Flagship Project
Preprints Aware of Information Hazards for Security Studies (PAIHSS)
There is a problem with security studies and academic publishing. Security Studies are often defined by periods of crisis: wars, terrorism incidents, pandemics. Something that all periods of crisis have in common, is that they are rapidly evolving with analysis and results often needed on time-frames of days or weeks. Peer reviewed journals, by contrast, function in a plodding and deliberate manner with even the most expedited publishing time frames being months.
Preprint severs have, in some disciplines, been the solution to accelerate the world of academic publishing by allowing the sharing of results to not be delayed by the glacial response times of academic publishing. However, preprint servers are a relatively new innovation (the earliest ones date back to 1991, but most were established after 2010) particularly outside of physics, computer science, and math. As of this writing, 73 preprint servers have existed and 65 are still in operation. Yet, zero preprint servers are devoted to or even significantly include security studies. (Cyber security being the notable if limited exception). There are two reasons for the absence security studies in preprint servers:
Security studies are intrinsically multi-disciplinary, as "security" comprises all factors that influence success or failure of either offense or defense of a given target and all consequences thereof. This has has been a soft barrier to penetration by preprint servers into security studies because they have, following the lead of peer reviewed journals and university departments, mostly divided themselves along lines of academic disciplines. As this amounts to nothing more than inertia and tradition, it does not qualify as a valid reason for security studies to not avail themselves of the accelerating properties of preprint servers. Indeed, the need to break down such academic-disapline silos is recognized as a crucial need for security studies.
More legitimately, there is a concern that in publishing for security studies, that scholarship might unintentionally provide bad actors with "blueprints" on how to do bad things. Effective altruist has been long aware of the potential for inadvertently revealing dangerous information to bad actors as one class of the larger concept of "information hazards". This IS a valid concern that must be overcome before security studies can avail themselves of preprint servers.
There is a solution: Effective altruists have studied information hazards more intensely and systematically than perhaps any other group with the possible exception of the intelligence community and are therefore the perfect community from which to develop a preprint server that mitigates the risk of information hazards while still gaining the advantages of a faster publishing cycle. (It is important to note that such a mitigation need not be perfect to be useful... current preprint servers may not be set up to facilitate security studies publishing, but they don't ban it, so any mitigation of preprint mediated information hazards is a net win). If you are interested in how we intend to mitigate preprint mediated information hazards please feel free to contact us directly!
The PAIHSS project aligns very closely with the underlying identity and conception of the Archimedes Network. We recognize the both the imperative and existential implication of academic study.
Imperative: We can not afford to let our knowledge, our understanding, our study, and our invention slow or falter for these represent our instruments of survival, and if we slow down, others perhaps less ethically motivated won't.
Existential: There is nonetheless danger from forging ahead too.
Any rational approach to security must balance the risks of both speeding up and slowing down. The Preprints Aware of Information Hazards for Security Studies (PAIHSS) is meant to provide us with the first of many tools to build such a balanced approach.