It should be fairly clear by now that the rate of development for
artificial intelligence isn’t slowing down… which to my mind means one thing (well several, but only one is the point of this article).
Global
governments can’t afford to sit on the sidelines and just watch it happen anymore.
Every month brings new breakthroughs that shift what’s possible, from generative tools that rewrite the rules of creativity, to
agentic systems capable of making decisions and acting on them.
For decision-makers in both the private and public sectors, the pace of change is creating a problem… by the time new laws/policies around AI are drafted, the technology has already massively moved on.
The question though, is what this means for decision-makers… in both government and in business.
Who’ll be held accountable for AI’s impact on the public?
It’s tempting to just leave the future of AI to the private sector.
After all, corporations have the investment power, the technical expertise and the hunger to innovate. But when that innovation is being driven by short-term quarterly results, shareholder profits are always going to outrun public accountability. That means the risks, bias, exclusion, lack of transparency… even destabilising job markets, will get pushed back onto us (society as a whole), in pursuit of quarterly gains.
Modern governments, on the other hand, (whether you like it or not) are all built around the concept of incremental progress. Contemporary economies are too complicated and interconnected to realistically allow for anything else. Policy moves slowly and the machinery of state is rarely designed for rapid adaptation.
Yet the role of governments shouldn’t be to just retroactively “
police” AI… chasing after problems once they appear. Their responsibility should be to help shape the trajectory of AI so that it benefits economies, societies and the public over both the short
and the long term.
This matters because AI is unlike any other technology that has come before it.
Unlike past tech waves, it’s capable of adapting, scaling and embedding itself into decisions that affect millions (billions)… in real time. Generative AI is already shaping sector wide hiring practices, healthcare and even our public services, with Agentic AI already being rolled out.
But left unchecked, it will evolve faster than the safeguards we all rely on to ensure fairness, trust and equity in our lives.
At FormusPro, I’ve seen firsthand how organisations can use AI responsibly to build transparency, improve governance and align innovation with real human outcomes.
And it’s those principles that need to be applied at both national and global scales.
Governments are going to need to work alongside the private sector to set those guardrails, not create roadblocks, ensuring that AI continues to develop in a way that creates opportunity without sacrificing trust (or progress).
The challenge is urgent, and the choices made in the next few years are going determine whether AI is remembered as a technology that served society… or one that society struggled to control and adapt to.
The Limits Of Private Sector Leadership
Up till now, it’s been the private sector that’s really been the engine of AI innovation.
From OpenAI, Microsoft, Google and Meta to fast-moving start-ups, businesses have invested billions into research, development and bringing new tools to market.
And the cold hard truth is, without them, we wouldn’t be seeing the breakthroughs we are today.
But when the private sector is the only voice shaping AI’s future, the direction it takes is going to be naturally skewed by incentives that don’t necessarily align with the public good.
Speed Of Innovation Vs Short-Termism
The private sector excels at speed.
It moves fast, adapts quickly and delivers products that capture market demand. But that agility comes with a trade-off.
Commercial pressure to show quarterly growth means decisions are often made with a short horizon. Ethical safeguards and long-term risks can be sidelined when the focus is on immediate returns for shareholders. The launch-first, fix-later approach we’ve seen with generative AI is a prime example… products reached millions of users before safeguards around bias, misinformation or safety were properly tested.
Self-Regulation Rarely Holds Up
Large tech companies frequently argue that self-regulation is enough. Voluntary codes of conduct, internal ethics boards and transparency reports are all steps in the right direction. But history shows us that when regulation is left entirely to industry, commercial priorities almost always outweigh social responsibility.
Google’s short-lived AI ethics board, which collapsed after just a week, and years of unchecked growth in social media platforms demonstrate how commercial priorities overshadow public accountability, all leading to issues that governments are still trying to untangle years later.
Innovation Without Accountability
Innovation is powerful, but without accountability it can destabilise markets and societies.
AI systems will, if we’re not careful, embed bias, automate inequality and / or disrupt entire industries before governments even realise what’s happening.
We’ve all read the case studies of AI powered facial recognition technology, for example, misidentifying people of colour at disproportionately high rates, leading to real-world harm with little recourse.
When the private sector holds both the steering wheel and the accelerator, there’s no guarantee the journey will take us somewhere sustainable.
And that’s why governments need to be more than just spectators.
The private sector brings the speed and creativity… but without public oversight and long-term guardrails, those strengths can easily tip into risks that society is left to clean up.
The Pros And Cons Of Government Oversight
Governments will never move at the speed of Silicon Valley, but that shouldn’t make their role any less critical.
Where businesses focus on innovation, governments are charged with protecting citizens, ensuring fairness and keeping long-term stability a priority (in a perfect world of course).
But that viewpoint creates both strengths and weaknesses in how they can, will and do approach artificial intelligence.
Incremental Change With Public Accountability
Unlike the private sector, governments answer to the people.
Policies need to be debated, scrutinised and voted on… often more than once. That makes progress much slower, but it also means decisions are made in the open, with public interest at the core. GDPR is a great case in point… massively criticised at first as too strict, it’s since become a global benchmark for data protection, forcing even non-EU companies to raise their standards.
