Why Governments Must Grab the Steering Wheel of Artificial Intelligence — Now, Not Later
By a concerned news reporter watching the warning lights flash
There is a feeling in the air right now. A quiet tension. Like the moment before a storm breaks.
Artificial intelligence is everywhere — writing, designing, diagnosing, predicting, deciding. Big tech companies call it progress. Investors call it gold. Politicians call it “innovation.” But many ordinary people are starting to ask a harder question:
Who is actually in control?
A recent letters piece published in The Guardian put that question front and centre. The message was blunt, uncomfortable, and urgent: if governments do not take control of AI now, they may never get the chance later.
The warning did not come from a sci-fi movie or a tech conspiracy forum. It came from citizens watching history repeat itself.
The ghost of 2008 is still here
One letter, written by Anja Cradden, cuts straight through the hype. She pushes back against the idea that when the so-called “AI bubble” bursts, society will magically regain control.
Her fear is simple — and painfully familiar.
When the AI bubble bursts, she argues, it will not be everyday people who get rescued. It will be the same wealthy tech executives, investors, and economic power players who helped inflate the bubble in the first place.
We have seen this movie before.
In 2008, banks collapsed after years of reckless behaviour. What happened next? Governments rushed in with public money. Jobs were lost, wages froze, public services were cut — but the richest institutions survived. Some even became richer.
Cradden’s point is clear: AI could follow the same script unless we change the ending now.
The real danger is waiting too long
One of the biggest mistakes governments make, again and again, is waiting for disaster before acting.
AI today is mostly regulated through soft promises, voluntary guidelines, and “innovation-friendly” rules. Tech companies are largely trusted to police themselves. Regulators are underfunded. Laws move slowly while technology moves fast.
Supporters of this approach say strict rules could slow progress. Critics say this is exactly how crises are born.
The Guardian letter argues that AI is not just another tool or app. It is a system that affects jobs, elections, privacy, energy use, culture, and even how truth itself is produced.
Once AI systems become deeply embedded into daily life, rolling them back becomes almost impossible. That is why waiting for a crash is not caution — it is surrender.
A bold idea: governments as shareholders
One of the most striking proposals in the letter is also one of the most controversial.
Cradden suggests that if major AI tech companies begin to crash, governments should be ready with an alternative plan — not bailouts, but ownership.
Her idea is this:
If a tech giant produces something genuinely useful to society, governments could coordinate internationally to buy majority shares at low prices during a crash. These shares would include full voting rights.
That means governments would not just rescue companies — they would control them.
As majority shareholders, governments could:
- Break up massive monopolies into national companies
- Force firms to pay taxes in the countries where they actually operate
- Enforce local copyright, content, and labour laws
- Invest in parts of the business that serve public needs, not just profit
- Later sell the shares back, potentially at a profit for taxpayers
It is a radical idea — but not an impossible one. Governments already step in during crises. The difference here is who benefits in the long run.
Another option no one wants to say out loud
Then comes the most uncomfortable suggestion of all.
What if we don’t rescue them at all?
What if governments decide that some AI systems simply cost too much — not in money, but in power, water, energy, and social damage?
AI data centres consume massive amounts of electricity and water. They strain local infrastructure. In some regions, residents already complain about rising energy prices and water shortages linked to these facilities.
Cradden raises a question many leaders avoid:
Should we shut some of this down?
Should governments refuse to build new data centres?
Should power and water be prioritised for people, not machines?
It sounds extreme. But in a world facing climate stress, resource shortages, and inequality, it may soon become a practical question, not a moral one.
“There is no alternative” — the most dangerous sentence
One phrase echoes throughout the letter, heavy with warning:
“There is no alternative.”
This sentence has been used before — during financial crises, austerity policies, and emergency laws. It is often spoken behind closed doors by powerful people, then presented to the public as unavoidable.
The letter argues that the only way to fight this is preparation.
If citizens and governments start discussing alternatives now — public ownership, stricter regulation, controlled shutdowns — then when the crisis hits, leaders cannot claim ignorance.
Ideas are power. Silence is surrender.
Why this debate matters right now
Since late 2025, concerns like these have only grown louder.
Around the world, AI systems have already caused real-world problems:
- Chatbots giving false legal advice
- AI tools hallucinating facts in court filings
- Automated systems reinforcing bias in hiring and policing
- Emotional manipulation through AI companions and gambling bots
Each failure chips away at trust. Each one proves the same point: controls added after deployment are always weaker than controls built in from the start.
Governments are falling behind — fast
In the UK and many other democracies, AI regulation still relies on existing sector regulators — bodies that were never designed to oversee systems this powerful or complex.
Critics say the government’s “pro-innovation” stance sounds good in press releases but leaves regulators toothless. Meanwhile, tech companies race ahead, setting rules by default.
Other countries are not waiting.
China, for example, has already introduced rules limiting emotionally manipulative AI and restricting harmful chatbot behaviour around self-harm and gambling. These rules are strict, fast, and enforced.
The irony is hard to miss: authoritarian states are acting faster than democracies to control AI.
This is not anti-technology — it is pro-democracy
Supporters of stronger AI control stress one thing again and again: this is not about stopping technology.
It is about who decides how technology shapes society.
Right now, a small group of corporations sets the direction. They decide what gets built, who gets replaced, what data is used, and what risks are acceptable.
The Guardian letter argues that this power should belong to the public — through democratic governments — not private boardrooms.
AI should serve people, not manage them.
The window is still open — but not for long
History shows that moments like this do not last forever.
Once systems become too big, too profitable, and too essential, they become untouchable. Regulation turns symbolic. Accountability fades. Crises become inevitable.
The warning is clear: act before the crash, not after.
Governments still have leverage. Public opinion is shifting. The failures are visible. The costs are rising.
What happens next depends on whether leaders choose courage over comfort.
Because if they don’t, the next rescue plan may already be written — and ordinary people may once again be asked to pay for a disaster they did not create.
And this time, the machines will not be the ones losing control.
Login with Google