that.

You need to be able to combat that.

You need to be able to, have some
way of at least understanding

what's happening to you.

And if you don't have as-- If you're
going into a gunfight with a knife,

then that's gonna be very challenging.

so yeah, it's hard to put-- I mean,
l- people should listen to this

thing, this Europe 2031, because
that paints one possible future,

and it does describe some of these
scenarios much better than I'm doing.

but in terms of tactic, I,
I'm not sure the right advice

is to not use AI, honestly.

But I think people could decide
where they get their AI from

in a more meaningful way.

So, for example, Anthropic is
considered a risk, to the US government.

you know, they, they want to
rip it out of all government

departments and things like that.

that is-- that means that Anthropic
and Claude is a potential risk to

us because it might get taken away.

You know, they might change
their strategy or have to change

their strategy to fall in line.

and generally the big US, tech
companies, you know, their incentives

are unclear and, and probably
not aligned with, with ours.

So using open-weight models, getting
our models from Europe, so for

example, Mistral, you know, you
can use their, their technologies.

probably not state-of-the-art, but
it's, it's good for a lot of things.

And actually, if you use open-weight
models, you'll find that increasingly 90%

of the work you need to do with AI is, is
fully possible with these, these things.

So, yeah, download the Mistral
app and, and use that, for

some of your AI questions.

and see if you can get on, on some of
the open models through, through some of

these, these other platforms, OpenRouter
and things like that, and, and just see

whether, whether they feel good and what
you can use them for, and don't build

a full dependence on US, US technology.

this sounds-- I feel like it sounds
like I'm wearing my tinfoil hat,

but I think that it's just pragmatic
actually, to at least know…

If you're not super reliant on it,
maybe you don't need to do this.

But as your reliance grows and you start
to think, "Actually, there's some critical

things that I'm using this for," then
reliance on any one party that you don't

feel that confident you control or has
your best interest at heart is, is a risk.

Tom: Yeah Yeah, totally.

makes a lot of sense.

I think my…

One of my takeaways from, from, listening
to Europe 2031 was the, I mean, it's, I

think, quite commonly understood at the
moment that the speed of AI acceleration

right now will only increase and only
get faster over the, the coming years.

I think there's this aspect of the
sort of existing mental models and

incentives for each of the actors in
this game will only, reinforce the

differences between these regions and
between these countries potentially.

even if the…

I- in the scenario laid out in
the, in the s- in the story,

the US has no, bad intentions.

are the natural consequences of a
structure and set of incentives that,

a certain way of approaching, their
AI development at the frontier labs.

So, yeah, super interesting.

very, very thought-provoking.

and I think the Fable five turning it
off for everyone outside of a country,

I think just highlights, oh, yeah, if,
if Claude stopped working tomorrow, and

in this future that they're describing
in the, in the story, the, the sort

of arc I'd say was around the speed of
AI development is of self-reinforcing,

but then that requires more commu- more
compute, and more compute means, more…

it's scarcer than before,

Jonny: Yeah

Tom: a, why would a country want to
share that more broadly with other

places outside of their core audience?

so yeah, fascinating,
thought-provoking, and a, a really

interesting to- topic as well.

I, I don't think I've even
really heard the word sovereignty

before the last six months.

but I think it's, it's, it's,
yeah, fascinating to think about.

There's a…

There's sort of a…

Not a flip side, but a couple
of other sort of news pieces

that have come up recently.

mentioned Cursor earlier.

Cursor, have just announced
over the last week that they're

building an office in London.

They have a headcount proposal of
200 people that they want to add.

and then I've also just been curious
recently, like Google DeepMind

is obviously based in London and,

when DeepMind was acquired by Google,
Demis Hassabis had, One of the terms

of that deal was about retaining a
London team and how important that

was to the running of DeepMind.

and, and I think now potentially, we've…

London has helped cultivate into the,
an era of technical talent, of that

contract deal with Google that has now
had this interesting knock-on effect of

AI companies considering London somewhere
that has a lot of top talent because of

Jonny: Yeah

Tom: that, Demis and DeepMind
seeded in the community.

so curious if you have any reflections
on that, Jonny, based on the Cursor news,

the sort of seeding of talent in London.

and let's not go down the path
of comparing London with Europe.

But, but I'm curious sort of gener-
generally as London as a, as a hub, if

you have any thoughts on the sort of
growing talent community, here or sort

of talent outside of the US this stuff?

Jonny: Yeah, I'm about to move
house to Hertfordshire, sunny

Hertfordshire, which is under half
an hour into Euston and King's Cross.

