How do transformative innovations emerge, and can we deliberately accelerate their discovery and impact? From electricity that powers modern civilisation to vaccines that have saved hundreds of millions of lives, innovations have reshaped human existence. This post explores innovation as an engine of human advancement, examining its successes and the flaws in how we generate knowledge.
These examples reflect a broader historical pattern. Agricultural revolutions enhanced food security, the internet connected billions to information, and smartphones put powerful computing in the hands of people worldwide. Innovation has consistently lifted societies and expanded opportunity, demonstrating a capacity to alter the trajectory of well-being on a large scale. For anyone concerned with global development, supporting such change offers valuable pathways to solve large problems.
Whilst the impact of successful innovation is clear, the process of generating and scaling it effectively is complex. Norman Borlaug's work on disease-resistant wheat, a cornerstone of the Green Revolution, was the culmination of decades of prior research and required coordination across international organisations, governments and farmers. Innovation typically depends on a confluence of factors, including scientific research, practical implementation strategies, contextual understanding and an enabling environment for adoption.
To better navigate this landscape we turn to metascience. Metascience is the study of science and innovation itself. Seeking to increase the quality of research by analysing how knowledge is created, validated, disseminated and implemented.
Metascience researchers typically focus on areas such as:
Methods: How researchers design studies and analyse data
Reporting: How scientific findings are communicated
Reproducibility: Whether studies can be replicated by independent researchers
Evaluation: How scientific work is assessed through peer review
Incentives: What motivates researchers and institutions
When we look at innovation successes and failures through this lens, we sometimes find surprising outcomes. Much research can't be replicated. Brilliant ideas often take decades to spread. Funding sometimes flows to popular but unproductive areas while neglecting promising alternatives. By studying these patterns, metascience could help us understand what works and what doesn't.
Sections
How Innovation Transforms Development
Electricity
Containerisation
Biomedical Breakthroughs
Mobile Communications
The Green Revolution
This all sounds great, but what’s the counterfactual?
Metascience
The Scope of Metascience
The Innovation and Metascience Landscape
Academia
Government
Private Sector
Philanthropy
Cross-sector examples
Metascience
Progress Studies
How Innovation Transforms Development
The few centuries have seen innovations completely transform human life. Cities built sewage systems and provided clean water sources, drastically reducing waterborne diseases like cholera and typhoid. Vaccines eliminated smallpox. The Green Revolution doubled crop yields, preventing mass starvation. Electricity extended productive hours beyond sunset. The printing press, then the telegraph, then telephones, then the internet dramatically reduced the cost of sharing information.
Today, billions still face challenges that have been solved in many parts of the world. Farmers struggle to access markets, mothers die from preventable causes, villages lack reliable power. Yet new innovations are addressing these old problems in novel ways, and continuing to reshape our world. Here are some prominent examples of how innovation drives development.
Electricity: Powering Modern Life and Productivity
Electricity is an undisputed lifeline of a modern economy and stands as one of history's most profound innovations. The positive relationship between energy and economic growth is stark: income and energy consumption are tightly correlated across every continent and time period.
Whilst causality can run both ways, mounting evidence indicates that energy consumption is a necessary enabler, and a powerful driver, of economic growth, critical for providing the necessities of human life - education, healthcare, food and jobs.
Access to reliable electricity fundamentally reshapes economic and social possibilities. Research from Ghana suggests that a doubling in electricity power consumption will cause real GDP per capita to increase by ~52%. Conversely, power outages, a common challenge in many regions, significantly curtail manufacturing output and business profitability.
In India, districts with low electricity access exhibited two times higher rates of age-standardised premature mortality in women, with one in five such deaths linked to poor electricity access. Reliable power is essential for functioning health clinics, refrigerating vaccines and medicines, and reducing reliance on polluting indoor cooking fuels.
Looking ahead, the innovation driving energy's impact continues at pace. Breakthroughs in renewable energy sources like solar and wind, advancements in energy storage such as improved battery technologies, the ongoing development of next-generation nuclear power and the rise of electric vehicles are all poised to further transform how we generate, distribute and consume energy. These technologies hold the promise of not only expanding access to affordable power globally but also mitigating climate change and air pollution.
