What risks come with increasingly powerful and autonomous AI systems? Professor Pekka Abrahamsson, Head of GPT-Lab at Tampere University, joined AI entrepreneur Mikko Alasaarela on Yle’s A-studio to discuss AI safety, the need for safeguards and Europe’s role in shaping the future of artificial intelligence.
The A-studio discussion focused on a question that is becoming increasingly important as artificial intelligence develops: how concerned should we be about AI?
According to Abrahamsson, the risks associated with AI should not simply be dismissed as hypothetical scenarios or narratives created by major technology companies. As AI systems become more capable and autonomous, they are increasingly being considered for critical applications where mistakes can have serious consequences.
One example discussed was the use of autonomous AI in military systems. Such applications underline a broader challenge in AI development: powerful systems need clear boundaries, safeguards and ways for humans to maintain meaningful control.
Abrahamsson illustrated the challenge with a rallying metaphor:
“We can drive 200 km/h on a gravel road, but we need a co-driver. The co-driver keeps the drive safe.”
The faster AI develops, the more important these safeguards and regulations become.
Europe needs strategic AI sovereignty
The discussion also turned to Europe’s position in global AI development.
Abrahamsson highlighted strategic sovereignty as one possible direction for European artificial intelligence. In practice, this means strengthening Europe’s ability to develop its own AI models, technologies and infrastructure instead of relying primarily on systems developed by large US technology companies.
Europe already has AI research expertise and technological capabilities. The question now is whether there is enough ambition and investment to turn that expertise into European AI models, infrastructure and applications at scale.
For Abrahamsson, Europe has an opportunity to take a more active role in shaping the development of artificial intelligence.
The question is ultimately about agency: who develops AI, who defines its boundaries and who gets to decide how it is used?
Professor Pekka Abrahamsson leads GPT-Lab at Tampere University, a Finnish artificial intelligence research laboratory studying generative AI and its impact from software development to organizations and society.
The A-studio discussion with Pekka Abrahamsson and Mikko Alasaarela aired on Yle on 15 September 2026.
Watch the show in Finnish here: Pitääkö tekoälyä pelätä? | A-studio | Yle Areena