Well, King's Cross, Euston, Euston Road
sort of area is, is really where a lot

of the London AI stuff is happening.

You've got Google and Anthropic
and, OpenAI, I think, and, and, and

various other labs that are setting
up shop there and, and planning

on growing their London teams.

It's really exciting, and I think,
yeah, you do wonder it's a sort of

butterfly flaps its wings moment.

If Demis had said, "Yeah, I'll just
move to, to SF," where would we be now?

and again, another reminder that,
you know, the actions of so few

people, positive and negative,
can impact so many downstream

But I think I'm, I'm,
I'm bullish on London.

I think, I've been London maxing.

I went to the first London maxing event.

It was great.

and hopefully will go to many more.

And I think that Europeans are
really terrible for-- including

Londoners, are, are, are really
terrible for, underplaying things

and only spotting the, the negative.

And I think there is a huge potential
for lots of opportunities for people

in our world, but also growth,
economic growth and, excellence

and invention and all of the things
that we're good at, I think, here.

And Cambridge and Oxford are,
are a big part of it as well.

I think Cambridge is the densest AI--
second densest AI talent in the world.

SF is first.

and that's quite
impressive, you know, that,

it's such a hotbed there.

I can't remember if Demis is Cambridge.

I think he was a potentially a Cambridge
grad or post grad or something like that.

But

Tom: connections there.

and I know he goes there a lot, but I,

Jonny: yeah,

Tom: know the exact history

Jonny: Yeah.

so yeah, it's, I mean, it's good.

It's, that's the strange, like,
dualism of all of this, and I think

AI generally, and where I, I think we
try and keep a nuanced, view of it.

We aren't

I don't think this podcast
and, and our existence is about

pushing AI on people necessarily.

It's about navigating the change
that is happening around you.

And I think that's-- we're
seeing two sides of that.

We're seeing this,

new hegemony growing
around AI, which is bad.

I think bad.

we get great technical and technological
benefits from it in the short term, but it

is not good to see concentration of power.

But at the same time, all of the benefits
are fascinating, and what new drug

discovery can we do, and what new benefits
to humanity will come out of this?

Probably very many.

yeah.

And so it's an endlessly fascinating
topic, but not obviously good or bad.

It just is, and then the
rest is what we make of it

Tom: Yeah.

Yeah, so true.

So true.

those are some good, deep
thoughts, for, for today's episode.

Jonny: Yeah

Tom: are you gonna, you gonna
say anything else, Jonny?

Jonny: No, I was going to just,
ask you, so one thing that we did

put in the content channel, we
did achieve or I achieved putting

one thing in the content channel,
which is, a Twitter thread about…

I'm, I'm gonna change the subject
basically, just to warn you.

which is a Twitter thread about
how you build a business in the

AI era as a, an AI-led business.

So I suppose this is like us
talking shop or, or exploring what

near future is live as a business.

But,

I think we're seeing two things.

One is like training stuff like, you
know, people really want to upskill.

They want their teams to upskill.

They want people to come in and show
people the art of the possible and all

of that kind of stuff, which is great.

And then there's separately buying
capability directly, which is like,

"Come and build agents for us,"
or, "Come and build AI products

for us," or, "Level up our team
by being inside it on some level."

And they are outwardly maybe the
same or like look quite similar, and

you see lots of people and lots of
companies offering both, but they're

such different of activities and
areas of specialty on the inside.

And I remain not entirely convinced
that a company, definitely a person,

but a company can do both really well.

So I'm, I'm interested in your thoughts,
Tom, like, predictions maybe for us,

but also thoughts generally on a new
business model, an emergent business

model for this era or like, yeah,
where do you think we're all going?

Tom: I, I don't wanna overclaim where I
think we're, we're all going, but I, but

I do think, been working together in this
AI capacity for a year now, feel like

one of the most interesting aspects of
working in this space together has been

on this question of what, what is our
business model and how does it evolve?

that's gone from everything from purely
focusing on workshops around AI adoption

through to, helping build out build
out and staff teams through to, more

graduate level, supporting people coming
into the workforce and helping, train,

train them up or, or coach or mentor,
into this sort of new era or expanding

more, experienced product ways of
working into energetic, young talent.

and so there's that, and there's many
others in between, but I think there's

this interesting tension between when
we talk with, potential companies that

we work with or even just friends in, in
design, like, are you a consultancy or

are you-- like, what, what do you guys do?

Do you run workshops?