Containerisation: Enabling Global Trade
Few innovations have had as dramatic an impact on the architecture of the global economy as the standardised shipping container. Introduced in the 1950s, this seemingly simple metal box revolutionised the movement of goods across oceans by drastically reducing costs and complexity. The expense of loading and unloading cargo plummeted to 2% of the costs pre 1960.
Econometric studies suggest that containerisation boosted bilateral trade by 900% over a 15-year period. By making long-distance shipping economically viable, containerisation enabled the rise of global supply chains. This facilitated the shift of manufacturing to more countries, creating millions of jobs and providing new development pathways for export-oriented economies.
Containerisation also highlights how some innovations aren’t necessarily academic, highly technical or scientific but can come from business and cultural practices.
Biomedical Breakthroughs
The pace of innovation in biomedical science has transformed human health and longevity. Foundational discoveries like antibiotics and vaccines laid the groundwork.
The rapid development and deployment of mRNA COVID-19 vaccines are estimated to have saved nearly 20 million lives in their first year alone. Beyond vaccines, gene editing tools like CRISPR hold promise for treating inherited diseases and potentially tackling infectious diseases.
In pharmaceuticals, the development of antiretroviral therapies transformed HIV/AIDS from an almost certain death sentence into a manageable chronic condition. Innovations in production and distribution saw treatment costs in low-income countries drop from over $10,000 per person annually in 2000 to under $100 in 2016, saving an estimated ~16.5 million lives.
More recently, new classes of drugs like GLP-1 agonists (Ozempic, Wegovy, etc) are showing remarkable efficacy in treating obesity and type 2 diabetes, offering new hope for tackling the growing global burden of non-communicable diseases and their associated economic costs.
Matt Reynolds - Salt, Sugar, Water, Zinc: How Scientists Learned to Treat the 20th Century’s Biggest Killer of Children
One of the most transformative yet underappreciated healthcare innovations of the 20th century is oral rehydration therapy (ORT). A simple solution of salt, sugar and water that has saved over 70 million lives since its introduction in the 1970s
Before ORT, diarrhoeal diseases were the leading killers of children worldwide. Treatment required intravenous fluids administered by trained medical professionals with sterile equipment - resources unavailable in most places where cholera and other diarrhoeal diseases struck hardest. As one of ORT's developers, observed: "It's better to reach 80 percent of people with something that's 80 percent effective than five percent of people with something that's 100 percent effective."
The key breakthrough came when scientists discovered that even during severe diarrhoea, the gut could still absorb water, salt and glucose when they were present in specific proportions. This simple solution slashed child deaths from diarrhoeal diseases from 4.8 million in 1980 to about 500,000 in 2023
What makes ORT remarkable isn't just its effectiveness but its accessibility, it can be prepared by anyone with access to clean water, sugar and salt, making it a key example of frugal innovation designed for maximum impact in resource-constrained settings
Mobile Communications: Connecting Billions
Few innovations have increased access to information and opportunity as swiftly as the mobile phone, particularly the smartphone. This device has rapidly rewired global communication and economic participation. By the end of 2023, mobile internet was in the hands of an estimated 4.6 billion people worldwide.
This explosion in connectivity ignited a powerful economic engine. In 2024, mobile technologies and services contributed ~5.8% to global GDP, ($6.5 trillion). In 2022, mobile services contributed ~$170 billion to economies in Sub-Saharan Africa, driven mostly by improvements in productivity and efficiency.
The smartphone also serves as a platform for other innovations. With mobile finance, health consultations, remote learning resources, real-time agricultural market data, accessible digital government services, etc.
The Green Revolution & Modern Agricultural Science
Ensuring global food security in the face of a growing population and environmental challenges remains a paramount development goal, and agricultural innovation is central to achieving it. The original Green Revolution of the mid-20th century, spearheaded by scientists like Norman Borlaug, introduced high-yield crop varieties that dramatically increased food production, particularly in Asia, and are credited with saving up to a billion lives from starvation.
Modern agricultural science continues to deliver advancements. Genetic improvements have yielded crops with enhanced resilience, such as drought-tolerant rice varieties. Disease-resistant crops, like Bt cotton, have increased yields and farm profits, helping lift people out of poverty.
