The collection
All ideas
Every proposal on the site, in one list. Open any card to read the full argument, leave a comment, or run a question against it.
68 ideas
"Legal Person"
Laws are urgently needed to prevent AI from gaining personhood or becoming a "legal person" like corporations or enterprises today
Global Concessions as Diplomatic Leverage: Bundling Solutions
The most intractable international problems become tractable when bundled together. Linking an AI safety treaty to a pandemic preparedness agreement to a trade deal creates a package where each party gets something they want more than what they give up. China wants trade access. The US wants AI safety commitments. Small nations want vaccine technology transfer. A grand bargain that trades all three simultaneously is achievable even when each individual agreement is politically impossible.
The Global Flourishing Summit: G20+ for Existential Risk
Convene an international summit for one or several annual 10-day summits focused on solving the world's biggest problems: Extinction risks from AI, Pandemics and global health, Factory farming, Making governments solve problems much more effectively, Global poverty, Cybersecurity, Climate change, Diplomacy and nuclear war, wild animal suffering, and more. Any organization that can get world leaders and organizations to go to such a summit can make this happen. For example, Anthropic could probably make this summit happen!
Sharing 'What We Owe the Future' With World Leaders
Will MacAskill's book 'What We Owe the Future' makes a compelling case, grounded in philosophy, history, and economics, that the most important thing any leader can do is ensure humanity has a long and flourishing future. A targeted campaign to ensure that every head of state, senior minister, and major religious leader receives a copy (translated into their language), along with a personal meeting to discuss its implications, could shift the policy agenda of governments worldwide.
The Longtermism Endowment: A Permanent Fund for the Future
A Longtermism Endowment, funded by a small tax on AI compute, fossil fuel extraction, or financial transactions, would provide permanent, independent funding for work on existential risk: AI safety research, pandemic preparedness, nuclear risk reduction, and effective governance. A 0.1% tax on AI compute globally would generate ~$1B/year. Over 30 years at 5% returns, this grows to $70B, enough to fund the entire global AI safety research field indefinitely.
AI-Driven Biological Engineering for Feed-to-Food Conversion
Use AlphaFold-style AI breakthroughs to accelerate development of organisms and enzymes that convert non-edible biomass into human food. Fungal fermentation with AI optimization, machine learning for enzyme engineering, and generative biology platforms could cut development time by 50-70%. Realistic timeline: proof-of-concept in 2-4 years, pilot-scale at $5-10/kg in 5-8 years, cost parity with conventional agriculture in 10-15 years.
Community Hub Infrastructure: Library++ Deployed Globally
Build integrated Community Access Hubs providing internet + computer workstations + water access + food distribution + training space + health services. Libraries are trusted neutral spaces, culturally familiar, and multi-generational. Target: every person globally within 20 minutes of a fully-equipped hub within 10 years. Cost model proven in post-conflict contexts (Rwanda, Liberia). Includes Starlink for immediate connectivity, solar backup power, and community oversight boards.
Employ the World's University Students in the Global Longtermism Economy
With open internet and a stable economy, the world's educated population represents an extraordinary talent pool for the world's most important problems. Connect them to: AI safety research (30-hour BlueDot Impact courses, then employment at Anthropic, DeepMind); biosecurity and pandemic preparedness; nuclear risk reduction; climate policy; cybersecurity. Remote work enables earning international wages ($40-80k USD) from anywhere, creating enormous positive economic feedback loops without requiring emigration.
Creating Change Through Art and Cinema
The most powerful tool for changing minds at scale isn't a white paper, it's a great story. A cinematic documentary about the global challenges facing humanity, told with the production quality of a major Hollywood release, could reach billions of people where policy papers reach thousands. Al Gore's 'An Inconvenient Truth' shifted climate policy globally. 'Seaspiracy' changed dietary behavior for millions. A 'Flourishing' documentary would cost $10-50M and could reach 500M+ people at $0.10 per person.
