The problem
Animal welfare funders, including Animal Charity Evaluators, Open Philanthropy, and others, must decide how to allocate resources across different approaches to reducing animal suffering. Corporate campaigns for cage-free commitments are seen to have clear, measurable impactsAnimal Charity Evaluators rates corporate campaigns (e.g., cage-free and broiler welfare commitments) among the most cost-effective animal welfare interventions, with documented and measurable outcomes.. Cultivated meat development is more speculative: potentially transformative, particularly if it comes close to cost parity with conventional meat, but highly uncertain.
The core question is deceptively simple:*Why cost matters for funding decisions — but isn't sufficient. "CM can plausibly achieve near price parity (or a reasonable hybrid cultured meat/plant-based product that is close to taste equivalent)" is highly correlated with "funding CM is likely to have high animal welfare impact per dollar." If CM can never get close to cost parity, additional funding is almost certainly wasted. But cost parity is necessary, not sufficient: CM funding could still have low impact if (a) CM would succeed soon regardless of marginal funding, or (b) even cheap CM fails to displace conventional meat due to consumer or regulatory barriers. We discuss this in more detail in our EA Forum post and PQ definitions. What will cultivated meat cost to produce? If costs fall dramatically, CM could displace a substantial share of conventional meat production and spare billions of animals.**There may also be important environmental benefits, reduction of animal-to-human disease vectors, and other consequences — but we are primarily focused on the animal welfare implications for this workshop. If costs remain high, funding CM development may have been a poor use of limited animal welfare resources compared to proven interventions.
This is part of The Unjournal's Pivotal Questions initiative: working with impact-focused organizations to identify their highest-value research questions, connect them to evidence, and commission expert evaluations that can inform real decisions.
What sparked this workshop
We recently completed an evaluation of Rethink Priorities' forecasting work on cultured meat production. The evaluators noted that while the original forecasts (from 2022) were valuable, the field has evolved: new TEAs have been published, some claiming dramatically lower costs (e.g., Pasitka et al. 2024 at ~$13/kg), while others remain pessimistic. See our EA Forum post for further discussion.July 2025: "Is cultured meat commercially viable? Unjournal's first Pivotal Question" — background on the PQ project, focal questions, and research pipeline.
Through The Unjournal's Pivotal Questions evaluator process, we're commissioning expert assessments of key TEAs, including Pasitka et al. and comparative analyses like the Goodwin 2024 scoping review. The workshop brought together TEA authors, evaluators, and funders to synthesize what the evidence tells us and what it means for funding decisions.
What the workshop aimed to achieve
The workshop brought together TEA researchers, evaluators, and stakeholders to work through the specific technical and economic disagreements driving the cost debate. The discussion was organized around key cruxes:
1. Where do the TEAs actually disagree — and why?
Pasitka et al.Pasitka et al. (2024), "Empirical economic analysis shows cost-effective continuous manufacturing of cultivated chicken using animal-free medium," Nature Food. Claims ~$13.7/kg via continuous perfusion with animal-free medium. Note: Believer Meats (the authors' company) shut down in Dec 2025. The paper uses spontaneously immortalised cell lines; gene-edited alternatives could substantially alter these projections. projects ~$13/kg; HumbirdHumbird (2021), "Scale-up economics for cultured meat," Biotechnology and Bioengineering. Commissioned by Open Philanthropy. Concludes CM faces "extreme, intractable technical challenges at food scale." projects $38–51/kg. This large gap isn't random — it traces to specific assumptions about media costs, bioreactor design, and scale-up pathways. Which assumptions matter most? See component-by-component TEA comparison →
2. Media costs and growth factor technology: How cheap can they get?
Media is the largest variable cost. GFI's recent amino acid supply chain analysisGFI (Dec 2025), "Amino acid cost and supply chain analysis for cultivated meat." Based on real quotes from amino acid suppliers and CM manufacturers. Reports that Humbird's amino acid prices may have been overestimated by 2–10x. suggests Humbird may have overestimated amino acid prices by 2–10x. Will plant or yeast hydrolysates replace expensive purified amino acids in the base media (CM_12)? What are realistic media costs per kg of output by 2036 (CM_14)? How TEAs treat media costs →