Innovations also focus on sustainable intensification. Precision agriculture techniques can save on inputs, whilst machine learning-based fertiliser recommendations show potential for significant yield gains.
The Haber-Bosch process for synthetic nitrogen fertiliser production, a 20th-century innovation, broke natural constraints on food production and is estimated to underpin the food supply for roughly half the world's population today.
This all sounds great, but what’s the counterfactual?
The transformative power of innovation is undeniable, from ships to computers to the internet. But this still raises a question: compared to what? How do we know if channeling resources into innovation is more impactful than alternative approaches to improving lives?
Consider the challenge of reducing childhood mortality in sub-Saharan Africa. Any of the below options sound reasonable.
Direct intervention - Distribute malaria treated bed nets
Economic growth support - As countries gets richer, there will be more funding for healthcare systems, better roads to transport medicine, more energy to power hospitals, and many more improvements that make it easier to eliminate malaria completely
Innovation - Fund research into malaria vaccines that could eliminate malaria within years of discovery
Each pathway has merit and opportunity costs. The $100 million spent developing a new vaccine is $100 million not spent on proven interventions that could save lives today. Yet without innovation, we'd still be treating diseases the same way we did centuries ago.
There is also a difference between innovation focused directly on issues faced mainly by LMICs and general innovation.
Directly-targeted innovation
Focuses explicitly on challenges faced in LMICs. The development of heat-stable vaccines, for instance, directly addresses the cold chain challenges in tropical regions. These innovations follow a clear theory of change from laboratory to lives saved
General advancement
Creates technologies without specific development goals that later prove transformative. Mobile phones weren't designed to bank the unbanked, yet mobile finance now serves millions. The internet wasn't created to educate rural students, yet online learning platforms now reach the most remote communities
Determining a more impactful allocation of resources and talent for these different paths has considerable uncertainty. Should we prioritise the immediate and tangible benefits of direct interventions, the systemic uplift of economic growth or the potential of innovation? Even if we lean towards innovation, the choice between funding a specific malaria vaccine versus foundational research, both with potential global benefits but via different mechanisms and timelines, is not clear.
The answer likely depends on several factors, here are a few that could impact those decisions.
Time horizons
Direct interventions help people today, innovation might help far more people tomorrow
Risk tolerance
Proven interventions have predictable impact, innovation involves many failures for each success
Comparative advantage
Some individuals and institutions are better positioned for discovery than delivery
Neglectedness
Markets often undersupply innovation for diseases of poverty but may oversupply it for profitable conditions
On a personal level, your background and interests provide important context. A biochemist might contribute more to vaccine development than to distributing bed nets. Someone with logistics expertise might save more lives optimising supply chains than pursuing laboratory research. And how much uncertainty can you tolerate? Innovation often means working for years on problems that might prove unsolvable.
We can also acknowledge what we don't see, the innovations that didn't happen because our research systems failed. How many potential breakthroughs languished because academic incentives rewarded quantity over quality? How many promising discoveries had very delayed impact because we lack mechanisms to bridge the lab-to-field gap?
This is where metascience enters the picture, as a multiplier that could improve returns across all pathways. If better research practices could accelerate discovery by even 10%, the cumulative impact over decades could dwarf almost any individual intervention.
Metascience
Metascience (or meta-research) is the use of scientific methodology to study science itself. Metascience seeks to increase the quality of scientific research while reducing inefficiency.
Early efforts can be traced back to the 1960s when researchers began examining the methodological foundations of scientific practice. In 1966, an early meta-research paper had examined the statistical methods of ~300 papers published in medical journals, finding that
“in almost 73% of the reports read…conclusions were drawn when the justification for these conclusions was invalid.”
A 1976 paper called for dedicated funding for “research on research,” recognising that “the very nature of research on research, particularly if it is prospective, requires long periods of time.” These early investigations revealed widespread methodological flaws and inefficiencies across numerous scientific fields, laying the groundwork for what would later become the formal discipline of metascience.
The replication crisis is an ongoing methodological issue in which it has been found that many scientific studies are difficult or impossible to replicate. While the crisis has its roots in the meta-research of the late 20th century, the phrase "replication crisis" was not coined until the early 2010s as part of a growing awareness of the problem.