10x Economic Growth Through Scientific Innovation Pipelines
Phase 1 (Years 1-3): Deploy ALLFED resilient food systems, build pilot SCP facilities, establish community hub infrastructure. Phase 2 (Years 3-7): Scale SCP to replace 20-30% of imported protein, expand precision fermentation, employ displaced agricultural workers. Phase 3 (Years 7-15): Full transition to resilient food systems, 50,000+ working in longtermism fields, economic output dramatically increased. Infrastructure conversion requires only 1/3 the lead time and 1/5 the capital vs new construction.
Clean Water for All: The $50B Solution
2.2 billion people lack access to safe drinking water. Yet the technologies to solve this exist: ceramic water filters reduce waterborne disease by 40-50%; point-of-use chlorination eliminates most pathogens for pennies per household. The $50B Global Water Fund, 0.1% of G20 GDP for 5 years, would provide safe water to every person currently lacking it. Every $1 invested in clean water returns $5 in reduced healthcare costs, increased worker productivity, and reduced child mortality.
Desalination + Renewable Energy: Solving Water Scarcity Forever
The world has an unlimited supply of salt water. Solar-powered desalination now costs $0.50-1.00 per cubic meter, approaching the cost of groundwater extraction in water-stressed regions. As solar panel costs continue to fall (down 90% in the last decade), solar desalination will reach $0.20-0.30 per cubic meter by 2030. A G20 commitment to fund desalination infrastructure in the world's 20 most water-stressed nations would eliminate water scarcity within 15 years.
Agricultural Waste to Protein: Turning 1.3 Billion Tons of Waste into Food
1.3 billion tons of food is wasted globally every year. Oyster mushrooms can be grown on coffee grounds, cardboard, straw, and sawdust, all currently going to landfill. A Global Agricultural Waste-to-Protein Initiative would: map waste streams in each country; fund pilot facilities converting waste to mushroom or SCP protein; and establish supply chains connecting waste producers to protein manufacturers. The result: 10-20 million tonnes of additional protein annually from currently wasted resources.
ALLFED Resilient Food Systems: Feeding the World After a Catastrophe
ALLFED has documented food production systems that can feed the world with minimal sunlight, relevant to nuclear winter and volcanic winter scenarios. These include: single-cell protein from methane ($3-5/kg), leaf protein concentrate from agricultural waste, cellulosic sugar from biomass, and rapid mushroom cultivation. Building these systems as non-emergency food production also displaces factory farming, they use far less land and water than conventional animal agriculture.
Far-UV Light: The Pandemic-Proof Building
Far-UV light at 222nm destroys airborne viruses and bacteria in seconds without harming human skin or eyes. Deploying far-UV fixtures in hospitals, airports, schools, and public transit globally would dramatically reduce the spread of every future respiratory pathogen, including ones we haven't discovered yet. Cost per fixture: ~$500. To cover every hospital room in the US: ~$2B. Global deployment to high-risk settings: ~$50B. Compare to COVID-19's cost: $16 trillion in economic damage. This is a 320x return on investment.
Library++: The Community Hub That Changes Everything
Take an existing building, a library, an old movie theater, a community center, and transform it into a Library++ hub: fast internet + computer workstations with pre-loaded courses (BlueDot Impact, Khan Academy, Coursera) + clean water dispensers + healthy food that actually tastes great + a nurse or doctor two days a week + overnight sleeping for those who need it. India has 60,000+ public libraries. Cost per hub: ~$50,000 to set up, ~$15,000/year to run. Cost per person-reached: under $50/year.
AI Biosecurity: Preventing Engineered Pandemics
The same AI tools that could accelerate drug discovery could help a bad actor design a pandemic pathogen. Kevin Esvelt at MIT has shown that current AI models can already provide meaningful uplift to bioweapon developers. The solution: a mandatory screening system for all DNA synthesis orders, AI models fine-tuned to refuse dangerous biosecurity requests, and an international treaty requiring biosecurity screening on gene synthesis. Cost: ~$500M to build the screening infrastructure globally.
AI Alignment Research Moonshot, $10B Fund
AI alignment research, making AI systems reliably pursue human-intended goals, is catastrophically underfunded relative to AI capabilities research. Current global alignment research budget: ~$100M/year. Current AI capabilities investment: ~$300B/year. A G20-funded $10B alignment research fund, distributed competitively to academic labs and nonprofits worldwide, would transform the field. Specific research priorities: interpretability, scalable oversight, and formal verification. This is likely the highest-return public investment in human history.