Do you-- are you

Jonny: Yeah

Tom: Are you-- what
is-- what are you doing?

The thread that you specifically
shared, Jonny, was, I forget who

the person was who, so maybe we
can add it to the show notes.

but the original question was, the,
an advisor to startups about, how

to determine, whether you're an
AI services business or not comes

purely down to how you charge money.

And if you're charging a day rate
or an hourly way-- rate, you are

basically an AI services business.

But if you charge based on outputs or
outcomes or, other more, nuanced form

of charging, that has a huge impact
that sort of underpins the rest of

the, the business that you build.

Jonny: Yeah

Tom: I think for us, we've been sort
of gone back and forth a bit on that,

and some clients we've worked with
have been day rates, and others have

been, on a sort of project-based fee.

don't know, really know the answer.

but I think the flexibility within
that space or sort of exploring

it and letting that expand, doing,
I think is a really interesting

space to be as a, as a business.

I think on the

Jonny: Mm-hmm.

Tom: it can look like we don't know
what we're doing 'cause we 'cause we

don't have a very clear answer to it.

But on the other, I think that is the
nature of operating in this space and

things are moving very quickly and being
adaptable and really focused on driving,

like, what is the core value that you're
delivering and helping teams unlock what

they want to be doing with that, I, I

Jonny: Mm-hmm.

Tom: is a better north
star, in my opinion.

Like who, who is the audience and what
is the value you're adding, as a focus?

And then the specifics of the
business model fall out of that.

having said all that though, I've been, on
the weekend, I was, a, friend, A friend is

a, is creative director, design director,
a type design studio called Grantly Type.

My friend

Jonny: Oh, cool

Tom: also has a, another podcast
which you should check out called,

I think it's called Makes Noises.

No.

Works But Makes Noises.

I think that's the name of it.

Anyway,

Jonny: Nice.

Tom: yellow artwork.

Jonny: Love Grellity

Tom: yes, so he, he is, is
a design director there.

And, anyway, but I chatted with him
recently and I was sort of thinking

about the, the nature of the type
foundry as a business and how there is

this sort of retail fonts side, where
I, I think most r- most people are

familiar with type foundries for like,
"Oh, they, they made this font and I

can buy that font through the website."

And actually you're not really
buying the font, you're buying

the license to use the font.

but then there is this whole other
side of many type foundries which

is around, custom type or adapting
typefaces or building completely

custom type for, brands that
w- that want that or need that.

so there is this inherent,
symbiosis between these two

elements of the business.

There is like making cool
stuff that people want and

Jonny: Yeah

Tom: because of that you build a
reputation around what that cool stuff

is and then people come to you to help
adapt and modify what that is to use

in more unique brand applications.

I don't think the space we're
in like super applicable.

Jonny: Yeah

Tom: but I think, I, I think there
is, there is some element to, having

something that is more that, is a bit
more like people have a clearer picture

of what they're getting and they, there's
a, a piece of what you're offering

that is available to a wider audience.

Jonny: Yeah

Tom: there is a more sort of
bespoke tailored experience that

is around, getting much deeper
into the nuances of how a team or

a company specifically operates.

And I think there is sort of a
shared, shared language around that.

But, but I think the thing with type
foundries that I think is kind of

interesting is, and I'm definitely not
an expert in this, I'm just sort of

observing from the outside, but it seems
like there a- there's It takes a long

time for a type foundry to build up a
base of retail fonts that are gonna, make

enough money to be a sustainable business.

So there's sort of this cat and mouse or
back and forth between those two aspects

of it to like get enough of a baseline
where you can, and enough of a reputation

to, persist on the retail font side.

Jonny: Yeah.

Tom: anyway

Jonny: yeah, I s-- it, it's a great
parallel because the way, the way we

started a year ago was with ticketed
events, and that almost definitely is

a way to fairly quickly make enough
money as a two-person organization

to run ticketed events and, and do
that as the whole thing you're doing.

But that only works if that's
what all you want to do.

and I think we quite quickly realized
that that wasn't all we wanted to do.

And so then how much do you do that
knowing that it's not the-- it's a loss

leader or it's even marketing, versus,
you know, really leaning into it.

And I think that's something that we've,
we talked about a fair amount, I suppose,

is, what do we put out into the world
for that's very accessible, cheap,

free, you know, resources, et cetera,
versus what do we keep for ourselves.

I, I'm sure that that is-- in
fact, I know that that is true

of every business everywhere.

In my previous guise as a SaaS
founder, we-- it was what, what's

in the pr- in the free tier?

