Effective Altruism Forum
EA Forum

Hide table of contents
Comment Permalink
freedomandutility5
0
0

How does this differ to citizens assemblies?

Reply
ClayShentrup1
0
0
  1. it's compulsory, so you know it's statistically representative without having to use any complicated algorithms based on arbitrary criteria like Democrat or Republican or gay or straight (which aren't necessarily binary).
  2. this body isn't writing the laws, they are electing people. or they are voting on ballot measures.
Reply
See in context
by ClayShentrup
8 min read 10

14

AI safetyPolicyImproving institutional decision-making
Frontpage

In 2017, Aaron Hamlin wrote a compelling article arguing that voting methods were an open target for effective altruism. He was right. As an original co-founder of the Center for Election Science, I've spent two decades working to advance better voting methods like approval voting. The movement has had remarkable success, with adoption in several cities and growing awareness of alternative voting methods.

But in the past year, I've shifted my focus to what I believe is an even more neglected and impactful reform: Election by Jury (EBJ). Through electionbyjury.org, I'm working to advance this transformative approach to democratic decision-making. While voting method reform remains crucial, I've come to realize that no other democratic reform can match EBJ's potential impact-to-cost ratio, particularly when we consider existential risks facing humanity.

The Existential Stakes

We live in an era of unprecedented technological advancement, bringing both extraordinary opportunities and catastrophic risks. Advanced artificial intelligence could either solve humanity's greatest challenges or lead to our extinction. Synthetic biology could cure diseases or create engineered pandemics. Our response to climate change could determine the habitability of Earth for centuries.

These aren't just policy challenges – they're tests of our collective decision-making capacity. And we're failing these tests spectacularly:

  • Only 3% of Americans rank AI risk as a critical threat, despite warnings from leading AI researchers
  • 41% of voters oppose increased pandemic preparedness funding
  • Carbon pricing repeatedly fails at the ballot box, even in environmentally conscious states

The problem isn't just voter ignorance – it's structural impossibility.

The Mathematical Impossibility of Mass Democracy

Here's the stark reality: it's practically impossible to educate society at scale. Consider the math:

If it takes 20 hours of careful expert presentations and deliberation to truly understand a complex policy issue (a conservative estimate for topics like AI governance or pandemic preparedness), that's 4 billion person-hours for a nation of 200 million voters. For comparison, the entire Manhattan Project required about 24 million person-hours.

Even if we could somehow create this time, we face three insurmountable barriers:

  1. Attention Economics: In an attention economy dominated by social media algorithms optimized for engagement, how do we compete with engineered dopamine triggers?
  2. Epistemic Bubbles: You can't force vaccine skeptics to sit through immunology lectures or make climate change deniers engage with climate models they've been primed to distrust.
  3. Adversarial Forces: Concentrated wealth and hostile state actors have enormous resources and incentives to spread misinformation. AI-powered microtargeting will only amplify this problem.

The Statistical Solution

This is where the power of random sampling enters the picture. A mathematically remarkable fact: you only need about 40 randomly selected people to achieve statistically reliable decisions that match what the entire population would choose if they went through the same deliberative process. The math behind this is rigorous and well-established – you can find a detailed analysis here.

This isn't about estimating exact percentages - it's about the probability that the jury makes the same choice the whole population would make under identical conditions. With just 40 people, we get remarkably high confidence that the jury's decision matches what would have happened if everyone had participated in the same thorough deliberation process.

The main reason to increase jury size beyond these basic statistical requirements is to increase resistance to tampering through bribery or coercion. Even here, the math works in our favor – with secret ballots and reasonable jury sizes (200-400 people), the cost of reliably bribing enough jurors becomes prohibitive.

Philosophical Foundations

EBJ addresses a fundamental tension in democratic theory between epistocracy (rule by the knowledgeable) and democracy (rule by the people). Pure epistocracy risks capture by elite interests, while pure democracy risks catastrophic decisions from an uninformed public.