Quantum-Safe Cryptography: Protecting the Internet Before 2030
A sufficiently powerful quantum computer will break most current encryption in seconds, including the encryption protecting banking, healthcare, military communications, and the entire internet. NIST finalized post-quantum cryptography standards in 2024. A crypto-agility requirement for all federal systems, combined with an industry mandate for financial and healthcare sectors, would protect 80% of sensitive data at risk. Cost: ~$10B in the US, $50B globally.
Open Courseware for the World: Every Course Free Online
MIT's OpenCourseWare has put 2,400 courses online for free since 2002. Khan Academy provides K-12 education to 155 million students. The bottleneck: internet access and a place to study. The Library++ model solves this, pre-loading courses on local servers so they're available even with limited connectivity. A G20 investment of $1B/year to expand open courseware, increase offline availability, and pay for translation into the 100 most-spoken languages would make world-class education available to every human being within 5 years.
Cultivated Meat: From Lab to Supermarket in 5 Years
Cultivated meat (real animal cells grown without slaughter) has reached price parity with premium beef in pilot production. Current cost: ~$17/kg for cultivated chicken. Target: <$5/kg by 2027, price parity with conventional by 2030. What's needed: $2B in government investment in bioreactor R&D, streamlined FDA/USDA approval pathways, and government procurement (military, school cafeterias) to create guaranteed demand that de-risks private investment.
Universal Pre-K: The Highest-Return Investment in Human Capital
Nobel economist James Heckman has shown that investments in early childhood education return $7-12 per dollar invested, the highest return of any social program. Brain development in the first 5 years determines 90% of lifetime learning capacity. Universal pre-K (ages 3-5) for all children would: close educational achievement gaps; reduce crime by 20-30% among affected cohorts; increase lifetime earnings by $50K+ per child; and pay for itself within 20 years in reduced social program costs.
Digital Jobs Pipeline: Training the Global South for $50K/Year Remote Work
The highest-leverage anti-poverty intervention may be connecting skilled workers in low-income countries to high-paying remote jobs. BlueDot Impact's 30-hour AI safety course leads to employment at Anthropic and DeepMind, positions paying $100K+. ALX Africa trains software engineers in 6 months; graduates earn $30-50K/year while living in Africa. The Library++ model makes this accessible to anyone within 20 minutes of a hub.
Universal High-Speed Internet: The $150B Infrastructure Plan
3.5 billion people lack reliable internet access. A global fund, $150B over 10 years, split between G20 nations, the World Bank, and tech companies receiving preferential spectrum access, would provide universal basic internet at no cost to users in low-income countries. A 1% increase in internet penetration in developing countries raises GDP growth by 0.2 percentage points. This fund would generate $10 in economic value for every $1 invested.
Scientific Grand Challenges: The Apollo Program for Global Problems
The US government's investment in the Apollo program generated an estimated $7-14 return per $1 spent in spinoff technologies. A modern equivalent: a $100B/decade Grand Challenges Fund targeting the 10 biggest solvable problems. Each challenge has a defined, measurable goal and a prize for the first team to achieve it. X Prize has demonstrated this model works for space travel and ocean exploration.
The Citizen Idea Submission App: Democracy That Actually Works
Imagine: any citizen can submit a policy idea through a government app. AI reads all submissions, identifies the highest-quality ideas, and surfaces the best ones to relevant policymakers. Policymakers are required to respond publicly within 30 days. Taiwan's vTaiwan platform crowdsourced legislation on Uber and telemedicine, producing laws that passed with 98% consensus. Finland's citizen initiative system requires parliament to debate any proposal with 50,000 signatures.
The Seasoning Package: Making Alternative Protein Delicious
The single biggest barrier to alternative protein adoption isn't cost, it's taste. The solution: a standardized seasoning package. MSG (one pinch per dish, pennies), dried chili (one plant seasons 350 meals), cumin, garlic powder, soy sauce. These transform bland mushroom mycoprotein or single-cell protein into food people actively choose to eat. The science: MSG activates umami receptors in a way that makes protein taste meaty. At <$0.01 per serving, it's the highest-leverage culinary intervention possible.