Other issues discovered include:
Widespread misuse of p-values and abuse of statistical significance
Perverse incentives that have created a publish-or-perish environment promoting junk science
Bias in peer review and funding decisions
Whilst concerns about scientific quality have existed for decades, modern metascience largely traces its origins to a 2005 paper by John Ioannidis titled "Why Most Published Research Findings Are False." This essay became the most-accessed article in the history of the Public Library of Science with over 3 million views and served as a catalyst for broader scrutiny of research practices.
Metascience gained institutional momentum in the 2010s. Stuart Buck at the Laura and John Arnold Foundation allocated over $60 million to metascience starting in 2012, funding metascience entrepreneurs including the Centre for Open Science.
Stuart Buck - Metascience Since 2012: A Personal History
(I would really recommend reading this post, it covers a lot of the history of modern metascience).
2012 - Arnold Foundation enters metascience after John Arnold asks about psychology replication crisis, leading Buck to investigate widespread irreproducibility across fields
2013 - Launch of Center for Open Science with Brian Nosek and METRICS at Stanford with Ioannidis to drive research on reproducibility
2014 - Academic resistance emerges, with replicators being called ‘shameless little bullies’ as replication efforts gain momentum
2015 - Reproducibility Project: Psychology published in Science, showing only ~40% replication rate - becomes most cited paper in the initiative
2016 - TOP Guidelines launched with 5,000+ journal adoptions
2018-2020 - Major infrastructure scaling with Chris Chambers' Registered Reports expansion and continued growth of Open Science Framework
As Buck notes “A decade ago, not many folks talked about 'metascience' or related issues like scientific replicability. Those who did were often criticised for ruffling too many feathers.” Today, the situation seems far different: pre-registration has become much more common across many fields, the American Economic Association employs a whole team to rerun data and code from every published article, and national policymakers at NIH and the White House have launched numerous initiatives on reproducibility and open science.
The Scope of Metascience
Metascience it not just about replication and pre-registration. Researchers tackle questions across several domains.
Research methods
How scientists design experiments, collect data and draw conclusions. Poor methodology can lead to false discoveries that waste resources and mislead future research
Which research designs are most likely to yield reliable results and how statistical practices can be improved to reduce bias
Investigating optimal sample sizes, exploring novel analytical techniques and understanding when different research approaches are most appropriate
Scientific communication
How findings are reported in academic papers, presented at conferences and shared with the public
Research shows that studies with more accessible titles and abstracts are more likely to be cited and potentially translated into practice
How scientific language and presentation affect interpretation, and how findings move from academic journals to broader public understanding
Evaluation systems
How scientific work is assessed through peer review, citation metrics and tenure decisions
These evaluation mechanisms shape what research gets funded, published and rewarded, ultimately determining the direction of scientific progress
Are there systematic biases in peer review?
Explores alternative models for evaluating research quality, and investigates how current systems may inadvertently discourage innovative or interdisciplinary work
Incentive structures
What motivates individual researchers, laboratories and institutions
Misaligned incentives can encourage quantity over quality, novelty over replication and publication over practical application
Studying how promotion criteria, funding mechanisms and professional recognition shape researcher behaviour, often in ways that may conflict with broader scientific goals
Research organisation and funding
How scientific institutions operate and allocate resources
Optimal research team sizes, the effectiveness of different funding models, the role of collaboration networks and how bureaucratic processes affect scientific productivity
Whether current models for organising science maximise discovery and innovation
Knowledge integration and translation
How scientific findings are synthesised across studies and translated into practical applications
Includes understanding why some discoveries rapidly translate into impact whilst others remain confined to academic literature, and developing better systems for integrating knowledge across fields and linking research to societal needs
There are plenty of suggested interventions but I would like to highlight1 this post by Brian Nosek from the Center for Open Science - Strategy for Culture Change.
