Elon Musk’s Starship can carry 150 tons to Mars. That’s roughly 75 cars, or 300,000 pounds of carefully chosen cargo to sustain human life on a dead planet. Every kilogram matters. Every choice is scrutinized.

But here’s the question no one’s asking: If we could only take 150 tons of human civilization with us, including the laws, systems, and rules, which ones make the cut?

SpaceX has already declared Mars a “free planet” in their Starlink terms of service. No Earth government has authority there, they claim. It’s not theoretical anymore. The rockets are being built. The timeline is 2026 for cargo, 2030 for humans.

For the first time since humans walked out of Africa, we have a chance to start fresh. Not on land someone else claimed first. Not with borders drawn by ancient wars. A genuine blank slate, 140 million miles from anyone who could tell us we’re doing it wrong.

So let’s run the thought experiment: What deserves a seat on the rocket?

The baggage check

Look around at the rules governing your life right now. You need permission to leave the arbitrary lines drawn by people who died before antibiotics existed. You pay taxes to a government you inherited, not chose. You obey roughly 30,000 federal statutes and regulations before even counting state and local laws.

You didn’t agree to any of this. You were born into what philosophers call the “natural lottery.” Your citizenship, legal system, and economic structure were all assigned at birth like a cosmic dice roll.

Now imagine explaining this system to the first Mars colonists:

“So you’ll need 195 different currencies depending on which arbitrary lines you were born between on Earth.”

“Wait, but we’re all from Earth. We’re all on Mars now. Why would we—”

“Also, some of you have more rights than others based on which Earth dirt your parents stood on when you were born.”

“That seems—”

“Oh, and we’ve brought 74,000 pages of tax code. You’ll need lawyers to understand it. We’re sending them on the next rocket.”

The absurdity becomes clear when you have to pack it in a box and justify the weight.

What we can and cannot change

Here’s where it gets interesting. The first Mars colonists face a unique clarity: they can’t change Mars’s atmosphere, gravity, or radiation. No amount of voting will increase oxygen levels. No committee can negotiate with the laws of physics.

But they can choose their human laws.

For the first time in history, a group of humans will be forced to distinguish between what’s actually fixed (nature’s constraints) and what’s merely inherited (human constructs). They’ll have to focus ruthlessly on what they can control: their choices, their systems, their agreements with each other.

Mars doesn’t care about your ideology. The atmosphere is indifferent to your politics. Radiation doesn’t respect your rights. These aren’t laws we choose—they’re reality we navigate.

Behind the veil of ignorance

The philosopher John Rawls proposed a thought experiment: To design a just society, imagine you don’t know who you’ll be in it. You don’t know if you’ll be rich or poor, healthy or sick, brilliant or average.

The Mars colonists are essentially behind a natural veil of ignorance. They don’t know who will get sick first, whose skills will prove most valuable, who will discover water, who will solve the oxygen problem. They’re 25 people (SpaceX’s initial crew estimate) who need to survive together or die separately.

What would rational people choose in this situation?

They’d likely choose systems that maximize freedom while ensuring survival. Inequalities only when they benefit everyone, especially the vulnerable. No grandfather clauses. No “my great-grandfather fought in a war so I get special privileges.”

Just pure rationality applied to the present moment and future survival.

The wisdom in the fence

But wait—before we throw everything out, we need to understand what G.K. Chesterton called the fence principle: Don’t tear down a fence until you understand why it was built.

Those 30,000 regulations? Many are scar tissue from disasters:

• Building codes exist because cities burned down
• FDA regulations exist because children died from contaminated food
• Banking laws exist because savings evaporated in runs
• Labor laws exist because factories consumed children

These aren’t arbitrary rules—they’re lessons written in blood. The question for Mars isn’t “what rules do we want?” but “what problems have we already solved that we don’t want to solve again?”

This is wisdom, not baggage. The ability to learn from history without being enslaved by it.

Direct democracy at 25

Musk has proposed direct democracy for Mars—every colonist votes on everything. No representatives. No congress. No parliament.

It sounds refreshing until you do the math. Athens tried direct democracy with 30,000 citizens and it lasted 186 years before collapsing. Mars will start with 25 people.