EBJ offers a unique synthesis:

  • Democratic legitimacy through random selection (pure lottery, no qualification tests)
  • Epistocratic benefits through structured deliberation
  • Protection against capture through:
    • Random selection (can't buy your way in)
    • Secret ballots (can't verify bribes)
    • Large enough juries to make tampering prohibitively expensive

Expected Value Analysis

Let's consider the expected value of EBJ implementation:

Costs:

  • Research & polling: $25,000-100,000
  • Pilot programs: $100,000-500,000
  • First city implementation: $1-2M
  • Scaling to major cities: $5-10M per city

Benefits (conservative estimates):

  • 10% improvement in policy decisions
  • Applied to city budgets ($100M-1B+)
  • Yearly benefit: $10M-100M per city
  • Net Present Value (30 years, 5% discount rate): $150M-1.5B per city

Even with extremely conservative assumptions about decision quality improvement, the return on investment is extraordinary. This doesn't account for existential risk reduction, which could multiply these benefits by orders of magnitude.

Addressing Key Objections

When presenting Election by Jury to policymakers and the public, several important concerns typically arise. The most common is that people simply won't accept such a dramatic change to our democratic process. But we already have compelling evidence to the contrary from Oregon's Citizens' Initiative Review (CIR), a groundbreaking program developed by Healthy Democracy in 2009.

The Oregon CIR demonstrates the public's willingness to trust their fellow citizens' judgment when given the opportunity to deeply study an issue. Under this program, randomly selected panels of Oregon voters spend multiple days studying ballot measures, hearing from experts and advocates on all sides, and producing clear, factual statements for the voters' pamphlet. These citizen panels have consistently produced high-quality analyses that voters find helpful and trustworthy.

The program remains on the books today and has been extensively studied by academics, who have documented its effectiveness. The only thing preventing its continued operation is funding – it requires a modest $100,000 to $150,000 per election cycle to run. This relatively small cost for demonstrated benefits suggests that more ambitious reforms could also be surprisingly cost-effective.

Critics sometimes argue that Election by Jury is simply too radical a departure from current practice. But it's worth remembering that we already trust randomly selected citizens with matters of life and death in criminal trials. Moreover, what we consider "normal" democracy today would have seemed radical just a few centuries ago. The secret ballot, universal suffrage, and direct election of senators were all once considered dangerous innovations.

A more sophisticated objection concerns capture by special interests. Couldn't wealthy individuals or organizations simply bribe or coerce enough jurors to swing the outcome? This is where the mathematics of random selection and jury size becomes crucial. With secret ballots and sufficiently large juries (200-400 people), the cost and risk of trying to influence enough jurors becomes prohibitively high. Multiple layers of protection – random selection, secret ballots, and careful size requirements – work together to create a remarkably tamper-resistant system.

Finally, some worry about the quality of deliberation. Can ordinary citizens really understand complex policy issues? The evidence from citizens' assemblies and Oregon's CIR is clear: when given time, resources, and a structured environment for learning, citizens are remarkably capable of understanding and making informed decisions about complex issues. The key is creating conditions that support genuine learning and reflection – exactly what Election by Jury is designed to do.

How Would It Work? Implementation Pathways

Election by Jury isn't an all-or-nothing proposition. There are multiple ways to implement it, from gentle advisory roles to fully binding decisions. Let's explore the two main applications: ballot initiatives and candidate elections.

For Ballot Initiatives: Building on Proven Success

The Oregon Citizens' Initiative Review (CIR) provides a proven model for ballot measure deliberation. Established in 2011 through the work of Healthy Democracy, the program randomly selects panels of voters to study ballot measures and provide recommendations to their fellow citizens.

The CIR remains law in Oregon today and requires only modest funding - about $100,000 to $150,000 per election cycle - to operate. We could build on this success in several ways:

  1. Advisory Role with Enhanced Visibility
    • Place jury recommendations directly on the ballot itself, not just in the voter guide
    • This "nudge" approach preserves voter autonomy while maximizing the influence of deliberative judgment
    • Could be implemented without changing existing voting rights
  2. Hybrid Binding/Advisory System
    • Require measures to pass both the public vote AND a citizen jury
    • This provides a check against misleading campaign tactics while maintaining broad participation
  3. Fully Binding Decisions
    • Have citizen juries make final decisions on ballot measures
    • Could start with specific types of measures (e.g., technical policy changes) before expanding

For Candidate Elections: A New Approach

Electing candidates through citizen juries would work differently from ballot measures, using a structured deliberative process:

  1. Advisory Implementation
    • Citizen juries interview candidates and make public recommendations
    • Their detailed findings appear in voter guides and on ballots
    • Voters retain final decision but benefit from thorough jury evaluation
  2. Hybrid Systems
    • Add new "jury-elected seats" to existing councils or legislatures
    • For example, a 9-member city council could add 2 jury-elected seats
    • This creates a controlled experiment in jury selection while maintaining voter-elected majority
    • Allows direct comparison of the two selection methods
  3. Full Implementation
    • Citizen juries make binding decisions for specific offices
    • Could start with offices requiring technical expertise
    • Uses the structured time-bank system for candidate presentations

The hybrid approach is particularly promising as a stepping stone. By adding jury-elected seats while preserving existing voter-elected positions, we can:

  • Demonstrate the effectiveness of jury selection
  • Build public confidence gradually
  • Allow direct comparison of decision-making quality
  • Maintain democratic legitimacy during transition

Each of these approaches can be tested at various scales - from small municipalities to state-level implementation. The U.S. system of local control and ballot initiatives provides an ideal laboratory for proving these concepts incrementally.

A Call to Action

The effective altruism community has demonstrated remarkable wisdom in identifying neglected opportunities for massive impact. Election by Jury represents exactly such an opportunity – a chance to fundamentally improve humanity's decision-making capacity at a time when good decisions have never been more crucial.

This isn't just another reform – it's potentially the most important reform possible. The challenges humanity faces in the coming decades will require unprecedented wisdom in our collective decisions. EBJ offers a mathematically rigorous, philosophically sound, and practically implementable path to achieving that wisdom.

The question now is: Are you ready to help create a democracy that can tackle humanity's greatest challenges? Visit electionbyjury.org to learn more and get involved. The future is watching.

Extending the Power of Secret Ballots

While the core EBJ proposal focuses on jury selection and deliberation, there's an intriguing possibility for further reducing corruption: extending secret ballots to elected officials themselves. Just as we protect jury independence through secret voting, we could require officials elected by EBJ to cast their legislative votes in secret.

This would make quid pro quo schemes practically impossible – how can you verify someone voted as promised if their vote is truly secret? Combined with EBJ's resistance to initial electoral corruption, this could create a system remarkably resistant to undue influence.

This extension remains speculative and would require careful consideration. But it demonstrates how the principles that make EBJ powerful – statistical representation, mandatory participation, and secret balloting – could potentially be extended to create even more robust democratic institutions.

Clay Shentrup was an original co-founder of the Center for Election Science and is a long-time advocate for democratic reform.

14

2
2

Reactions

2
2

More posts like this

Comments10


Sorted by
New & upvoted
Click to highlight new comments since:
freedomandutility5
0
0

How does this differ to citizens assemblies?

Reply
ClayShentrup1
0
0
  1. it's compulsory, so you know it's statistically representative without having to use any complicated algorithms based on arbitrary criteria like Democrat or Republican or gay or straight (which aren't necessarily binary).
  2. this body isn't writing the laws, they are electing people. or they are voting on ballot measures.
Reply
Matt_Sharp3
0
0

Benefits (conservative estimates):

  • 10% improvement in policy decisions
  • Applied to city budgets ($100M-1B+)
  • Yearly benefit: $10M-100M per city
  • Net Present Value (30 years, 5% discount rate): $150M-1.5B per city


I upvoted because I think you're touching on some interesting ideas. But I think you have a lot to do to demonstrate the scale of benefits you describe - if you have a more detailed analysis, I'd encourage you to link to it in the above section.

In particular: 

  • What evidence there is for (at least) a 10% improvement in policy decisions? I can see how the process would be a substantial improvement over a 'pure democracy' (as you describe). But what you describe sounds very similar to what policy-focused civil servants should already be doing (at least in the UK, for national policy), in terms of assessing evidence and listening to various experts and advocates. Perhaps there is a gap at city-level?
  • You appear to have taken the '10% improvement', and then multiplied it by the annual city budget in order to get the yearly benefit, and then multiplied this by 30 (with a 5% discount rate) to get the benefit over 30 years. This makes sense if '10% improvement' literally means 10% improvement in outcomes (rather than 'reallocating 10% of the budget').
    • But if this does mean 10% improvement in outcomes, it seems extremely implausible to me that such improvements can reoccur year-on-year. If a city's budget is horribly misallocated at the start, then there could be substantial improvements in the first few years - but you'll pretty quickly get to the point where there are relatively minor differences in the marginal cost-effectiveness of different activities.
      • [edit: I think this is wrong - a specific 10% improvement made in year 1 could of course reoccur in subsequent years. But there is an assumption that this 10% improvement wouldn't have been made at any other time without Election by Jury]
        • Could you provide some real (or even theoretical) examples to support your claims?
  • You also mention x-risk. But even if what you are proposing is successful, I imagine it would take at least several decades to become widespread. Given the costs you indicate per city, I imagine it would be easier and much faster to try to influence existing politicians and civil servants?
Reply
John Huang*1
0
0