Cybersecurity Workforce Development: Closing the 3.5M Job Gap
There are currently 3.5 million unfilled cybersecurity jobs globally. A Library++ model for cybersecurity training: free, high-quality online courses pre-loaded on Library++ computers; government-funded cybersecurity apprenticeship programs at critical infrastructure operators; and fast-track immigration visas for cybersecurity professionals. Countries that implement this program can fill their cybersecurity workforce gap within 5 years.
mRNA Vaccine Platform, 100-Day Vaccine for Any Pathogen
The COVID mRNA vaccines demonstrated we can design a vaccine in 48 hours once we have a pathogen's genome. The bottleneck is manufacturing and regulatory approval. The solution: pre-approve mRNA vaccine manufacturing facilities globally, establish regulatory fast-track protocols that activate automatically when WHO declares a Public Health Emergency, and stockpile lipid nanoparticle delivery components. CEPI's 100-Days Mission needs $3.5B in additional funding.
Mandatory Pre-Deployment Safety Testing for All Frontier AI
Require any AI system trained above a compute threshold (e.g. 10^26 FLOPs) to pass a battery of safety evaluations, including dangerous capability evals, deceptive alignment testing, and corrigibility checks, before deployment. California's SB 1047 was the first serious attempt. The path forward: a federal law in the US, a binding EU regulation, and a voluntary treaty among major AI labs (Anthropic, Google DeepMind, OpenAI, xAI) to maintain standards even if governments don't act. Labs that sign get preferential regulatory treatment.
International Pandemic Prevention Treaty with Real Enforcement
The WHO has no enforcement power. Countries can ignore pandemic reporting requirements with no consequences. A new International Pandemic Prevention Treaty would: require member nations to report novel outbreaks within 24 hours; grant international inspectors access to outbreak sites; impose automatic trade and travel restrictions on non-compliant nations; and fund a global stockpile of PPE, vaccines, and antivirals. The enforcement mechanism: nations that violate lose access to the shared vaccine stockpile.
Open Source Medicines: Breaking the Patent Barrier to Global Health
Most diseases killing people in low-income countries have effective treatments, they're just unaffordable. A Global Health Patent Pool, modeled on the Medicines Patent Pool (which already exists for HIV, TB, and Hepatitis C), would allow generic manufacturers to produce essential medicines without royalties. Expanding this to include cancer drugs, diabetes medications, and cardiovascular drugs would save an estimated 2-3 million lives per year.
Open-Source Government: The Estonia Model for Every Nation
Estonia built its entire digital government, e-voting, e-health, e-tax, e-police, digital ID, in 10 years after independence. Every Estonian can file taxes in 3 minutes, vote from anywhere in the world, and access their full medical history with one click. Estonia open-sourced its X-Road data exchange platform, which is now used by Finland, Namibia, and others. A Global Digital Government Initiative would give every nation an Estonia-quality digital government within 20 years.
Firewalls 2.0: AI-Powered Network Defense for Critical Infrastructure
The next major cyberattack on a power grid or water system will not be stopped by traditional signature-based firewalls. AI-powered network defense (Darktrace, Vectra AI) monitors all network traffic for anomalies and responds in milliseconds, faster than any human team. The policy ask: require all operators of critical infrastructure (power, water, hospitals, financial systems) to deploy AI-powered network monitoring. Cost: ~$2B/year for the US, $10B globally.
Global Pathogen Early Warning Network
We need a global network of metagenomic sequencing stations, at airports, wastewater treatment plants, and hospitals, that continuously scan for novel pathogens. Current wastewater surveillance detected COVID variants 1-2 weeks before clinical cases appeared. Scaling this globally would give us a 2-4 week head start on every future outbreak. Cost: ~$5B to build, ~$1B/year to operate. Countries that join share data and receive early warnings.
