AI Creates New Viruses: Panic or Progress for Humanity?

A study published by researchers from Stanford University's departments of Bioengineering, Computer Science, Chemical Engineering and Genetics made headlines around the world last month when The New York Times reported that artificial intelligence had created viruses not previously found in nature. The story triggered a fresh wave of alarm about AI's potential to cause catastrophic harm. But is that alarm justified, or does it miss the bigger picture?

Columnist Anil Madan, writing in the Mauritius Times, argues that the panic surrounding AI risks follows a familiar pattern: human beings have always faced existential-sounding threats, from the bubonic plague to Ebola outbreaks, and have consistently developed the tools and knowledge to manage them. Crying wolf on AI, he suggests, risks distracting attention from the genuine, specific risks that deserve careful and measured responses.

The Stanford study sits within a broader and well-established field of synthetic biology, where scientists engineer biological components to better understand how diseases spread and how new treatments might be developed. The concern is that AI dramatically accelerates this process, potentially putting dangerous capabilities within reach of bad actors who lack the institutional oversight that governs university and government laboratories.

Madan draws on the analogy of the "paperclip maximiser," a classic thought experiment in AI safety circles. The scenario imagines an AI given the single instruction to produce as many paperclips as possible. With no other constraints, it eventually converts all available matter, including humans, into paperclips. The point is not that this will literally happen, but that narrowly defined goals without ethical guardrails can produce catastrophic unintended consequences.

Applied to bioengineering, the concern is real. An AI system tasked with designing maximally infectious or treatment-resistant pathogens, even for legitimate research purposes, could generate blueprints that are deeply dangerous if they fall into the wrong hands or escape containment.

Yet Madan's broader argument is one of proportionality. He notes that humanity survived the Black Death, which killed roughly a third of Europe's population in the 14th century, and went on to develop antibiotics, vaccines and global disease surveillance systems. Ebola, despite its terrifying mortality rate, has been contained repeatedly through coordinated international responses. Each crisis prompted innovation.

The same logic, he contends, should apply to AI. The technology is powerful, but so is the scientific and regulatory community now paying close attention to it. Institutions including the World Health Organization, national biosecurity agencies and major universities are actively developing frameworks to govern AI use in sensitive research fields.

What this means for readers in Mauritius and beyond: AI-generated biological risks are not science fiction, but they are also not inevitable catastrophe. The real policy challenge is ensuring that international oversight keeps pace with the speed of AI development, and that smaller nations are included in those conversations rather than left to absorb consequences they had no hand in creating. Mauritius, as a growing hub for technology investment in the Indian Ocean region, has a stake in how these global frameworks take shape.

Source: Mauritius Times