What's, you know, what's available to

everyone?

Is it free resources
that aren't the product?

Is it a part of the product, or,
or is it just our wise words on our

blog that we're giving away for free?

So, but then when it comes to-- you're
right in this, in this thread, it,

it speaks about exchanging time for
money versus more value-based pricing.

And being conscious about where you're
gonna end up is important, but at the

same time, we're so new, just gotta
muddle through a bit and see what works

and see there's pros and cons to each.

You know, if you say a number and then
it turns out it was harder than you

thought, then you, you lose your shirt.

whereas if you price by time, then the
incentives aren't really as well aligned

with your, your client or whatever it is.

So, it is just muddling through, and
neither of us have run a consultancy

before, so we're learning stuff
that many people have learned over

many years, many times before.

Some of whom may listen to this and
roll their eyes, but here we go.

It's a fun journey.

Tom: Yeah, I think there's a sort of a
blessing and a curse to have not done

this before in the sense of, I think being
too rigid about what-- how to operate or

what a business model looks like right
now, I think is potentially a limiting

belief as well at this moment in time.

I do think we also keep coming
back to these three pillars

in the work that we're doing.

There's the, the working with leaders on
shaping strategy and helping leadership

teams, exec teams align on and figure
out what is relevant for them in terms

of, tooling, infrastructure, what kind of
investments to make today versus later.

There is the, the and minds work of
running workshops, running enablement

pro-programs, people across a team
to, to help level up in the work

that they're doing and weaving
AI into how they're operating.

and last of all, there is the building
piece, which I think is really part

of as builders and makers of things.

Like, we're both design-designers by
background, but I think making is a really

important part for both of us in that

Jonny: Yeah

Tom: so I think th-that is kind
of a complex picture for many

people to be like, "Oh, so I…

Do we hire you as…

Like, are you basically
freelancers that we bring

Jonny: Yeah.

Tom: build something with AI?

or do you come and run workshops?"

Anyway,

Jonny: Yeah

Tom: we've talked enough about
this, but, but I think it's a, it's

a really interesting space to keep
fluid and be open to, to ideas.

I think the other challenging aspect
of this is that companies also

don't really know what they're…

Like, many companies don't have
a clear idea of exactly what they

need other than, "We're getting
pressure to do more with AI.

Can you come and help us do that?"

a not a bad starting point, but then I
think it really leads to the interesting

unpacking of that question and what
that really means for the business.

And I think we have found that the
questions we've had with companies we've

worked with where we've been building
real tools, has led to conversations

that are much more focused on, the
alternatives to, to us doing that.

And that could be using an existing
product, which has its own licensing

fees and, and annual renewal rates.

or it could be hiring extra
people to run a certain function

because it's very, inefficient
in the way it's set up right now.

There's different avenues there.

anyway, let's move on.

I know we're tight on time.

I did, Let's-- we jump to the end?

Jonny: Well,

Tom: do you

Jonny: I did, I, I did have
a seamless transition that

I can attempt at this point

Tom: Perfect.

Jonny: f-from what, yeah, from what,
what do you give away for free,

transitioning to something that
you're working on, Tom, which we've

talked about for ages and finally…

Well, a few things have happened
this week actually, in our world.

One of which is we put our website live,
and it's still a work in progress, but

it's better than it was, and it's actually
has words on it and, things like that.

So, we're using modern technology
like words and pictures.

there's still, there's still a way to go.

we haven't got any real case studies
or anything like that up on there yet,

but, it's, it's progress, and this
podcast can now be found on that website.

Maybe you found this podcast episode on
our website, so you know this already.

But Tom, what else are you building?

Tom: I, thanks for that
update, by the way.

And I wanted to, hearing you say
that makes me think we should have

a, a convert to caveman button
where you can convert all the copy

on a website into caveman speak,

Jonny: Ship it.

Let's … I'll get an agent on it.

Tom: okay, so what I've been,
what have I been working on?

So I've been, working on an open
source design frame that helps people

who are building these one-off micro
design tools to things as part of their

work, and basically consolidate them,
pull them all together into a single

coherent, product-like experience.

So imagine Figma, but, a really stripped
down minimalist version of Figma

that has a consistent with consistent
controls and interactions, and then

a selector at the top where you can
choose between the different tools

you've built, and then this panel on
the right, which is the, the tool.

And I've been building a bunch of
tools this way to create, visual

assets for, things we've been doing
as Near Future, dithered gradients

and, information design treatments
for circles and shapes and things.

but I think it's just a, a really
fun way of, of working too.