Voluntary Safety Treaty Among AI Labs
Anthropic, Google DeepMind, OpenAI, Meta, xAI, and Mistral could sign a voluntary treaty committing to: (1) not deploy systems that score above a threshold on dangerous capability evaluations; (2) share safety research with each other and with regulators; (3) halt deployment if a credible safety threat is identified. Precedent: the Asilomar conference on recombinant DNA in 1975, where scientists voluntarily paused research until safety protocols were established. Labs have economic incentives to sign: it establishes them as responsible actors and heads off heavier government regulation.
Welfare Legislation: Banning the Worst Factory Farming Practices
California's Proposition 12 (effective 2024) banned the sale of pork from pigs kept in gestation crates, affecting the entire US pork supply. The model works: companies reformulate rather than lose the California market. The next steps: (1) Federal legislation banning battery cages, gestation crates, and veal crates, already banned in the EU; (2) mandatory pre-slaughter stunning for all animals; (3) a carbon tax on meat that reflects its true environmental cost.
International Cyber Incident Response System
When a major cyberattack hits, there is no international coordination mechanism. A Global Cyber Emergency Response Team (GCERT), modeled on Interpol but focused on cybercrime and state-sponsored attacks, would: share threat intelligence in real-time; provide technical assistance to nations under attack; attribute attacks using a credible international process; and coordinate sanctions against attacking nations.
Cryptographic Chip Licensing for Frontier AI
Every advanced AI chip (Nvidia H100 and above) should require a cryptographic license before it can run training runs above a certain compute threshold. The license is issued by a new international body, modeled on the IAEA, only after safety audits. TSMC and ASML, who control chokepoints in chip manufacturing, are the enforcement mechanism. This approach doesn't require China's cooperation to be effective: if the chips don't work without the license, no one can train dangerous models without detection. Cost: ~$500M to set up the licensing infrastructure. Benefit: prevents the single highest-risk scenario facing humanity.
Affordable Nutritious Food at <$1/Day: The Global Food Solution
Combining ALLFED food technologies with seasoning science creates food that is nutritious, delicious, and affordable. A daily diet based on mushroom mycoprotein, leaf protein concentrate, or single-cell protein, seasoned with MSG, chili, cumin, and garlic, provides complete protein at under $0.50/day. Add carbohydrates and you have a full day's nutrition for $0.80. Deploy this through Library++ hubs and school feeding programs to eliminate hunger as a barrier to education.
International Cybercrime Treaty with Real Enforcement
Most cybercriminals operate from countries that don't extradite to the US or EU. A binding international cybercrime treaty would require member nations to: criminalize ransomware and infrastructure attacks; extradite or prosecute cybercriminals; share evidence for cybercrime investigations. The enforcement mechanism: nations that harbor cybercriminals are cut off from SWIFT and the global financial system.
Open Source Security: Bug Bounties for Critical Software Infrastructure
The Log4Shell vulnerability (2021) affected hundreds of millions of systems and was maintained by a volunteer. Critical open source software, the backbone of the global internet, is maintained by a tiny group of underfunded volunteers. A $1B/year government-funded bug bounty and maintenance program for critical open source projects would protect systems that every government and business depends on.
Precision Fermentation for Animal-Free Dairy and Eggs
Perfect Day makes whey protein identical to cow's milk using precision fermentation. Impossible Foods' heme comes from the same process. These products are already available, the bottleneck is cost and scale. If the US military switched 20% of its dairy procurement to precision fermentation products, it would create enough demand to push costs down by 40-60%, making these products cost-competitive for the mass market within 5 years.
US-China Treaty on AI Safety Standards
The US-China rivalry is the biggest obstacle to coordinated AI governance. But both nations share an existential interest in not building systems that kill everyone. A bilateral treaty modeled on nuclear arms control could: (1) set agreed compute thresholds above which safety testing is mandatory; (2) establish a joint hotline for AI incidents, modeled on the nuclear hotline; (3) create shared evaluation protocols so each side can verify the other's claims. China's cooperation is obtainable: their AI safety researchers already cite the same risks as Western ones, and their leadership has shown openness in previous Bletchley-style summits.
