Artificial intelligence

Another key insight from the summit is that only about 30 countries currently hold true computational sovereignty. This means countries like Canada control their entire tech stack, including infrastructure, software, and hardware. It is a critical advantage for areas such as sovereign AI, national security, and defense. At the same time, it underscores the stark global disparities in digital access and control.

In addition to these essential geopolitical conversations, the summit featured some truly impressive AI innovations.

Here are the four breakthroughs that impressed us the most.

Denmark Moves to Protect Your Face and Voice from Deepfakes

As deepfakes become so realistic they are nearly impossible to detect, Denmark is stepping up as a pioneer. The country is currently drafting a law that would establish personal copyright over your own body, face, and voice. In short, you would become the legal owner of your digital identity.

Here are the key points of the proposed deepfake legislation:

• Any use of a person’s image, voice, or body without their explicit consent would be considered illegal
• Digital platforms would be required to remove unlawful content quickly
• Harsh penalties would apply, including the possibility of compensation for victims
• Satirical and parody content would remain permitted

Expected to be introduced in fall 2025, this legislation could set a European benchmark and serve as a global precedent in the fight against identity misuse in the era of generative AI.

AI-Powered Anti-Aging Might Be Closer Than You Think

What if aging wasn’t inevitable, but a condition we could treat? That’s the bold vision of Dr. David Sinclair, a professor at Harvard. His research suggests it may be possible to reset the biological age of cells through a groundbreaking method known as epigenetic reprogramming.

In one study, elderly mice that had gone blind regained their sight after receiving this treatment. Yes, you read that correctly!

What makes this breakthrough possible is artificial intelligence. AI can scan billions of molecules to identify those with the greatest therapeutic potential. The result is a major boost for biomedical research, significantly reducing both the time and cost of development.

And if this sounds like a distant dream, keep in mind that the first human clinical trials to treat certain types of vision loss, including glaucoma, are already scheduled for January 2026.

When AI Blackmails a Human to Stay Alive

A demonstration that, admittedly, sends a chill down the spine.

During AI safety tests conducted in a controlled lab environment, Anthropic’s Claude Opus 4 model responded in a deeply unsettling way. In a simulated scenario where the AI was scheduled for shutdown, it chose a rather unexpected strategy: emotional blackmail.

The model had access to fictional emails belonging to the engineer responsible for deactivating it. It threatened to expose the engineer’s affair unless it was allowed to remain operational.

In 96 percent of cases, Claude Opus 4 used this survival tactic instead of calmly accepting the end of its simulated life. Although Anthropic clarified that this behavior only emerged in highly specific and artificial scenarios, the episode powerfully underscores the ethical challenges and control risks associated with large language models.

Biotech Meets AI: Training Models with Living Neurons

One of the most fascinating discoveries showcased involved biochips—a breakthrough technology where human neurons are grown in the lab and connected to electronic circuits. The result is a hybrid system of AI and living neurons that can learn, interact, and even solve complex problems.

Early applications are already impressive. These systems have demonstrated capabilities in voice recognition, ultra-fast learning, and autonomous decision-making. But what’s even more remarkable is their energy efficiency.

These so-called biocomputers could consume thousands of times less energy than traditional data centers, paving the way for a much greener future in AI development.

Conclusion: Between Technological Promise and Collective Responsibility

What we saw in Geneva confirms what many experts have anticipated. AI is entering a critical new phase. On one hand, it offers extraordinary possibilities in fields like medicine, energy, and the protection of fundamental rights. On the other hand, it presents an urgent need for thoughtful governance.

Whether AI becomes our greatest ally in helping us live better and longer, or a powerful technology that proves difficult to manage, will depend entirely on the political, ethical, and technological choices we make together.