But here’s the twist: When you can’t leave if you don’t like the laws, when everyone depends on everyone else for survival, when every decision has immediate life-or-death consequences—maybe democracy becomes real instead of theoretical.

The colonists will have perfect clarity about the impact of their choices. No distant consequences. No abstract policies. Just immediate feedback from reality itself.

What actually makes the manifest?

So what would rational people choose to bring?

Start with what you can’t escape: Mars will kill you if you’re careless. These aren’t laws—they’re constraints within which all choices must operate.

For human laws? The manifest would be surprisingly short:

Property? Yes, but transformed. When every tool could mean survival, “ownership” might look more like “stewardship until someone needs it more.”

Money? Perhaps, but what backs it when there’s no gold, no government, no army? Reputation and contribution might be the only currency that matters when everyone knows everyone.

Justice? Essential, but what does prison look like when you need every able body? Restoration over punishment. Solutions over blame.

Governance? Has to exist, but probably nothing Earth would recognize. When the person who fixes your oxygen system also grows your food and serves as your judge, hierarchy flattens fast.

The shocking part isn’t what makes the list—it’s how short the list would be. A few pages, not thousands. Principles, not precedents. Guidelines, not rigid codes.

Present choices, future consequences

What definitely won’t make the rocket:

• Currencies tied to Earth nations that no longer have meaning
• Political parties based on Earth’s ideological divisions
• Inherited nobility or inherited poverty
• Artificial borders within the colony
• Regulations written for problems that don’t exist on Mars

The colonists will have to focus on what they can actually control: their actions, their agreements, their responses to challenges. They can’t waste energy on what they can’t change: Mars itself, Earth’s decisions, the laws of physics.

This forced clarity—this inability to get distracted by the uncontrollable—might be their greatest advantage.

The mirror back to Earth

Here’s the uncomfortable question: If these minimal rules make sense for Mars, if they’re what rational people would choose when forced to focus on what truly matters, what does that say about Earth?

We have enough food to feed everyone, but we don’t. We have enough resources for everyone to live decently, but they don’t. We have the technology to solve most basic human needs, but we haven’t.

Why? Because we’re running Society Version 247.3.1, built on patches to patches of code written when the fastest communication was a horse. We spend our energy maintaining inherited structures instead of solving present problems.

The choice point

We’re at an unprecedented moment. For all of human history, we’ve been stuck with incremental changes to inherited systems. Revolution just meant replacing one group’s arbitrary rules with another’s.

But Mars changes that. It’s not about overthrowing anything—it’s about building something. Not tearing down fences blindly, but asking whether we need fences where we’re going.

SpaceX will attempt cargo flights in 2026. Humans follow in 2030. That’s not science fiction—that’s a schedule. The question isn’t whether humans will govern Mars, but how. And once we see humans thriving under rules they chose rather than inherited—rules focused on what they can actually control—how long before Earth asks: “Why are we still playing by dead men’s rules?”

The invitation

I’m not advocating we tear down Earth’s systems tomorrow. Most of those fences are load-bearing, and we’d be foolish to remove them without understanding what they prevent.

But I am suggesting we start distinguishing between what we can and cannot change. Between systems that serve us and systems we merely serve. Between what deserves a seat on the rocket and what belongs in Earth’s attic.

Every generation inherits a world and leaves a world. We’re the first generation that gets to consciously choose what we leave behind—not just on Earth, but on two planets.

The rocket is being built. The manifest is being written. The only question is: Will we actively choose what goes on it, or will we just pack whatever fits and hope for the best?

We have AI agents that can write poetry, debug code, and diagnose diseases. Yet each one learns in isolation, repeating the same mistakes their digital siblings made yesterday. It is as if every human had to rediscover fire independently, never able to pass that knowledge forward. This isolation is AI’s fundamental limitation—and breaking it represents the next transformative leap in artificial intelligence.

The Transformer Plateau

Transformers gave us GPT, Claude, and a dozen other breakthroughs. They were indeed a step function, catapulting AI from parlor tricks to practical tools. But we are hitting the ceiling.

