Tl;dr: Technical progress in DNA synthesis has outpaced regulatory safeguards against negligent and malevolent misuse of DNA synthesis technology. The need for intervention from the community has been on our collective radar for sometime and still coordination on the hottest issues looks underwhelming.
Epistemic status: Low - Master's degree in epidemiology and broad expertise in public sector/health policy but <5 hours reading on this issue, synthetic biology, bioterrorism etc
The conversation on biosecurity in the EA community appears to somewhat infrequent and when it does to me it appears immature - short of leadership and expertise proportionate to the imminence, scale, neglectedness and tractability of the threat and solution, and the standard of analysis we expect in analyses and ideas proposed by EAs in the global health and development space.
The problem is that DNA synthesis procedures enable persons to render pathogens undetectable by established methods, vaccines ineffective and increase virulence or lethality [source]. In a pandemic simulation by the Johns Hopkins Center for Health Security in 2018, a terror group modeled on Aum Shinrikyo engineered a virus that killed 150 million people [source]. At the 2021 Munich Security Conference, the Nuclear Threat Initiative wargamed an exercise where a terrorist group engineered a strain of monkeypox in a country with weak oversight that left 270 million dead and more than three billion sickened [source].
Engineered viruses are already cheap and could be even more readily available to bioterrorists or negligent entities in the immediate future:
- DNA synthesis technology lets you “print” the sequence and create your own copy of the virus once you have the digital sequence for a biological agent. [source]. The cost of DNA sequencing dropped from $2.7 Billion to $300 in under 20 Years [source]. The cost of gene synthesis per base pair has dropped from approximately $10 a base pair to approximately $0.10 a base pair over the past 10 years [source]. The smallest viral genomes are 1000-2000 base pairs [source].
- By one estimate [source] well short of AI Impact’s survey standards, tabletop DNA synthesis that could widely proliferate biological weapon capabilities are less than 20 years away.
U.S. federal government funding for civilian biosecurity in 2012 was around $560 million [source]. The only philanthropic funders of biosecurity identified by Open Philanthropy in their pre-COVID-19, 2014 shallow investigation was the Skoll Global Threats Fund, the Gates Foundation and Rockefeller Foundation which together funded biosecurity work by the Nuclear Threat Initiative. Since the advent of COVID-19 the landscape has changed but appears inadequate. To illustrate, the Nuclear Threat Initiative spent just under $6 million on the biological program in 2021 [source].
Does anyone have a more up-to-date characterisation of the synthetic biological terrorism funding landscape?
It was reported in September 2021 that the United Nation’s Biological Weapons Convention has 3 staff members [source] whereas the Chemical Weapons Convention has hundreds (and has been how effective, to consider tractability). Screening DNA synthesis orders is not legally required by any national government [source] and The Johns Hopkins Center for Health Security found in 2021 that 94% of countries have no national oversight, laws, regulation, agency or assessment of dual [source]. Should we agitate within our home countries to strengthen biosecurity regulatory regimes?
An estimated 80% of global DNA synthesis providers belong to the International Gene Synthesis Consortium which voluntarily screen DNA synthesis orders and the remainder are believed to go unscreened [source].
The Nuclear Threat Initiative are coordinating international dialogue around biosecurity and DNA synthesis [source]. However, I see nothing published that could indicate a timeline for concrete regulatory strengthening, in their patch or otherwise. Is there any known work on this? I know they are developing a software package for their "common mechanism’’ [source] for every DNA provider to be able to screen DNA sequences, making it easier and cheaper for them to screen DNA sequences and customers efficiently and at lower cost. However, it's unclear whether anyone is seriously working towards a global regulatory mechanism to compulsory that could control against opting out of screening, or illicit practice in under-regulated jurisdictions.