-
Middlesbrough
Aisha designs flood models at the Tees Valley Compute Centre.
Her grandmother worked at the steelworks before it closed. Her mother cleaned offices in London for twenty years. Aisha's models have saved an estimated £3 billion in avoided damage since the network went live.
She cycles past the old estates, solar on every roof, the hum of cooling from the datacentre. A generation ago this place was dying. Now it is essential.
-
Edinburgh
A budget is tested against twelve futures before anyone votes on it.
"What happens if energy prices spike 40%?" The model surfaces three interventions that soften the worst outcomes. Two weren't obvious. One contradicts the Treasury's first instinct.
It is wrong sometimes. But it is wrong in ways you can examine, probe and improve, and the decision stays with people.
-
Newport
Dafydd runs a shift at a chip fab that didn't exist in 2025.
Sensors, power electronics, automotive parts: mature nodes Britain makes now. Eight thousand people across three sites. He started as an apprentice in 2028. His kids are doing engineering degrees.
-
Westminster
Nine days' warning of a flood, instead of two.
The coupled ocean–atmosphere model puts severe flooding in the North West at 34% over nine days. Not certainty, but enough to pre-position resources and prime the response. Not predictions: probabilities, scenarios, options.
-
Belfast
A drought in Spain, traced to children's dinners in Sheffield.
Caoilfhionn couples climate to farming to food prices to household budgets to health. Every assumption is arguable and the validation is never finished. But it makes the invisible visible, and what you can see you can act on.
-
Manchester
A games studio builds its world on public tools.
Capability that once lived only in Californian megacorps is a public utility now. Insurance, logistics, planning, film: a dozen industries spun out, and the investment repaid several times in licensing alone.
That isn't the point. The point is that Britain builds things again.
We don't know if this will work. The scale is terrifying. The politics could collapse at any moment.
But we decided that managed decline wasn't acceptable. If you're reading this in 2045 and the datacentres are humming and people have work that matters, we did something right. If not, at least we planted something.
The shade isn't for us.
How we get there
Sovereign simulation capability as the anchor tenant for an industrial rebuild in every region. £100 billion over twenty years: £5 billion a year, less than HS2, less than the 2008 bailouts.
What we build
Ten to fifteen regional compute centres, two or three mature-node chip fabs, a national research backbone, university programmes across dozens of institutions.
Who works on it
65,000 to 100,000 direct jobs, 100,000 to 200,000 with the supply chain. Teesside to Belfast to Newport. London does coordination only.
Who pays
Those who can afford it. A 1.5% wealth tax on billionaire assets alone covers it. Windfall levies on privatised utilities could add £2 to 5 billion a year. The "millionaire exodus" was manufactured: 75% of UK millionaires say they would pay more tax.
What comes back
Licensing, allied contracts, less waste on failed infrastructure: a conservative £1 to 3 billion a year at maturity. Investment, not spending.
Who takes part
Anyone. Submit an idea and the modellers will have a go; groups compete for the best answer with support; children gather real data at home and school.
It's a choice.
We could keep arguing about boats and managing decline. Or we could decide our grandchildren deserve to inherit something other than our arguments. Let's plant the bloody trees.