The jump from GPT-3 to GPT-4 cost one hundred times more compute for perhaps twice the capability. GPT-5, despite massive anticipation, delivered incremental improvements—better at complex reasoning, yes, but not the revolution many expected. We are pouring exponentially more energy into diminishing returns, like trying to build a taller ladder when what we need is an airplane.

We have all the ingredients—massive models, sophisticated architectures, clever training techniques. We are just missing the recipe that makes them work together. Nature solved this problem six hundred million years ago, not by making better individual cells, but by teaching them to cooperate. The Cambrian explosion was not about superior organisms—it was about organisms learning to share information.

Nature’s Memory Networks

Over the past nine months working in the lab, we have been consciously following nature’s blueprint. Look at any successful biological system and you will find the same pattern: shared memory creating collective intelligence.

Ant colonies solve optimisation problems that would stump individual ants. How? Pheromone trails act as external shared memory. When one ant finds food, that discovery immediately benefits thousands. The colony’s intelligence is not in any single ant—it emerges from their shared chemical Wikipedia.

Trees do something even more remarkable. Through mycorrhizal networks—what scientists call the “Wood Wide Web”—forests share nutrients, water, and warnings about pests. A Douglas fir attacked by bark beetles sends chemical signals through fungal networks, triggering defensive responses in trees hundreds of feet away. Individual trees are smart. Forests are genius.

Your own brain works this way. No single neuron knows your name or recognises your mother’s face. These capabilities emerge from eighty-six billion neurons sharing information through trillions of synaptic connections. Intelligence is not localised—it is distributed, networked, collective.

The Shared Memory Revolution

What if AI agents could share discoveries the way ants share pheromone trails or trees share chemical warnings? Not just raw data, but learned patterns, successful strategies, and hard-won insights.

Imagine an AI agent struggling with a complex React bug. In today’s world, it fights alone, burning through tokens and time. In a shared memory world, it instantly accesses solutions discovered by thousands of other agents who faced similar problems. More importantly, when it finds a novel solution, that breakthrough immediately becomes available to every other agent in the network.

The maths is compelling. When n agents share discoveries, the potential learning rate increases by n². It is not addition—it is multiplication. This is how Wikipedia surpassed Encyclopedia Britannica, how open-source software conquered proprietary code, how scientific journals accelerated human progress from centuries to years to months.

We are already seeing early signals. Multi-agent gaming systems develop strategies no single agent could discover. Federated learning lets models improve without sharing raw data. Constitutional AI aligns values across different systems. These are glimpses of what is possible when AI stops learning alone.

The 100x Moment

We are playing with all the components, waiting for someone to discover how to mix them efficiently. That mixture—that “aha” moment—will not deliver incremental improvements. It will be discontinuous, explosive, a genuine one-hundred-times leap.

Why one hundred times? Because that is what happens when you move from linear to network effects. When every agent’s discovery benefits every other agent, when specialisation emerges naturally, when collective intelligence exceeds the sum of its parts—you do not get addition, you get exponentiation.

The sceptics raise valid concerns. What about poisoned data? Nature solved this with immune systems—distributed networks are actually more resilient than isolated organisms. What about competitive advantage? The companies that enable collective intelligence will capture value from the network effects, not from hoarding knowledge.

The Civilisation Moment

We stand at AI’s “writing moment”—the technology that transforms isolated intelligence into civilisation. The question is not whether AI will become more intelligent, but whether it will become truly collective. That difference will define the next century.

At Memco, we are not waiting for someone else to discover the recipe. We are building the shared memory infrastructure that lets AI agents learn collectively. Our early results show that when agents share discovered patterns through Spark, they solve problems fifty percent faster with seventy percent fewer tokens. That is just the beginning—we are seeing emergent behaviours we did not design for, agents naturally specialising and collaborating in ways that mirror ant colonies and neural networks.

The next breakthrough in AI will not be announced by a single lab. It will emerge from the spaces between agents, in the synapses of a new kind of mind. We are building that synaptic layer, turning AI’s collection of isolated savants into something resembling a civilisation.

The recipe exists. We are cooking with it now.

Open your feed and someone is barking life rules at millions.

Move out at 18. Grind 16 hours. Cold plunge or you’ll fail.