AI is going to need that same kind of public accountability, ensuring that questions of bias, inclusion and transparency aren’t left solely to market forces.
The Risk Of Over-Regulation
The danger however is that governments regulate too tightly or too soon as a reactionary measure in the wrong direction.
Heavy-handed rules stifle innovation, drive investment elsewhere and create complex compliance burdens that smaller companies can’t afford. The debate around the EU AI Act illustrates this risk perfectly. Whilst it sets important ethical boundaries, industry leaders have bene pushing back, warning that excessive restrictions could push AI development to less regulated regions in a ‘
brain drain’, leaving behind slower growth and fewer opportunities.
Coordination Across Borders
AI also doesn’t (likely can’t) stop at borders.
A model trained in one country can easily be deployed globally in seconds. That makes national-level regulation fragile on its own. Without international cooperation, rules risk becoming patchy, with some countries prioritising safety and others racing ahead for economic advantage. The G7’s recent code of conduct for AI is an early attempt to align standards across nations, but much more collaboration will be needed if oversight is to match the technology’s global reach. Achieving true oversight is going to require governments to collaborate in ways that are often politically difficult, but technologically unavoidable in the long run.
Governments around the world therefore are facing the kind of balancing act we’ve never seen before. They need to be cautious without being paralysing, protective without being protectionist and cooperative without losing sovereignty. And all this in a global economic climate that doesn’t exactly lend itself to co-operation.
The challenge is immense, but so too is the responsibility and need.
So Why AI Is So Different From Past Technology Waves
Let’s face it, every wave of technology brings disruption, from the very first wheel, to steam powered mills, to the internet… but artificial intelligence represents a step change that outpaces anything we’ve ever seen before.
Unlike all the above, unlike smartphones or even machine learning, modern AI has a capacity to evolve, adapt and scale exponentially in ways that make traditional regulatory approaches look outdated (at best).
More Than Just Another Tool
Previous technologies have all extended human capabilities in some way… faster communication, better storage, improved logistics etc.
But AI goes further by taking on tasks we’ve always thought of as uniquely human; tasks that machines were previously incapable of… such as reasoning, decision-making and problem-solving. In
healthcare, AI is already analysing scans more quickly than doctors; in
finance, algorithms are deciding who qualifies for loans. None of these examples are extensions of human work, they’re replacing it.
That makes AI’s place in the world much less predictable, because outcomes can be shaped by factors even its creators can’t fully explain, predict or plan for.
Speed Beyond Regulation
Cars, planes and pharmaceuticals were all regulated successfully because those technologies had much, much longer development cycles. Governments had time to react.
AI, by contrast, develops at digital speed. ChatGPT had 100 million users withing just two minutes of launch.
New models can now be trained, released and adopted by millions in a matter of weeks. By the time a policy response is drafted, the landscape could have already shifted to something completely different.
Borders Mean Little To Algorithms
Perhaps the most striking difference is how little borders matter.
A single system can be deployed globally with no friction, creating worldwide effects overnight.
That makes purely national oversight insufficient and almost pointless. Without shared standards, the world risks fragmented approaches… with some regions prioritising ethics and safety whilst others focus only on rapid growth.
The combination of adaptability, speed and reach means AI isn’t just another technology to regulate.
It’s almost a force unto itself, that will demand new ways of thinking about governance, accountability and responsibility from all parties involved.
Looking Ahead To Future AI Developments
And the AI we see today is only the beginning.
Tools that generate text, images or code feel transformative, but they’re still incredibly narrow in scope. Over the next decade or two, researchers expect AI to move through several new stages… you can think of them as reflective, collaborative and then, even embodied.
Each stage raises the stakes for how governments and businesses shape the future of this technology.
Reflective AI
Reflective AI will refer to systems that can examine their own reasoning, identify flaws and adjust their own approach to something without any human intervention. In healthcare, for example, a reflective diagnostic tool could analyse scans, flag a potential misdiagnosis, and correct itself before the result reaches a doctor.
That ability to self-critique would make AI far more reliable but also reduce the need for as much (if any) human oversight.
If a system is constantly improving itself, who decides when its reasoning is good enough? Someone will need to ensure that “
self-correcting” doesn’t mean “
unaccountable.”
Collaborative AI
Today’s AI tools largely act alone. The next generation will work in networks, with multiple AI agents exchanging information and making joint decisions. Imagine supply chain AIs negotiating delivery contracts in real time, or fleets of autonomous vehicles coordinating traffic without human input.
Collaborative AI will undoubtedly streamline industries, reshape supply chains and accelerate research. But it will also create complex systems that move faster than regulation and policy can track. And if governments can’t keep pace, decisions that affect millions, or even billions, will be made in machine-to-machine negotiations, with little to no transparency.
Embodied AI
The penultimate progression before we hit the singularity of
AGI, would likely be embodied AI.
AI systems out in the real world, interacting physically with us, not just digitally.
Automated checkout kiosks, automated, self-drive taxis, pizza delivery drones, train station help desks… they’ll all bring new efficiency savings and new services, but they also all carry risks of widespread job displacement and raise fundamental questions of safety and liability.