3. Bioreactors & scale-up: What's the real cost of capacity?
Pharma-grade vs food-grade vs custom-built bioreactors (CM_20). Some companies claim custom bioreactors for under $1M. What cell densities are achievable at 20,000-liter scale (CM_16)? How does process choice (Fed-batchFed-batch: nutrients added incrementally during the run, but without continuous media exchange; cells harvested at the end. This differs from true "batch" culture (no media addition at all), which is not viable for CM because cells rapidly deplete nutrients and accumulate toxic waste products. Fed-batch is simpler to operate than perfusion but achieves lower cell densities — a key cost disadvantage for high-volume CM production. vs. perfusionPerfusion: continuous media exchange with cell retention (e.g. via a filter); cells harvested at end of run. Higher cell densities than fed-batch, used in Pasitka et al.'s optimistic projections. vs. continuous harvestingContinuous harvesting: both cells and spent media are continuously removed; fresh media replenishes the reactor in a steady-state operation. Highest potential productivity but operationally complex and less common in CM at present.) change the picture? Cell density & process assumptions across TEAs → · Capital costs →
4. From TEA to reality: bridging the gap
Optimistic cost projections don't always translate into commercial success — and there are many reasons a company might struggle beyond production costs alone. Some companiesSeveral cultivated meat companies have experienced commercial difficulties in recent years, including Believer Meats (late 2025). The reasons behind any specific case are rarely fully public and are typically complex — production costs, regulatory timing, market conditions, and funding environment can all play a role. These developments shouldn't be read as simple refutations of any particular TEA methodology. that made significant technical progress have nonetheless faced commercial difficulties. What can the gap between modeled costs and commercial realities tell us about what TEAs might be missing or underweighting? Cautious synthesis across TEAs →
5. Cell line technology: does gene editing change the picture?
Most publicly available TEA data — including Pasitka et al. and Believer MeatsBeliever Meats used spontaneously immortalised cells. Pasitka et al. described their spontaneous immortalisation protocol in avian and bovine cells. Gene-edited cell lines can grow without or with minimal growth factors, at high densities, in single-cell suspension — dramatically changing cost projections. See: Riquelme-Guzman et al. (2024), iScience. — is based on spontaneously immortalised cell lines, not gene-edited ones. Gene editing can generate cells that grow without expensive growth factors and at much higher densities, potentially eliminating the largest cost driver. Yet current models rarely model gene-editing scenarios explicitly. Gene editing is permitted in most jurisdictionsGene editing is legally permitted in the US, UK, and most non-EU jurisdictions. Upside Foods received US regulatory approval using gene-edited cell lines. The EU is more restrictive, but is not representative of the global regulatory environment. See: CRISPR regulations tracker. outside the EU. How should cost models account for gene-editing scenarios — and what is the realistic timeline for gene-edited lines to enter commercial production? Growth factor assumptions across TEAs →
These cruxes are interconnected — media costs depend on cell density, bioreactor choice affects media efficiency, and TEA methodology choices cut across all of them. Together they feed into The Unjournal's focal question CM_01: What will CM cost per edible kg in 2036?
Explore the Evidence
Explore the cost modeling and background resources that informed the workshop discussion:
📊
Cost Projection Dashboard
Interactive Monte Carlo simulation — adjust parameters and see how they affect cost projections across 30,000 runs.
📚
How Cultured Chicken Is Made
Cost-focused production overview with video explainers, key papers, and interactive calculation tools.
Technical documentation: model methodology & parameter reference →
Join the discussion
Quick correction or inline note? Use Hypothes.is — click the < tab on the right edge of any page. No GitHub account needed. The team monitors annotations and implements obvious fixes within hours.
Substantive question or methodology debate? Post in GitHub Discussions — the Substantive hub is the main event. Resolved discussions are committed back to the model and site.
For suggested reading and key papers, see our Resources page.
How the workshop was structured
The workshop was held fully online on May 8, 2026 — three sessions over a 4-hour window with breaks. S1 and the public portion of S3 were recorded; edited recordings and a transcript are being prepared. See the Workshop Summary for what was discussed. Asynchronous participation remains welcome — beliefs submissions and Hypothes.is annotations are still open.
See the full agenda — including session details, speaker information, and recording policy →
Session overview — click to expand
All times US Eastern (May 8, 2026). S1 and the public portion of S3 were recorded — edited recordings are being prepared. Async participation remains welcome. Full agenda → · Workshop Summary →
Recording: S1 and the public first portion of S3 were recorded and will be shared publicly (with transcript), similar to our recent wellbeing workshop. S2 was fully off-record — not recorded, and nothing shared beyond participants unless explicitly requested. Edited recordings are being prepared. See the full agenda for session details.
Outputs: A working workshop summary is now live, covering session discussions and pre-workshop written submissions. Expert beliefs are being collected via a post-workshop update round, and the full belief distribution will be published here once responses are in. Commissioned evaluations of key TEAs are being finalized through The Unjournal's PQ evaluator process. Belief responses are published publicly — contributors can choose to be named or remain anonymous.
Participants — RSVPs and Attendance
For suggested discussants matched to each agenda topic, see the Suggested discussants by topic table → on the agenda page.