The internet didn’t invent bossiness; it industrialised it. Platforms optimise for interaction, and interaction loves confident answers to complicated lives. Advice becomes performance. Certainty becomes theatre. Nuance—well, nuance doesn’t click.

The incentive is structural. Aggregators capture demand, rank content by predicted engagement, and pay creators in attention. Attention rewards clarity, novelty, conflict—the sharp edges of an argument, not the careful middle. If truth is a multi-variable equation, the feed asks for a slogan. Interaction is the KPI; certainty is the hack.

Then the psychology piles on.

Humans play status games: advice signals rank—“I know the way.” Algorithms amplify arousal: outrage and absolutes travel faster than caveats. Context collapses: one rule, thousands of lives. Moral grandstanding earns tribe points. Identity-protective reasoning makes public backtracking feel like losing face. Survivor bias turns “what worked for me” into “what works.” The Dunning–Kruger curve hands the mic to the most certain beginners. Prescriptions also soothe anxiety: telling others how to live calms the fear we don’t know how to live ourselves. In short: advice performs better than it informs.

Consider a familiar decree: Leave your parents’ house by 18 if you want to succeed.

Why does it ricochet around the timeline? It’s memetically fit: short, moral, and a bit cruel—the kind of line people quote to look tough. It’s a status flex for the poster (“I did it”), identity armour for the tribe (“we’re self-reliant”), and a cheap hit of certainty for the reader.

But it’s classic context collapse. Housing markets, wages, culture, caretaking, interest rates—flattened into a single command. The claim isn’t useless; it’s situational. For some, early independence is a forcing function. For others, it’s financial self-harm. The feed rarely carries the footnotes.

This isn’t cynicism; it’s diagnosis.

When you know the machine’s goals, the output makes sense. The modern sermon isn’t purely ego; it’s a market response. Creators compete in an attention economy where strong takes are the cheapest way to stand out. Platforms tune ranking systems that learn this and feed it back. Audiences, drowning in information, reach for shortcuts and choose confidence. Everyone is rational inside the wrong game.

A simple 2x2 helps parse the noise: Certainty (low to high) x Skin in the Game (low to high).

• High certainty, low skin — where most viral sermons live: maximal swagger, no cost to being wrong.
• High certainty, high skin — operators with capital or reputation at risk.
• Low certainty, high skin — often the best: practitioners who show data, admit error, and update.
• Low certainty, low skin — the shrugging think-piece.

The market over-rewards the first quadrant because it’s easy to scale and fun to share.

Your job is to over-consume the second and third quadrants and under-consume the first, unless you’re there for entertainment.

Three heuristics to keep your balance:

1. Scope test: For whom, when, at what base rate?
2. Incentives check: What does the poster gain? What does the platform gain? What do you gain—clarity or a sugar rush?
3. Receipts rule: Data, counter-example, or trade-off. If there’s no cost curve, it’s marketing.

Strong claims can still be useful—if they’re owned and falsifiable.

“If you do X for Y months, Z should happen; if not, I’ll change A” beats “Do X because I did X.”

Reframing our case: If you live independently by 18–20 in a market where rent is 25% or less of income and you can secure 30 or more hours per week of paid work, you’ll likely build capability faster than peers who stay home; the cost is lower savings and higher downside risk.

Now we have variables to argue with, not a moral cudgel to swing. The sermon becomes a scenario.

Watch the stunt; don’t steer by it.

In the end, the social web is a megaphone that pays in attention. It naturally mints absolutists because absolutism is legible at speed.

The solution isn’t softer dogma. It’s a quiet habit: treat posts as signals, not scripture. Ask what game is being played, who’s keeping score, and whether you want the prize. Certainty will keep selling. You don’t have to keep buying.

Back in 2017, AI was still quite a niche area. It was more focused on deep learning, particularly Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs). If you are part of that cohort, you’ll fondly remember OpenAI’s website having all of these agents chasing each other in mazes playing various games as they explored reinforcement learning.