Microfinance 2.0: Mobile Credit for a Billion Unbanked People
M-Pesa in Kenya enabled 96% of Kenyan households to use mobile money, lifted 2% of households out of poverty, and created a $10B+ fintech industry, all from a simple mobile payments system. Applied to just 20 resource-rich African nations, a Global Sovereign Wealth Fund Framework would generate $500B+ for public services by 2040. An international mobile financial access standard, with a $5B implementation fund, would replicate M-Pesa's results globally within a decade.
Evidence-Based Policy: What Works Centres for Every Government
The UK's What Works Centres evaluate which policies actually work and make the results publicly available to policymakers. Mandating that every G20 nation create equivalent What Works Centres, funded at 0.1% of the budget for each policy area, would dramatically improve government effectiveness. A policy that gets 5 stars gets funded; one that gets 1 star doesn't. This sounds obvious, yet it's extremely rare in practice.
Algorithmic Transparency Requirements for Government AI
Governments increasingly make decisions using algorithms: who gets benefits, who gets parole, who gets audited. An Algorithmic Accountability Act would require: (1) disclosure of all government algorithms affecting individual rights; (2) independent audits for bias and accuracy; (3) a right to human review for algorithmic decisions; (4) open-source release of non-security-sensitive government algorithms.
Vertical Farming + AI: Producing 10x More Food per Acre
Vertical farms produce 10-100x more food per acre than conventional agriculture, use 95% less water, require no pesticides, and can be located anywhere. Current bottleneck: energy costs. The solution: co-locate vertical farms with renewable energy sources, use AI optimization of light recipes to maximize yield per watt, and focus on highest-value crops where vertical farming is already cost-competitive. A G20 subsidy of $5B/year for vertical farming R&D would compress the timeline to cost parity for staple crops from 20 years to 8 years.
International AI Safety Agency (IASA), An IAEA for AI
The International Atomic Energy Agency monitors nuclear programs in 173 countries with 2,500 staff. We need an equivalent for AI. The IASA would: inspect frontier AI labs for compliance with safety standards; maintain a registry of all training runs above threshold compute; provide technical assistance to nations building safe AI; and recommend sanctions against violators. Budget: ~$2B/year, paid by member nations pro-rata by GDP.
Single-Cell Protein: Feeding the World from Methane and Agricultural Waste
Single-cell protein (SCP) is protein produced by bacteria or yeast fed on methane, agricultural waste, or CO2. ALLFED estimates SCP could fulfill 7-11% of global protein needs within 1 year of full deployment. Current cost: $3-5/kg for methane-based SCP, already competitive with soy protein isolate. The policy intervention: the same subsidies currently given to factory farming ($500B/year globally) redirected to SCP facilities would make SCP cheaper than chicken within 5 years.
Clean Air Act for Buildings, Pandemic-Proofing Indoor Spaces
Air quality standards for outdoor air exist in most countries. No equivalent standard exists for indoor air, where people spend 90% of their time. A Clean Indoor Air Act would mandate minimum air changes per hour, HEPA filtration standards, CO2 monitoring (CO2 is a proxy for viral aerosol concentration), and far-UV installation in high-risk spaces. This would reduce transmission of every respiratory pathogen by an estimated 50-80%.
Solar Microgrids: Energy Independence for 1 Billion People
760 million people have no access to electricity. Traditional grid extension costs $10,000-50,000 per kilometer. Solar microgrids can provide reliable electricity for $200-500 per household, cheaper than grid extension and faster to deploy. Bangladesh has deployed 6 million solar home systems; Rwanda built a national microgrid in 5 years. A Global Solar Microgrid Fund of $30B would electrify every currently unelectrified community on Earth.
Citizens' Assemblies for Hard Political Problems
Ireland legalized abortion and same-sex marriage not through partisan political battles but through Citizens' Assemblies, randomly selected citizens who spent months hearing expert evidence and deliberating before making recommendations. The model works because citizens chosen by lottery are free from campaign finance pressure. Citizens' assemblies have now been used successfully in Belgium, France, the UK, and Canada.
