The Genetic Evolution Tournament (GET) is a Metaculus human judgment forecasting tournament established to generate forecasts and scenarios pertaining to the use of human genetic and reproductive technologies for treatment and enhancement.
Hosting
Contact:
On Feedback: All feedback is welcome; however, feedback that enhances
is particularly desired.
Disclaimer: This description, which possibly contains inadequacies, is a minimally viable placeholder. Beyond the content detailing GET, there are many unsubstantiated claims. At some point, this tournament might be published as research, in which case the author will more strongly evidence claims regarding the landscape of current capabilities for human enhancement and treatment via genetic and reproductive means. GET questions often contain references to research literature, but these are far from comprehensive. The Extended Introduction section is liable to change.
Overview: _This introduction starts with some motivating remarks on possible future(s) for human reproduction and human genetic constitution. Following this, the author briefly describes GET, which is currently under development and due to be released on Metaculus (as a public tournament)._
Scientific progress in genetics, genetic engineering, and assisted-reproduction continues at a rapid and accelerating pace, with current technological capabilities far exceeding those imagined by researchers and policy-makers in biotechnology at the start of the 21st century.
Frontier events in genetic engineering, such as the Jiankui He affair, have required all of humanity to consider scenarios and possibilities—ranging the gamut of expected value from fantastical to catastrophic—concerning the manipulation of human biology.
Although extreme-fidelity human gene modification (as in precisely exacting desired genetic outcomes) or, more broadly, modification of human biology at a caliber similar to what we might expect to be employed in _Brave New World or Gattaca_, is yet unrealized, developments in the aforementioned fields steer evermore towards this capacity.
Fortunately, considerable anticipatory work in ethics and governance have coincided with scientific and technological progress in genetic and reproductive research fronts. Ethical work on directed use of technology on human biology has involved some degree of implicit and or explicit scenario analysis, but this analysis (in the author's sampling of the literature) has infrequently employed forecasting.
Handling anticipatory blind spots and reducing uncertainty on futures involving humanity's possible genetic evolution via forecasting seems important for coordination between researchers, the public, and policy-makers in directing outcomes and safeguarding certain aspects of humanity's future.
Despite the expected outcome for GET being that most community traction will occur within the Effective Altruism, Less Wrong, and Metaculus communities, GET seeks to occupy the general niche of uncertainty reduction on futures involving varying levels of human-targeted DNA and assisted-reproduction technologies.
GET will ideally contribute, via the forecasts collected across its lifespan, to a broader scenario modelling effort occurring across discussions on the future(s) of human genetics and reproduction. These forecasts and their analysis might inform individual and policy decision-making on the technologies and procedures involved in human genetics and reproduction.
With respect to the forecasting platform for GET, given both Metaculus's status as a reputable human judgement forecast (HJF) aggregator (this is the author's impression) and the author's familiarity with the site (the author is biased in this way), Metaculus was chosen as the site to host GET. The particular language and norms adopted in public tournaments on Metaculus typically incentivize forecasters to make clear and well-reasoned comments and to forecast as honestly and as accurately as possible. For example Metaculus tournaments, see this page.
GET currently consists of 5 question categories:
Each category was intended to have between 15 and 25 questions; the intention was to between 75 to 125 questions in total.
To incentivize forecaster participation, there are prizes totaling 15K USD, though this may increase in the future. Prizes are partitioned into a forecasting accuracy component, with questions scored via Metaculus's default scoring procedures (peer scores), and a commenting component.
Originally, within the commenting component, 2.5k USD was allocated to comments on questions that are part of the forecasting accuracy component (shorter-term questions) and 7.5k USD was allocated to comments on longer-term questions. However, after some deliberation, a decision was made to break the commenting section of the tournament into 3 stages, each with 2k USD. The tournament is expect to last roughly 20 years. There is the possibility for additional funding altering the structure of the tournament.