The original 16 questions now have 45–66 forecasters each (mean 51). Five added on September 20 have 15–23 each (mean 18), up from 11–15 in the first check. The forecasts still point to a gap between milestones and changes in production or welfare at scale.
Feedback welcome: Use the Hypothes.is sidebar at the top right to annotate this dashboard. On the short report, use the sidebar there to comment on that page. Corrections, missing sources, and suggestions are especially useful.
The first wave averages 51 forecasters per question (range 45–66). The five questions added on September 20 average 18 (range 15–23) after 13 days, up from 11–15 after their first three days. These are question-level counts, not unique participants or forecast revisions.
Question counts make coverage easier to compare. They count each account on every question it forecasts; the tournament leaderboard is a separate measure and includes bot accounts. Check current counts and standings.
| Tournament | Forecasters per displayed question | What limits the comparison |
|---|---|---|
| Cultured Meat | 23–31 across its three questions | Older, much smaller, no prize pool |
| Climate Tipping Points | 14–62 among the first 19 shown | Started in 2023; broader subject |
| Global Health Monitor 2026/27 | 8–17 across ten questions | Newer, no prize pool |
| AI Pathways | 58–143 among 19 questions | Older and draws a larger AI forecasting audience |
These are context, not a ranking. Ages, topics, question types, and prizes differ. Counts were read from public question cards on October 3.
These are current crowd estimates for precisely worded questions. One outcome has resolved; the rest are judgments that can change. Each claim links to its question and resolution rule.
A new-product approval question resolved Yes on August 4 after Aleph Farms’ Singapore approval was identified. This is a regulatory milestone. GFI says actual availability remains limited by geography and production volume.
The crowd gives 2% to plant-based meat and seafood reaching 10% of US retail packaged-meat dollar sales before 2031, down from 3% on September 23. It gives 2% to approval of Nueva Pescanova’s Canary Islands octopus farm before June 2027.
On the question’s stated definition, forecasters put about 83% combined probability on at least one further US state cultivated-meat ban before July 2027. They give 3% to the Save Our Bacon Act becoming law in that period.
The crowd estimates 42.9% of EU commercial layers and 3.06% of the US hen flock will be sexed in ovo in Q1 2027. The question definitions and data sources matter for this comparison; see the discussion below.
The crowd estimates 12.7% cage-free and 5.24% broiler commitment fulfillment under the questions’ Chicken Watch definitions. These questions now have 17 and 15 forecasters.
The crowd estimates 10.5 of Europe’s top 20 grocery retailers will have public shrimp welfare commitments before July 2027. A separate question estimates 8.26 billion shrimp per year covered by SWP Humane Slaughter Initiative agreements.
In the 2034 AI-lab guidelines question, two labs remains the most likely answer (22.4%). The crowd spreads probability across several outcomes. This is a long-horizon indicator rather than a near-term welfare outcome.
There are 32 comments across the 21 question cards; nine questions have none. Recent comments surfaced a timing inconsistency in a new question as well as issues that sharpen resolution criteria.
A forecaster flagged Louisiana’s signed “trigger law”, which depends on a future Supreme Court ruling. Metaculus clarified that it counts only once the condition is met. A signed bill need not be an operative restriction.
A commenter found a revised historical EU estimate and a difference between commercial-layer and broader measures. The criteria were clarified to use commercial layers and the first 2027 report.
On the new cage-free law question, a commenter flagged conflicting cutoff dates. A follow-up noted the mismatch, and a later reply said the fine print had been changed. It now agrees with the July 1, 2027 cutoff in the title and criteria.
Forecasters pressed for a precise approval definition before the cultivated-meat question resolved. The eventual Yes is a specific product and jurisdiction milestone, not evidence of broad market access.
On the 2040 aquaculture question, a participant flagged an older OECD-FAO background dataset and linked newer figures. Check this baseline before using the forecast in a funding or policy brief.
These are proposals for the organizers and participants, based on this snapshot. They are open for comment.
The tournament rules say three longer-horizon questions are excluded from leaderboard calculations, while prize eligibility requires active forecasts on at least two of them for most of the tournament. Explain this prominently when recruiting participants.
Open the relevant Metaculus question to add a prediction, evidence, or a resolution-rule comment. That keeps the forecasting discussion with the forecasters.
Highlight the relevant passage here and use the Hypothes.is sidebar to add a public annotation. Especially useful: a correction, a missing source, a funding or policy decision this could affect, or a question we should cover next. How to annotate.
This page is reviewed every ten days. The RSS feed records updates; the short report is the link to share when a full dashboard would be too much.
Snapshot taken October 3, 2026 from the public Animal Futures tournament, its question cards and comments, the Metaculus announcement, and the linked comparator tournament pages. Forecaster counts are per question, so a person forecasting several questions appears several times. Means are simple averages across 16 original or five September 20 questions. Comments count the question-card comments, including replies; the announcement’s eight comments are separate.
One question had resolved at the check. All other figures are Metaculus community forecasts, not observed outcomes or independently validated probabilities. Three longer-horizon questions are excluded from leaderboard scoring. We did not establish how often individuals revised their forecasts or find a publicly documented funding or policy decision citing one.
The Unjournal landing page still says 16 questions. Metaculus now lists 21.