Mushroom Mycoprotein Revolution: 54.5% Protein, Zero Suffering
Oyster mushroom mycoprotein achieved 54.5% protein yield in submerged fermentation trials using agricultural waste as feedstock. King trumpet mushrooms grill like scallops. Lion's mane shreds like pulled pork. Industrial mycelium production cycles run in 12 days vs 6 weeks for chicken. Combined with MSG and spices, these become everyday staples people genuinely enjoy. What's missing: scaling investment and a government mandate to serve these in school cafeterias.
Global Apprenticeship System: From School to High-Paying Career
Germany's dual apprenticeship system produces the world's lowest youth unemployment rate (5.8%) and the highest proportion of workers with industry-specific skills. Companies sponsor apprentices and pay them during training; the government subsidizes employer costs. Graduates enter the workforce with 2-3 years of experience and no student debt. A G20 commitment to implementing apprenticeship systems, with a $10B/year fund to subsidize employer participation, would transform global youth employment.
Wellbeing Economy: Beyond GDP as the Goal of Government
New Zealand replaced GDP as its primary measure of success with a Wellbeing Budget that tracks: life satisfaction, loneliness, housing quality, child poverty, mental health, and environmental sustainability. Scotland, Iceland, Wales, and Finland have joined the Wellbeing Economy Alliance. A G20 commitment to adopt Wellbeing Economy frameworks would reorient government priorities toward what actually matters.
BlueDot Impact: The 30-Hour Course That Gets You a Job at Anthropic
BlueDot Impact runs a 30-hour online AI safety fundamentals course that has led directly to employment at Anthropic, DeepMind, OpenAI, and other leading AI labs. Cost to the student: free. Employment outcome: $100K+ jobs in the field most likely to determine whether AI goes well for humanity. Scale needed: to adequately staff AI safety research, we need 10,000+ new researchers over the next decade. Funding ask: $50M/year to scale to 10,000 graduates/year.
Carbon Tax With Full Dividend: Climate Action That Pays Everyone
A carbon tax with a full dividend to citizens, where all revenue is returned equally to every person, is the most economically efficient way to reduce emissions while being politically durable. British Columbia's carbon tax reduced emissions 15-20% while boosting economic growth. At $50/tonne CO2, the US government would collect ~$300B/year. Divided equally, every American household receives ~$2,000/year. The bottom 60% of households come out ahead.
Direct Cash Transfers at Scale: GiveDirectly Model for 1 Billion People
GiveDirectly has run the world's largest basic income experiment in Kenya, giving $1,000 to ultra-poor households with no strings attached. Results after 3 years: $2.60 returned to local economy for every $1 given; child nutrition improved; psychological wellbeing increased; no evidence of increased alcohol use or reduced work. The intervention works because it trusts poor people to know what they need. Scaling to 1 billion people in extreme poverty would require ~$1 trillion over 10 years.
School Feeding Programs: Universal Free Nutritious Meals
Universal school feeding programs improve attendance, cognitive performance, and lifetime earnings, while providing a guaranteed market for alternative protein producers that drives down costs. Universal coverage would cost ~$20B/year. Every $1 invested in school feeding returns $3-10 in increased adult earnings. Using these programs to introduce alternative proteins normalizes these foods and creates the demand needed to drive production costs to parity with factory-farmed meat.
Creating Change Through Beauty and Art
Create a compelling speech, song, or cinematic video aimed at leaders of authoritarian regimes to help them accept a treaty or diplomatic deal. Think of all the visuals showcasing the bright futures a good peace plan could bring, and the horrors that might follow without one. A song, a speech, a cinematic documentary, nobody would refuse a well-framed diplomatic deal, especially with a beautiful video to accompany it. Al Gore's 'An Inconvenient Truth' shifted climate policy globally. 'Seaspiracy' changed dietary behavior for millions. A high-quality cinematic piece highlighting the vision of a flourishing, peaceful future could be a powerful diplomatic tool alongside formal peace deal documents. Budget: $10-50M. Potential audience: 500M+ people.