So we had a conversation earlier this
week about it, and that kind of lit

the fire under me to spend, a bunch
of time actually, moving this forward.

And yeah, it's exciting.

I feel, both completely out of my
depth at r- working, running an open

source project or sort of preparing
to launch an open source project.

and I'm sure I've made many fails, and
I'm gonna be criticized by any software

developer on anything that I'm doing.

But at the same time, I think,
it's been a really good experience

just to go through the process
of, oh, I want to package this up.

What's the most useful way I can do that?

Jonny: Yeah

Tom: Claude Code has been helping
me, helping coach me through the

process and talking about MIT
licenses and, kind of stuff to put

in a README file and, all of this.

but I think the, the core of this
idea, just told you about the what,

but the, the why for it, came from, a
workshop that we ran or two months ago,

Jonny: Hmm

Tom: a, a big global fintech company
where we helped them, helped 60

designers across the team build 60
individual design tools in three hours.

And it was this workshop format that
really resonated for what they were doing,

and we saw some incredible results off
the back of it, these, amazing tools

that were, were built and shipped.

the idea of what c- what could
companies do, what could teams do,

or even what could individuals do
if they had this approach which

encouraged these super custom tools
that helped them design things that

were unique to how they operated.

And so I f- I still find that
idea re- really inspiring.

And so the whole architecture of this
project is based on the idea that Each

of these tools can operate on its own.

So you can just drop, it's basically
a-an HTML file with JavaScript and CSS.

So I could literally send you, over
Slack you an HTML file, and then you

could download it and double-click on
it, and it will work as a design tool.

Jonny: Yeah

Tom: but you could also add it into
this, this frame, and then it sort

of collects them all together and
allows you to switch between them.

So the key is that it's super
local, su-super portable, super

lightweight, frictionless.

Jonny: Right

Tom: start…

You don't need to do NPM install.

You don't need to have a server running.

Jonny: Yeah

Tom: do that, and there might be
some benefits if you did do that.

but it should be as simple as downloading
a zip of this GitHub repo, on it, and

then just running an index file, and
then everything is available to you.

And you can just point Claude at it and
have Claude build your own design tools.

yeah.

It's been really fun to put together
and, I'm sure once it's released

more openly it will be humbling to
see what people complain about or,

Jonny: Yeah

Tom: or, or whatever.

But, yeah, super cool

Jonny: I think loads of
complaints would be a good sign.

It means people are excited about it.

Tom: that is true.

Very

Jonny: The, the thing--
I haven't tried it yet.

I'm very excited to.

The thing I'm excited about
especially, and this was sort of

a build on what we did in that
workshop, is the how lo-fi it is.

I think, you know, if you ask
Claude to build a website, it's

gonna start with a package.json

with like React and Vite and all
of this other stuff that it sort of

builds this, this dependency tree,
and Tailwind and blah, blah, blah.

And actually, you don't
need any of that stuff.

Like, the fact that this is an HTML
file means someone on a computer that

isn't allowed to run a single command
in terminal, that isn't allowed to open

a single website that it doesn't trust,
should-- that person should be able to

get some value out of this HTML file.

And it's so simple, like HTML
has been around for decades.

but it's still what
everything compiles to.

When you build your mad React
thing, it ends up as HTML in

one great big file anyway.

So, it's just cool to advocate
for something that is, so simple

and a simple piece of technology,
but that can actually do all of

the things that we want it to do.

We've got-- We could put one
example of something that we built.

It's may-maybe an earlier version we
could put it in the show notes, which is

the, our, like, dithered shape generator
thing, which is, live on the internet.

So we could, we could post
that somewhere in the meantime.

But looking forward to this going live

Tom: Yeah.

Yeah.

So true.

And, last thing, it was named, well,
current working title is Kazam,

named after the Kazam machine by
Charles and Ray Eames that was the

invention required to invent the
thing they actually wanted to invent.

And I really like the idea of,
helping give tool builders the

tool to unlock more tool building.

So

Jonny: Yeah

Tom: to do that or just this is,
something that is encouraging of them

to build their own unique tools, they
sort of download it, rip it apart,

try something completely different,

Jonny: Yeah

Tom: all fine.

The, the goal is raising the craft and
having craft win and explore all these

new a- new angles that are now possible.

So

Jonny: Blynt

Tom: with that, let's wrap up.

Jonny: Let's wrap.

Thank you, Tom.

Another episode in the
can, and, see you next week

Tom: Yeah.

See you next week.

Thank

Jonny: Bye.