Advances in artificial intelligence (AI) and biotechnology, the growth of high‑risk, high‑reward research, rising geopolitical tensions, and strains on public health systems all increase the chance that a human‑made lethal pathogen could emerge — either accidentally or intentionally — and that we might not be prepared.
Why this matters
The same technological progress that promises medical breakthroughs also raises the risk of catastrophic laboratory incidents. Experts disagree about the magnitude of that risk. Ashish Jha, the Biden administration’s COVID response coordinator, warned that a ‘‘very small tweak’’ can turn gene therapies aimed at a disease into a virus that is much more harmful. Jha has since founded a company focused on detecting emerging biological threats.
Key concerns
- Worst‑case scenarios include malicious actors using AI‑enabled biological agents to cause mass casualties. Major AI companies have acknowledged this danger and urged stronger safeguards.
- Jha said his primary worry is not necessarily a nation‑state deploying a biological weapon, but an accidental release stemming from a deliberate program.
- Despite political backlash around ‘‘gain of function’’ research, the number of labs conducting potentially risky biological research has increased, not decreased, according to a former senior official at the U.S. Department of Health and Human Services (HHS).
- AI tools can allow people with limited experience to perform or navigate experiments with a higher chance of success, which raises concerns about misuse and accidents.
Recent developments
News that Stanford researchers used AI to design new synthetic viruses capable of infecting and killing the microbe E. coli heightened worries about similar techniques being attempted by bad actors. The Stanford work may have important medical uses, but it also illustrates how the same methods could be dangerous in the wrong hands.
Reports have also shown that chatbots have been manipulated into providing users with instructions on how to create deadly biological agents.
Reality check: practical limits
- AI‑assisted biology still requires substantial manual laboratory work, which remains a barrier to weaponization. David Manheim, founder of the Association for Long Term Existence and Resilience, noted that AI cannot eliminate the need for physical checks and hopes that labs will not permit AI systems to perform arbitrary operations that are not fully understood.
- Creating a virus that targets bacteria is an important step but remains far from the task of engineering a virus that infects humans.
Why concerns persist
AI is rapidly evolving. The Stanford findings were first revealed in a preprint nearly a year earlier, suggesting additional progress could have occurred since. Jha frames the issue this way: given the growing number of people working in this space, ‘‘what are the chances that no one will succeed in the next five years?’’
Intelligence and government assessments
This year’s U.S. intelligence community threat assessment states that "China, North Korea, and Russia probably maintain the knowledge and capability to produce and employ traditional biological pathogens and toxins," and that historically these actors have pursued pathogens with high infectiousness or contagion. The report adds that advances in fields such as nanotechnology and genome editing "could lead to novel biological threats or raise the potential for biological safety events resulting in the unintentional release of pathogens."
Defense and mitigation
- AI can also strengthen biodefense. Google DeepMind announced last month a bioresilience program aimed at preventing, detecting, and responding to biological threats.
- A former senior HHS official said the upside is that AI and advanced biology give us better tools today than in 2020 for responding to pandemics.
- HHS said in a statement that the U.S. "remains fully equipped to respond to emerging public health threats" and protect the public, citing its responses to a hantavirus outbreak on a cruise and to an Ebola emergency in Africa.
Conclusion
AI and advanced biotechnology present both opportunities for medical progress and increased risks of severe lab accidents and biological threats. While experts debate how large the overall risk is, recent developments and the growing number of researchers in the field have prompted calls for stronger safeguards, improved detection capabilities, and continued investment in biodefense.
