Executive Summary

• Our Moral Parliament Tool allows users to navigate normative, metanormative, and empirical uncertainty. It does so by representing diverse worldviews as delegates in a democratic decision-making process about how to distribute resources to various causes or projects.
• We have added functionality to the original Parliament Tool, which allows it to be adapted to novel resource allocation problems. In this sequence, we demonstrate the versatility and utility of the tool by using it to analyze various types of problems.
• Here, we present a parliament to set funding priorities across different kinds of EA movement-building projects, based on worldviews that reflect different beliefs about movement-building strategy.

Introduction

In this sequence, we explore how RP’s Moral Parliament Tool can be adapted to support deliberations about a range of resource allocation problems. The original version of the Parliament allows users to compare various philanthropic projects in light of different moral theories. Here, we consider how it can be used to address a more strategic (and less directly moral) question: how should EAs distribute resources among different movement-building projects?

This application of the Parliament is far more speculative than the others in this series. We lack reliable data on the effectiveness of various movement-building projects in attracting new members and the degree to which those members align with EA values. We present this parliament as a template that could be used to facilitate decisions among EAs with diverse sets of priorities and strategies for growing the movement.  

Movement building worldviews and projects

There are lively debates within EA about how best to invest in the movement^[1], some of which we attempt to capture through the following dimensions:

First, there is often (though perhaps not inevitably) a trade-off between the total number of members and how value-aligned or engaged those members are. It may be possible to attract more members by having lower standards for inclusion or by making some movement commitments less stringent. Other strategies may attract fewer adherents but filter out those who are not highly committed to central EA principles. Delegates may disagree in how strongly they prioritize member number vs. member alignment.

Second, debates over cause prioritization within EA often spill over to debates about how to build the movement. Some delegates may prioritize growing the movement in ways that prioritize particular EA cause areas (e.g., AI safety or animal welfare). In contrast, others may prefer to build a “big tent” movement. We capture this through weights assigned to different AI cause areas: GHD, Animals, GCR, and AI.

Third, delegates may have varying levels of risk aversion. Some may be willing to take actions that could backfire and harm the movement, as long as they also offer a decent chance of yielding substantial returns. Others may be more cautious, favoring projects that more reliably and steadily grow the movement.

We use these dimensions to characterize four worldviews (though others are possible, and users are encouraged to build their own):

• Mass Movement: a risk-neutral worldview that prioritizes having a large number of EA members distributed across a range of cause areas
• AI Elites: a risk-tolerant worldview that prioritizes having a smaller number of highly aligned EAs who are devoted primarily to AI safety
• Broad Elite: a risk-neutral worldview that prioritizes having highly aligned EAs working on a broad set of cause areas
• Mainstream Philanthropy: a risk-averse worldview that prioritizes EA’s appeal to mainstream philanthropic audiences

We consider five hypothetical projects, which are not intended to model real projects accurately:

• AI videos: produces AI safety content for YouTube
• Giving commitment ad campaign: an advertising campaign to attract mainstream donors to an effective giving pledge
• Campus groups: invests in building EA groups at elite universities
• AI safety fellowship: funds fellowships to support career transitions to AI safety
• Press engagement: focuses on building positive perceptions of EA through press engagement

In this model, these projects are assumed to have similar scales (i.e., expected value added, were you to fully value everything the project does). Projects differ in the variance of their effects, with projects having greater or lesser chances of failure, great success, or backfire. These scale assumptions are not realistic, and the results of the parliament are quite sensitive to changes to them. Without reliable data that would allow us to set more realistic scales, we don’t endorse the model's conclusions as they stand. We offer it as an illustration of how the parliament could be used and a template for anyone who wants to input their own numbers.

The resulting Parliament can be found here (spreadsheet here). Given the speculative nature of the project, we will leave it to users to explore the effects of changing parameters in the model. The model can also be adapted, via the spreadsheet method, to handle other kinds of movement building strategies and resource allocation problems.

Conclusion

 We intend the examples in this sequence as a proof of concept that the basic framework of the Parliament Tool can be applied to a diverse array of problems. We do not intend them as the final say in modeling these domains. Here, we are even more restrained: we intend this as a template for reasoning about movement building and do not endorse some of the key parameter settings currently in the model. We invite users to explore alternatives to these assumptions by constructing their own worldviews and setting their own project estimates.

If you would like assistance in setting up a custom Parliament application for your organization or to make allocation decisions, please contact us.

Acknowledgments

The Moral Parliament Tool is a project of the Worldview Investigation Team at Rethink Priorities. Arvo Muñoz Morán and Derek Shiller developed the tool; Hayley Clatterbuck created the particular parliaments in this sequence. We’d like to thank David Moss and Urszula Zarosa for helpful feedback. If you like our work, please consider subscribing to our newsletter. You can explore our completed public work here.

1. ^{^}

    There are also debates about how much EA should spend on movement building vs. direct work. We will not be taking up this question here (though a parliament could be designed that does so!).