This article is part of the World Economic Forum's Geostrategy platform

A vast array of technologies is rapidly developing and converging to fundamentally change how research is performed, and who is able to perform it.

Gene editing, DNA synthesis, artificial intelligence, automation, cloud-computing, and others are all contributing to the growing intelligence and connectivity of laboratories, finds a report from The Wilson Center, The Intelligent and Connected Bio-Labs of the Future: Promise and Peril in the Fourth Industrial Revolution.

It is currently possible to perform a growing number of research tasks automatically and remotely with a few clicks of the mouse. And with the barriers of entry to synthetic biology tools decreasing, they will no doubt be subject to automation as well, and may even be coupled with artificial intelligence to optimize the power of genetic engineering.

While this may be a boon for the development of novel vaccines and therapeutics by parties that have traditionally not had access to the necessary tools, it also opens the risk of nefarious use to engineer or edit biological agents or toxins.

While there have been attempts at governance to limit the avenues by which a bad actor may gain access to the pathogens or tools to create biological weapons, the ever-increasing pace of innovation has left gaps that may be exploited. Fortunately, investment in technologies such as artificial intelligence and sequencing may also function as the best defence against the growing threat of misuse of biological agents.

Examining our vulnerabilities

Growing Accessibility

Unlike when the Biological Weapons Convention (BWC), and other initial governance aimed at biological attack was formulated, the greatest threat may no longer be a biological agent escaping a BSL-4 laboratory (the highest level of biosafety lab). Instead, the potential for peril is increasingly accessible and global.

While the remaining strains of smallpox are only believed to be stored at two highly-secured BSL-4 labs (one in the US and one in Russia), the complete genetic sequence is available to anyone with internet access. While the process to bring the virus ‘to life’ may currently take technical know-how, it will not always remain this way. The reconstitution of horsepox proves that this feat is currently possible when performed by experimentally capable actors. And the global number of capable actors is increasing at a staggering rate due to technological advances.

Global BSL4 Capability. Red dots indicate a laboratory with BSL-4 (or the local equivalent) capability. Increasingly, the threat is not a biological agent leaving one of these laboratories, but instead an agent being engineered using publicly available sequences, synthetic biology tools, and automated and intelligent processes. This is greatly increasingly who has access to dangerous toxins and agents for malicious use.

Taking Note of the Threat

The potential of new biotechnologies to become weaponized has not escaped the notice of the US Intelligence Community. Their 2016 Worldwide Threat Assessment noted:

“Research in genome editing conducted by countries with different regulatory or ethical standards than those of Western countries probably increases the risk of the creation of potentially harmful biological agents or products. Given the broad distribution, low cost, and accelerated pace of development of this dual-use technology, its deliberate or unintentional misuse might lead to far-reaching economic and national security implications.”

Further, the President’s Council of Advisors on Science and Technology (PCAST) noted in a 2016 report the potential for intentional misuse of technology developments such as CRISPR, the gene editing tool. PCAST notes that “a pathogen might be deliberately modified to affect its spread or be resistant to current preparedness and response capabilities”.

Yet, while these reports work to show the acknowledged threat posed by genome editing and synthetic biology, they fail to acknowledge the increased risks and decreased timeframe posed by the convergence of biology with other emerging technologies. The bioterrorism events we have seen in the past largely relied on the weaponization of known pathogens, such as anthrax or glanders – an infectious disease affecting horses – that were taken from laboratories or hospitals.

As many of these agents have been studied in the context of response to an attack, we have readied responses for them. The United States, for instance, keeps a Strategic National Stockpile housing vast quantities of an anthrax vaccine.

With the convergence of synthetic biology with artificial intelligence, though, we may be unable to mount such a rapid response. Soon, a pathogen may be engineered to evade current vaccines and therapies, causing an event intended to reach weapons of mass destruction (WMD) potential.

Exploiting the Gaps

While the technologies that are emerging and converging to form more intelligent and connected labs have great promise, they do indeed also have great risk.

The combination of synthetic biology with AI and automation, all of which are becoming more pervasive, means that tools traditionally siloed in academic and government labs are slowly becoming accessible to a wider audience. Indeed, the technologies are already nearly in place for an actor to co-opt synthetic biology tools to re-create a biological agent or a toxin, and largely accomplish this in a hands-free manner.

Scenario

By purchasing a used DNA synthesis machine online for less than $10,000, an actor may wield the power to engineer biological agents without the need to order pre-assembled oligos or genes from a vendor. While this machine could be used to assemble much larger genes or even genomes, a much faster way to engineer a sufficiently dangerous agent would be to synthesize short guide RNA sequences. When coupled with the easily available Cas9 enzyme, an actor could edit a virus or bacteria to be more pathogenic using the CRISPR system.

In possibly only a few edits, a pathogen could be made to be more virulent or transmissible. The bird flu virus H7N9, for instance, appears to require only three mutations to gain the functionality of easily spreading among human populations.

In a virus that already causes a death rate exceeding 40%, it is easy to imagine that an engineered form of H7N9 would cause havoc across the healthcare system and the economy. And shortly, AI advances may allow anyone to sequence the genome of an obtained virus or bacteria, and determine edits that may optimize a number of dangerous traits.

The cost would be exceedingly low and could easily be affordable by a small group or even a lone-wolf actor seeking to do harm. With increasingly advanced bioinformatics tools along with automated and cloud-connected lab capabilities, one may not even need much tacit knowledge of biological experimentation to obtain an agent capable of widespread incapacitation and even death, nor would they need to handle the agent extensively before deployment.

While it would currently be much easier to procure a reconstituted organism or toxin through its available genetic sequence than edit it to make it more virulent, the barriers to edit genes and genomes are steadily decreasing.

With other advances forthcoming, current governance surrounding the access to and sales of biological agents and technologies could be easily circumvented.

Exploiting the gaps. Given advances in DNA synthesis techniques and the advent of robotic cloud laboratories, one may find ways to circumvent current governance barriers. Example companies to complete each step are listed in parenthesis.