When AI moves from our screens into the streets, accountability and safety is going to take on an entirely new meaning.
All of these stages represents both opportunity and disruption.
The private sector will welcome efficiency gains and cost reductions, but governments will face the responsibility of managing social consequences. How they prepare now will decide whether these shifts are stabilising or destabilising for economies and communities alike.
Economic Fallout — Why This Isn’t Just Tech Policy
A lot of the debate I see around Artificial intelligence is focussing on it as a tech issue, but to my mind at least, its economic impact is going to be felt far, far beyond just IT departments.
As new waves of AI become reality over the next ten, fifteen years, the biggest disruption won’t be technical… it’ll be on the job market, on a scale we’ve never seen before.
Governments and business leaders alike need to start thinking and preparing now for the inevitable shifts that will redefine the relationship between labour, productivity and social stability.
Mass Job Displacement On The Horizon
AI isn’t simply going to automate individual tasks; it’ll replace entire categories of work.
From customer service to legal research, logistics to medical diagnostics, many roles are already being redefined as we speak.
Over the next ten to twenty years, large sections of the workforce could well find themselves displaced, not just by single tools, but by networks of intelligent systems working faster and cheaper than people… with no obvious roles they could pivot into it.
Think about losing your job today… you could likely start working for uber, delivering pizza or a number of other manual jobs quite quickly. But if these all become automated, what do the people currently working them do? And what roles do future waves of people facing redundancy take?
The Case For Universal Basic Income
For the private sector, reduced labour costs is always going to mean higher efficiency and profit margins. For governments though… the story is entirely different.
Mass displacement is going to erode income tax bases, increases welfare demand and risk leaving millions without stable livelihoods. Universal Basic Income (UBI), once seen as a radical idea, may become a practical necessity to keep economies functioning and citizens secure.
The trick will be in figuring out how to pay for it.
Diverging Interests Between Sectors
This is what will create the biggest tension point in the years to come.
Businesses gain immediate benefit from automation, but governments will be left to handle the long-term consequences. Without coordinated planning, the result could be instability: widening inequality, social unrest and economies struggling to adapt. Bridging that gap will require both sectors to accept shared responsibility, which might look like private AI companies contributing more to the safety nets that sustain the societies they rely on.
AI is therefore not just a story of innovation.
It’s a question of economic resilience, social equity and whether governments and private companies can work together to ensure the benefits of automation are widely shared.
A Shared Responsibility Model
Neither governments nor the private sector can shape the future of AI alone.
The speed of innovation makes unilateral oversight impossible, and the scale of economic disruption makes laissez-faire approaches unsustainable. The only viable path then is shared responsibility, where both sides commit to steering AI toward outcomes that serve society as well as markets.
The Role Of Governments
Governments around the globe are going to need to act as the architects of trust.
Their role is to set ethical guardrails, enforce transparency and create policies that protect citizens whilst still encouraging innovation. That means building frameworks that prioritise equity, inclusivity and accountability, whilst also preparing safety nets, such as reskilling programmes or universal basic income, to cushion the disruption ahead.
The Role Of The Private Sector
Businesses bring the innovation, the speed, and the technical expertise. But with that power comes responsibility (sorry – I’ve tried really hard not to quote Spiderman, but it was just too fitting).
The private sector has to be willing to operate within frameworks that ensure fairness and long-term stability. That could mean contributing financially to the safety nets that automation will require, sharing data to enable oversight, and designing systems with transparency and explainability built in, or it could be something else entirely. What is for sure is that a level of co-operation will be required.
The Need For International Cooperation
AI doesn’t respect borders.
To avoid a fragmented future, where some regions emphasise safety and others chase rapid growth, international cooperation is essential. Shared standards — whether through the UN, G7, or other coalitions — will be critical to preventing a global “race to the bottom.”
The shared responsibility model is not about slowing AI down, but about shaping its trajectory so that it accelerates progress without undermining stability. It requires courage, compromise, and collaboration on both sides. Without it, the divide between short-term corporate goals and long-term societal needs will only grow wider.
Final Thoughts
Artificial intelligence will not only transform sectors. It will, without any doubt, reshape the fabric of society.
The decisions made in the coming years, about who sets the rules, how accountability is enforced and where responsibility lies, will determine whether AI becomes a tool that amplifies human progress or a force that deepens inequality and instability.
If the private sector alone drives the agenda, the risk is short-term gain at the expense of long-term security. If governments move too cautiously, innovation will stall, or worse, fracture into competing systems with uneven safeguards.
Neither path is sustainable.
But future of AI can’t be left to chance either. It requires global partnership where governments provide vision and guardrails, the private sector delivers innovation within those boundaries, and international bodies ensure consistency across borders.
Most importantly, it requires that citizens, the people most affected, remain at the centre of the conversation.
AI won’t be just another technology to manage.
It’s a turning point that will define how economies, governments and societies function for decades to come.
The question isn’t whether AI will shape the future, but who will shape AI’s future… and whether we can act quickly enough to make that future one worth living in.