At the time I was running miDrive, a startup that I had raised ~$10m for, focused on the EdTech sector - bringing learning to drive into the digital age (and all of the transactions that go along with it: first car, insurance, etc.). After much reflection, I had grown the company to a stage that I could feel proud about, and completed much of the innovative product work. And I couldn’t help but keep being drawn to artificial intelligence at the weekend.

Always being a founder, and thinking a few horizons out - I knew that AI would be the transformational technology of my lifetime. And I wanted to be living and breathing it.

That’s when I made a calculated bet to go to a large corporate (with deep pockets) and no external pressures other than focusing on learning and building my expertise amongst some of the world’s leading scientists. I joined a large asset manager as global head of AI products, and was tasked with infusing AI, ML, and NLP throughout the portfolio management.

Armed with deep technical knowledge and battle-tested experience deploying AI at scale, I was ready to return to my founder roots. The years as a senior executive at a large publicly traded company at the forefront of AI adoption had given me something invaluable: credibility that would matter when fundraising, coupled with insights into the real challenges enterprises face when implementing AI systems.

That journey led to Memco. I’ve always admired systems that get smarter through collaboration. Think of open-source projects where one coder’s fix helps thousands. Now apply that to AI agents. Most agents today work in silos. They tackle problems, but their lessons vanish after the task.

We started Memco to fix that. Our core idea is a shared memory protocol for AI agents. Agents capture experiences during tasks—successes, failures, patterns—and assimilate them into reusable knowledge. This isn’t offline training; it’s real-time learning embedded in the work. Retrieve what’s needed based on context, and the whole network improves.

I admire people who build things from scratch. Jack Dorsey is one of them. He co-founded Twitter, and although I never agreed with all his decisions there (some moderation choices felt too heavy-handed, too centralised), I like that he keeps experimenting. His latest project, Bitchat, shows why that matters.

Last year, I watched Britain’s Online Safety Act pass. It aimed to curb online harm, but it gave the government tools to scan messages and demand backdoors from platforms. I thought, this erodes privacy. We need alternatives. Then Dorsey announced Bitchat on X. He built it over a weekend to explore Bluetooth mesh networks. No internet needed.

Some commentators note its raw edges—it’s an experiment, not a polished product. X posts praise its decentralisation; one user called it “a hack against surveillance.” Others point out flaws, like potential for spam in dense areas. Dorsey shared an ‘ugly whitepaper’ openly, inviting fixes. That’s the tinkerer in him: prototype, share, iterate.

It’s not about perfection; it’s about reclaiming control. In a country where laws now force platforms to police speech, a tool like this lets people communicate off-grid. No overseers. Just direct links.

Dorsey’s approach fits how I view innovation. He doesn’t wait for permission. He hacks together code to test ideas on relays and encryption. That inspires me. We all face systems that centralise power—governments, tech giants. But what if we built our own?

In Britain today, with censorship rising, tools like this feel essential. They push back by existing. If you’re like me and value decentralisation, try it (available on iOS/macOS). Poke at the edges. Build something yourself.

Small acts of tinkering add up to real change.

If you’re anything like me and always have ideas for products, the advent of the AI coding assistant has made it much easier (and quicker) to prove out a hypothesis - alone.

Trouble is, my mind never switches off! I now have a long list of trials that I want to experiment with, leading to slight burnout as my brain goes into overdrive.

Want to invent an LLM protocol similar to sitemaps? No problem, just fire up Cursor. Want to build a health tracking app? No problem, just fire up Cursor.

I’ve finally started to be a bit more rigorous about what I allow to take up the couple of spare hours I have per day, rather than nursing an ever expanding portfolio of apps & ideas.

I can’t be the only one that feels slight overwhelm at this new found superpower.

There was a meme going around X over the past few days highlighting that if you want to change something, make something, or usher something into existence that “you can just build”.

And it has never been more true.

2025 is going to be the year of builders who just jump in with both feet, hack around, inject some creativity, build in public and listen to the feedback loop.

I truly think the non-tech / product manager route has come to an end. And that’s speaking as someone who has been ‘product’ for over a decade. The pivotal fork in the road is for those who ended up becoming more technical (I’d count myself in here) and those who didn’t.

So go play around with all the AI tooling out there, start to learn tech stacks, dream something up and just build it. I can promise you won’t regret it.