Session discussion interest — from participant forms
Loading…
| Participant | Affiliation | Role |
|---|---|---|
| Elliot Swartz | Good Food Institute | TEA Author / Discussant |
| Claire Bomkamp | Good Food Institute | Industry perspective |
| Derrick Risner | UC Davis | TEA Author (Risner et al.) |
| Ellie Contreras | Tufts University | Researcher (tissue engineering) |
| Tom Bry-Chevalier ApologiesUnable to attend May 8 — in hospital (Victory in Europe Day complicates discharge). Has sent warm apologies and will watch the recordings and engage async. | Université de Lorraine | Researcher |
| Jakub Kozlowski | — | Cost model developer |
| David Manheim Pre-sessionParticipated in the May 6 informal pre-session. Not confirmed for the May 8 main workshop. | Technion / ALTER | Unjournal Evaluator |
| Oana Kubinyecz | Atova Regulatory Consulting | Regulatory Scientist / Cell Line Development |
| Valentin Klotzbücher | University of Basel / Unjournal | Clinical Research Statistician · UJ Management |
| Andrew Stout WithdrewUnable to attend May 8 due to a conflict. May engage asynchronously. | Tufts University / Kaplan Lab | Researcher (Beefy-R TEA co-author) |
| Natalie Rubio | Tufts CCA / Deco Labs | Researcher / Industry |
| Matt McNulty | Tufts Center for Cellular Agriculture | Strategy & Operations |
| Jordi Morales-Dalmau | Cultimate Foods | Industry R&D / Scale-up |
| Aleksandra Fuchs | ACIB (Austrian Centre of Industrial Biotechnology) | Researcher (bioprocess / media) |
| Nike Schiavo | Bruno Cell | Industry |
| Joana T. Rosa | S2AQUAcoLAB | Researcher |
| Mirjam Capuder PartialJoined the May 6 pre-session. May join May 8 for a short period. | University of Maribor | Researcher |
| Tarka Abraham | Ivy Farm Technologies | Industry |
| Stefano Lattanzi | Bruno Cell | Industry (CEO) |
| Zhuoran Du | University of New South Wales | Researcher / RA |
| Affif Grazette | Wageningen University | Researcher |
| Dexter Oelrichs | Altera Food Initiative | EA / Stakeholder |
| Anaiz Gul Fareed | Univ. of Naples Parthenope / Unjournal | Doctoral Researcher · UJ Team |
| Hannah McKay | Rethink Priorities | Animal welfare researcher |
| Breanna Duffy PartialPlanning to join May 8 for as much of the session as possible. | New Harvest (US) | Director of Responsible Research & Innovation |
| Bert Frohlich Late RSVPRegistered day-of. CM consultant since 2018; co-author of GFI cell growth modeling paper (Nov 2025). Available for all sessions. | Biopharm Designs | TEA Author / CM Consultant |
| David Reinstein | The Unjournal | Organizer |
The May 8 workshop has taken place. To contribute asynchronously — through the beliefs form, Hypothes.is annotations, or GitHub discussions — see the links above. To be notified of future workshops or suggest participants, get in touch.
Stay Involved
The workshop has taken place — here's how to continue engaging with the project:
PQ Evaluators
Compensated ($400–800) to assess TEAs, state calibrated beliefs on CM_01 and subquestions, and evaluate key papers. Still actively recruiting. Learn more →
Future Workshops
We aim to hold follow-up synchronous sessions as analysis progresses and additional questions emerge. Register interest → or get in touch to be notified.
Stakeholders & Funders
Animal welfare funders, policy analysts, and industry representatives. We want to help you identify your pivotal questions and refine your beliefs based on research and evidence. Get in touch →
Know someone who should be involved in future workshops? Suggest a participant or share this page.
The Key Questions
We've operationalized the cost debate into specific, quantifiable questions. The focal question:
CM_01 — Focal Question
What will be the average production cost (per edible kg) of cultured chicken meat in 2036, across all large-scale plants in the world?
Defined as (annualized capital + operating costs) ÷ annual kg output, in 2025 USD. Workshop beliefs form → · Forecast on Metaculus →Metaculus is a public prediction platform — forecasting there contributes to crowd-sourced wisdom on CM_01 and is visible to the broader community. Workshop participants have a dedicated compact beliefs form (link above) for use within this workshop.
Accounting boundary: The cost of cultured cell mass at bioreactor harvest — factory-gate, before texturisation, blending with plant-based ingredients, or packaging. Not the cost of a finished consumer product. The "edible kg" refers to the harvested cell biomass itself (wet weight), which is what enters any downstream processing.
Workshop cruxes map to specific subquestions that feed into CM_01 (click to expand)
CM_12: Will most CM be produced using hydrolysates (replacing purified amino acids in the base media) by 2036?
CM_14: What will cell media cost per kg of CM output in 2036?
CM_16: What cell density is achievable in a 20,000-liter bioreactor by 2036?
CM_20: What share of companies will build their own bioreactors by 2036?
Full questions and beliefs form for this workshop →
Public Metaculus forecasting (wisdom of the crowd)
These link to Metaculus, a public forecasting platform where the broader community can forecast alongside workshop participants. CM_01 on Metaculus → · Expert aggregation (CM_03) →
Other Pivotal Questions Workshops
About this page
Some content on these workshop pages was drafted with AI assistance based on our own source materials, notes, and research. We've reviewed it carefully, but errors or unclear passages may remain. If you spot anything incorrect or confusing, please let us know — ideally via a Hypothes.is annotation on the relevant passage, or by emailing contact@unjournal.org.