The advantage of "reform" vs "lobby" is a potential permanent change in 10% improvement year-on-year. If the decision making is actually superior, then we can expect repeated improvements in decision making and budgeting for all subsequent years. 

>I imagine it would take at least several decades to become widespread

Comparing to the pace of change with regards to any world problems, decades-long timespans, yes ridiculously long, are about on-par with many political battles. How long did it take for example to decriminalize marijuana? After 60 years, the fight is ongoing. How long did it take to eliminate lead from gasoline? Leaded gasolines started being banned in 1925, yet it wasn't fully banned until the 1970s to 1990s in the US. 

The fact that needed reforms have a 60+ year turnaround is an indictment on the incompetence of the status quo in my opinion. If we care about long term planning, we need something more performant. 

Let's imagine a hypothetical new and improved decision making process can reduce the turnaround time from 60 years to only 10 years. What's the cost-benefit of for example, having unleaded gasoline 50 years sooner? 

Reply
ClayShentrup1
0
0

great points John!

Reply
ClayShentrup1
0
0

these were ballparky estimates created by claude. to me, it seems obvious this is the biggest issue for humanity, because it affects every single other policy issue we care about. as i point out, you can't educate people at scale. but you can absolutely do it with a small statistically representative sample. so no matter what public policy you care about, this is the #1 issue with a bullet.

of course we want to do more to give this the kind of "objective" impact analysis we get via e.g. voter satisfaction efficiency metrics with voting methods. that would require a pretty substantial research budget and involve a massive amount of ballparky estimation. my point here is just to lay out the case at a high level. i've worked in electoral reform and "human welfare optimization" and economics for 20 years, and it seems so obvious to me that this is the solution, that i'm merely trying to pose the idea and get more people thinking about it. if someone thinks there's any other reform that can come close to competing with this for impact, i'd be floored.

Reply
Arturo Macias1
0
0

Sortision is not used even for expert panels (including high courts), where objections are truly inexistent.

https://forum.effectivealtruism.org/posts/PyqPr4z76Z8xGZL22/sortition

Reply
ClayShentrup1
0
0

I'm not sure what point you're making.

Reply
Arturo Macias1
0
0

Oh, it is not an objection! I mean, you have given some arguments that go against the use of citizen juries for general policy issues. Still you think juries are good for that use case.

But for expert panels, there are not substantial arguments against choice by sortision!. Still the US President (and often european parliaments) chose supreme justices. I am quite pessimistic because this mechanism is not used even when its superiority is almost impossible to dispute.

Reply
ClayShentrup1
0
0

you could have said the same thing about approval voting, but then we got it passed in Fargo by a 64% supermajority.