Employ Developing World University Students in the Global Longtermism Economy
Many developing nations have millions of university-educated citizens with high talent and few high-quality job opportunities. With open internet access and stable infrastructure, this represents an extraordinary talent pool for the world's most important problems. The opportunity: connect educated populations in the Global South to: • AI safety research (30-hour BlueDot Impact courses, then employment at Anthropic, DeepMind, etc.) • Biosecurity and pandemic preparedness • Nuclear risk reduction and arms control • Climate policy and renewable energy • Cybersecurity • Global food security research Remote work enables earning international wages ($40-80k USD) while living in their home country, creating an enormous positive feedback loop for local economies without requiring emigration. The Library++ model makes this accessible to anyone within 20 minutes of a hub.
ALLFED-Based Resilient Food Production: Feeding Nations in Crisis
Deploy ALLFED (Alliance to Feed the Earth in Disasters) food production technologies in food-insecure or crisis-affected nations: Single-Cell Protein from methane ($3-5/kg, could fulfill 7-11% of global protein needs by Year 1), Leaf Protein Concentrate from agricultural waste, cellulosic sugar from biomass, and rapid greenhouse deployment. Combined with seasoning science (MSG + dried chili + cumin at <$0.01/serving), these foods become not just edible but genuinely delicious. Pilot these alongside mushroom mycoprotein facilities using agricultural waste as free feedstock. Any nation can implement this, the technologies are modular, scalable, and work with locally available inputs.
AI-Driven Biological Engineering for Feed-to-Food Conversion
Use AlphaFold-style AI breakthroughs to accelerate development of organisms and enzymes that convert non-edible biomass (86% of livestock feed is material humans can't eat) into human food. Fungal fermentation with AI optimization, machine learning for enzyme engineering, and generative biology platforms could cut development time by 50-70%. Realistic timeline: proof-of-concept in 2-4 years, pilot-scale at $5-10/kg in 5-8 years, cost parity with conventional agriculture in 10-15 years.
The Seasoning Package: Making Resilient Food Delicious
The single biggest barrier to alternative protein adoption is taste. Solution: a standardized seasoning package deployed alongside every food innovation. MSG (pinch per dish, pennies), dried chili (one plant seasons 350 meals), cumin (one plant = 100 meals), garlic powder, soy sauce. These transform bland single-cell protein or leaf protein into food people choose to eat. Total agricultural footprint for seasoning: negligible. Total cost per meal: <$0.01.
Precision Fermentation Facilities for National Protein Independence
Build precision fermentation facilities in food-insecure or protein-import-dependent nations that produce protein at scale using agricultural waste feedstock. Current costs: $10-20/kg with production yields of 15-25 g/L. Technology requires production titers above 50g/L for cost-competitiveness. Each facility costs $125-500M but provides protein independence from imports. Nations with large agricultural waste streams and educated workforces are ideal early adopters. Deploy alongside ALLFED food systems for complete food sovereignty. This is viable for any nation, the inputs are local waste, the outputs are nutrition independence.
Mushroom Mycoprotein Revolution for National Food Supply
Oyster mushroom mycoprotein can be produced by submerged fermentation using agricultural waste as feedstock, achieving protein yields of 54.5%. A mushroom farm doubles as an ALLFED-style single-cell protein facility. King trumpet, lion's mane, and oyster mushrooms grill, shred, and smoke like meat. Industrial mycelium production runs in 12-day cycles. Infrastructure costs are a fraction of traditional protein production. This model works in virtually any climate, mushrooms grow on straw, cardboard, coffee grounds, and sawdust available in every country. Combined with MSG and spices, these become everyday staples people genuinely enjoy. Any nation can deploy this within 2-3 years to dramatically reduce dependence on factory-farmed meat.
Transform Any Nation into a Global Hub for AI Safety, Biosecurity & Longtermism
With internet, food, clean water, and workspace, citizens of any nation can work in the most impactful fields: AI safety (30-hour BlueDot Impact courses), cybersecurity, biosecurity, nuclear risk prevention, diplomacy, and scientific innovation. Remote work enables earning international wages ($40-80k USD) while living at home. The infrastructure for this is achievable via Library++ community hubs providing internet + workstations + training + water + food. Target: 50,000+ people per participating nation working in longtermism fields within 10 years. This model works in any country with an educated population and reliable internet, and Library++ makes the internet part achievable almost anywhere.