It seems the Pope and a professor of astrophysics at Cambridge University may share similar views about the damage to humanity inherent in the AI project.
They join other household names - Stephen Hawking, Geoffrey Hinton and Paul McCartney among them - who have warned of the dangers of enabling agentic AI to proceed without human oversight.
Writing recently in the Bulletin of the Atomic Scientists, Professor Hiranya Peiris, who holds the Professorship of Astrophysics at the University of Cambridge’s Institute of Astronomy, described the current behaviour of AI systems as “laying down the tracks ahead of a speeding train”.
She concluded that the lack of safety guardrails governing overall decision pathways means AI-driven systems can chart a course to catastrophe faster than humans can detect the danger of the trajectory, never mind reverse it.
The problem – that while no individual move was itself concerning, the overall direction of play was devastating – is a pattern present across all AI models and not governed or restrained by any current safety mechanism, Prof Peiris warns.
“Currently,” she noted, “this problem does not have a solution.”
The analysis has been praised by *Forbes *magazine, which drew parallels to Pope Leo XIV’s recent encyclical, titled Magnifica Humanitas. The encyclical suggests that AI should be “disarmed” because it threatens human dignity, agency, and social justice. The AI arms race - military competition is creating fearsome military weapons - will result in automated warfare, which will make wars easier to start and harder to control, it says. The technology’s water requirements will damage the ability of tens of millions of people to survive, it notes. And corporations are creating a world of “digital slavery”, where economic profit systematically sacrifices jobs.
The analysis in *Forbes *argued both publications highlighted the current moment as “a hinge in history, the kind where the choices of a few determine the fate of many”.
The Murray Edwards alumna and Fellow has admitted to being “surprised” that her analysis of the failings of AI oversight explored similar themes – and reached similar conclusions – to the recent Papal encyclical.
She tod the Cambridge Independent: “What surprised me was where the coverage landed, being read alongside the Pope’s AI encyclical in Forbes. I didn’t expect a piece about AI safety to travel that far.”
Prof Peiris is now advocating a more rigorous testing process for AI development.
“I am working now to make this case to the technical audience, solidly enough to change how these systems are tested,” she notes.