Reply
Curated and popular this week
LintzA
 ·  · 15m read
 · 
Cross-posted to Lesswrong Introduction Several developments over the past few months should cause you to re-evaluate what you are doing. These include: 1. Updates toward short timelines 2. The Trump presidency 3. The o1 (inference-time compute scaling) paradigm 4. Deepseek 5. Stargate/AI datacenter spending 6. Increased internal deployment 7. Absence of AI x-risk/safety considerations in mainstream AI discourse Taken together, these are enough to render many existing AI governance strategies obsolete (and probably some technical safety strategies too). There's a good chance we're entering crunch time and that should absolutely affect your theory of change and what you plan to work on. In this piece I try to give a quick summary of these developments and think through the broader implications these have for AI safety. At the end of the piece I give some quick initial thoughts on how these developments affect what safety-concerned folks should be prioritizing. These are early days and I expect many of my takes will shift, look forward to discussing in the comments!  Implications of recent developments Updates toward short timelines There’s general agreement that timelines are likely to be far shorter than most expected. Both Sam Altman and Dario Amodei have recently said they expect AGI within the next 3 years. Anecdotally, nearly everyone I know or have heard of who was expecting longer timelines has updated significantly toward short timelines (<5 years). E.g. Ajeya’s median estimate is that 99% of fully-remote jobs will be automatable in roughly 6-8 years, 5+ years earlier than her 2023 estimate. On a quick look, prediction markets seem to have shifted to short timelines (e.g. Metaculus[1] & Manifold appear to have roughly 2030 median timelines to AGI, though haven’t moved dramatically in recent months). We’ve consistently seen performance on benchmarks far exceed what most predicted. Most recently, Epoch was surprised to see OpenAI’s o3 model achi
Dr Kassim
 ·  · 4m read
 · 
Hey everyone, I’ve been going through the EA Introductory Program, and I have to admit some of these ideas make sense, but others leave me with more questions than answers. I’m trying to wrap my head around certain core EA principles, and the more I think about them, the more I wonder: Am I misunderstanding, or are there blind spots in EA’s approach? I’d really love to hear what others think. Maybe you can help me clarify some of my doubts. Or maybe you share the same reservations? Let’s talk. Cause Prioritization. Does It Ignore Political and Social Reality? EA focuses on doing the most good per dollar, which makes sense in theory. But does it hold up when you apply it to real world contexts especially in countries like Uganda? Take malaria prevention. It’s a top EA cause because it’s highly cost effective $5,000 can save a life through bed nets (GiveWell, 2023). But what happens when government corruption or instability disrupts these programs? The Global Fund scandal in Uganda saw $1.6 million in malaria aid mismanaged (Global Fund Audit Report, 2016). If money isn’t reaching the people it’s meant to help, is it really the best use of resources? And what about leadership changes? Policies shift unpredictably here. A national animal welfare initiative I supported lost momentum when political priorities changed. How does EA factor in these uncertainties when prioritizing causes? It feels like EA assumes a stable world where money always achieves the intended impact. But what if that’s not the world we live in? Long termism. A Luxury When the Present Is in Crisis? I get why long termists argue that future people matter. But should we really prioritize them over people suffering today? Long termism tells us that existential risks like AI could wipe out trillions of future lives. But in Uganda, we’re losing lives now—1,500+ die from rabies annually (WHO, 2021), and 41% of children suffer from stunting due to malnutrition (UNICEF, 2022). These are preventable d
Rory Fenton
 ·  · 6m read
 · 
Cross-posted from my blog. Contrary to my carefully crafted brand as a weak nerd, I go to a local CrossFit gym a few times a week. Every year, the gym raises funds for a scholarship for teens from lower-income families to attend their summer camp program. I don’t know how many Crossfit-interested low-income teens there are in my small town, but I’ll guess there are perhaps 2 of them who would benefit from the scholarship. After all, CrossFit is pretty niche, and the town is small. Helping youngsters get swole in the Pacific Northwest is not exactly as cost-effective as preventing malaria in Malawi. But I notice I feel drawn to supporting the scholarship anyway. Every time it pops in my head I think, “My money could fully solve this problem”. The camp only costs a few hundred dollars per kid and if there are just 2 kids who need support, I could give $500 and there would no longer be teenagers in my town who want to go to a CrossFit summer camp but can’t. Thanks to me, the hero, this problem would be entirely solved. 100%. That is not how most nonprofit work feels to me. You are only ever making small dents in important problems I want to work on big problems. Global poverty. Malaria. Everyone not suddenly dying. But if I’m honest, what I really want is to solve those problems. Me, personally, solve them. This is a continued source of frustration and sadness because I absolutely cannot solve those problems. Consider what else my $500 CrossFit scholarship might do: * I want to save lives, and USAID suddenly stops giving $7 billion a year to PEPFAR. So I give $500 to the Rapid Response Fund. My donation solves 0.000001% of the problem and I feel like I have failed. * I want to solve climate change, and getting to net zero will require stopping or removing emissions of 1,500 billion tons of carbon dioxide. I give $500 to a policy nonprofit that reduces emissions, in expectation, by 50 tons. My donation solves 0.000000003% of the problem and I feel like I have f
Recent opportunities in AI safety
4
Last week to apply for the Futurekind AI Fellowship! (deadline: April 1)
Jay Luong
· · 1m read
0
0
20
Hiring pre-docs
Eva
· · 1m read
0
0
57
Quick nudge to apply to the LTFF grant round (closing on Saturday)
calebp
· · 1m read
7
7
View more