Effective AI regulation and international cooperation are essential to harness AI’s benefits while ensuring safety

Many turn to National Newswatch for insights on policy and politics. Here’s why artificial intelligence (AI) should now be central to these themes.

Discussion of AI splits people into two camps: optimists and avoiders. I propose a middle path – AI pragmatism – as our best way forward. To see why, let’s start with what readers like you already know.

Maybe you’ve tried a chatbot or seen apps like Sora create stunning videos from a few lines of text. You might know AI is set to manage travel plans, drive cars, and enhance investments.

But what if I told you AI is also our best shot at revolutionizing health care, combating climate change, and modernizing governance? Sound far-fetched? Let’s see.

First, consider the incredible scale and pace of AI development. Three years ago, training a chatbot might have cost $10 million. Now, advanced models cost hundreds of millions and can use trillions of words to train—yes, trillions. And there’s no sign of slowing down.

Microsoft and OpenAI are investing $100 billion in “Stargate,” a supercomputer to support vastly superior AI. This could be one of the biggest infrastructure projects ever.

But the breaking AI story isn’t about the mind-bending tech; it’s about a coming knowledge storm. Chatbots already speak in human-like voices, in hundreds of languages. Google’s AI can outperform doctors in diagnosing ailments. AI has been used to pilot F-16 fighter jets in dog fights. Meta uses AI to translate brainwaves into eerily accurate pictures of the images in someone’s mind.

And that’s just the beginning. Soon, “agentic” AIs -- AI systems that possess a degree of autonomy and the ability to not only make decisions but also enact them -- will plan and execute complex tasks using teams of specialized bots; humanoid robots will work alongside us on the job; and quantum computers, thousands of times more powerful than today’s supercomputers, will supercharge machine learning. This isn’t science fiction.

Of course, these reports can be disturbing, as AI avoiders rightly insist. Among their main concerns:

• Jobs: AI will soon read and analyze legal contracts in seconds, design neighborhoods and cities, and execute complex business tasks, making many white-collar workers fear obsolescence.
• Deep Fakes: AI can create fictional videos or voice recordings indistinguishable from the real thing. Will bad actors use these tools to spread misinformation and undermine trust in journalism, politics, and ultimately, truth?
• Terrorism: AI could be used to create devastating chemical or biological agents – or even nuclear devices.
• Terminator: We’re all familiar with the scenario where computers become super-intelligent and see no further need for humanity.

These fears have convinced some that AI is spiraling out of control, leading them to urge governments to hit the pause button before something catastrophic happens. The famous “pause letter,” signed by AI leaders worldwide, pushed these concerns to the fore. Yet the AI machine spins faster than ever.

There are real risks, and AI avoiders are right to point them out, but let’s be clear: no one can stop this juggernaut. The momentum behind AI is massive, driven by huge economic interests and geopolitical forces, and not all are friendly.

China is a global leader in AI and won’t stop investing or developing it just because “the west” does. There’s a real arms race logic here, and halting development would simply hand the lead to authoritarian governments. That’s not going to happen.

Moreover, optimists are right to insist on the extraordinary gains and AI’s immense potential for innovation. AI can analyze vast data sets to find solutions previously beyond our reach, from discovering new materials to tackling climate change and personalizing education.

So, who should we believe: the optimists who say AI will lead to a utopian future, or the avoiders who see AI as a looming threat to be suppressed or locked away?

The pragmatist’s answer is that no one knows the future, but history is plunging ahead. We are far from John A. Macdonald’s vision of a sprawling, bucolic nation-state, bound by a transcontinental railway. And we are never going back.

Societies like ours must rise to a new challenge, creating new forms of social organization to manage our economy, healthcare system, and geopolitical relationships effectively and democratically. And that’s where pragmatism comes in.

On one hand, we need new tools to tackle these challenges. On the other hand, the only thing on the horizon with the power and breadth to support such a transformation is AI. So basically, we need AI to respond to the challenges that AI is creating.

Pragmatism manages this dilemma by looking at AI as a new generation of incredibly powerful tools that could transform society and democracy, but also realizing that nothing is guaranteed. Everything turns on how we use these tools.

The first step is to recognize that AI is much more than a tech issue. It is a policy imperative of the highest order that demands urgent attention. Effective AI regulation and international cooperation are essential to harness its benefits while ensuring safety.

To their credit, big tech firms are now pouring billions into AI safety. But the broader policy community must engage and lead. Policymakers must articulate and prioritize the issues, develop the rules and frameworks to guide AI development, and align it with public interest and ethical standards.

Members of the AI community routinely speak of AI as “the most powerful technology ever”—bigger than fire, the wheel, electricity, or the internet. They say it will change everything.

I agree.

This weekly column will be a space to probe, discuss, speculate, and inform. I want to confront the risks and explore the opportunities to renew and transform our society and governments.

I hope you’ll join me.

Don Lenihan PhD is an expert in public engagement with a long-standing interest in how digital technologies are transforming societies, governments, and governance.