Individual motivation and ability to change behaviour are insufficient when cultural barriers exist. Researchers operate within systems that provide incentives and impose policies that may conflict with desired behaviours. Despite researchers generally valuing transparency and having motivation for reproducible practices, academic culture rewards novel, positive results through publications and career advancement, creating incentives that discourage showing work and sharing materials
In other words, if you want scientists to change their behavior by sharing more data, you need to start at the bottom by making it possible to share data...Then try to make it easier and more streamlined, so that sharing data isn’t a huge burden. And so on, up the pyramid.2
You can’t start at the top of the pyramid (“make it required”) if the other components aren’t there first. For one thing, no one is going to vote for a journal or funder policy to mandate data sharing if it isn’t even possible. Getting buy-in for such a policy would require work to make data sharing not just possible, but more normative and rewarding within a field.
That said, I might add another layer at the bottom of the pyramid: “Raise awareness of the problem.” For example, doing meta-research on the extent of publication bias or the rate of replication can make entire fields aware that they have a problem in the first place.
-Raise Awareness: fundamental research on the extent of irreproducibility
-Make It Possible and Make It Easy: the development of software, databases, and other tools to help improve scientific practices
-Make It Normative: journalists and websites that called out problematic research, and better standards/guidelines/ratings related to research quality and/or transparency
-Make It Rewarding: community-building efforts and new journal formats
-Make It Required: organisations that worked on policy and advocacy
The Innovation and Metascience Landscape
Research and development (R&D) has annual investment reaching ~$2.75 trillion in 2023. The United States leads with $784 billion invested, while China follows closely at $723b. This funding flows through an interconnected ecosystem of businesses, academia, government agencies, and philanthropy, with the private sector contributing both the majority of funding and conducting most research activities.
R&D can include a wide range of activities ranging from basic science, specific innovations that would help with global development and research that may only have profit motivations and could be neutral or even negative for humanity.
National Science Foundation -R&D: US Trends and International Comparisons
The business sector dominates the U.S. R&D landscape, performing $693b (78%) of total R&D in 2022, while also providing the majority of funding at $673b (76%). Higher education institutions performed $91.4b (10%), and the federal government performed $73.3b (8%).
Federal government funding accounted for 18% of US R&D funding ($159.8b) with 40% of its funding directed toward basic research and higher education institutions receiving the largest share (30%)
Five industries dominated US business R&D in 2021, accounting for 79% of the total
Information (25%)
Chemicals manufacturing (18%)
Computer/electronic products (17%)
Professional/scientific services (11%)
Transportation equipment (8%)
The semiconductor industry increased ~10% in 2021 to $47.4 billion after a ~23% increase in 2020
Global Trends
Global R&D expenditures have more than tripled in real terms since 2000, reaching approximately $2.75 trillion in 2023. With R&D as a percentage of GDP increasing from under 1.5% to 2% in 2023.
Asia now accounts for approximately 46% of global R&D (up from 25% in 2000). North America's share has declined to 29%, followed by Europe at 21%.
The United States ($784b) and China ($723b) now dominate global R&D spending, with a substantial gap to the third-placed Japan ($184b), followed by Germany ($132b), South Korea, the UK ($88b) and India ($71b). When considered as a single entity, the EU's R&D expenditure ($410b) is approximately half that of the US or China.
Academia
Academic institutions provide spaces for fundamental research and serve as training grounds for future scientists and innovators. Universities occupy a position between theory and application where researchers can pursue questions that may lack immediate commercial application but hold potential for long-term impact.
Scale
Higher education institutions perform ~13-17% of global R&D. Within the US, in 2022, academic institutions performed ~$91.4b (10% of the US total). Whilst it isn’t a high percentage of R&D the academic sector performs roughly 60% of basic research in the US. Governments remain the primary supporter of academic research globally, though industry partnerships and philanthropic grants are increasingly happening.
Examples
The World Wide Web originated at CERN but was developed into its recognisable form through Tim Berners-Lee's work at MIT
Though building on earlier research, developments at the University of Aberdeen led by John Mallard produced the first full-body MRI scanner and useful images in 1980, allowing non-invasive medical diagnosis without radiation exposure
Recombinant DNA Technology - Developed in the 1970s at Stanford University this technique for enabled the production of human insulin from bacteria
Challenges
The pressure to secure external funding can direct research toward short-term, applied goals over fundamental science
Academic incentive structures often prioritise publication quantity, sensationalism and statistical significance over quality, replicability or practical implementation
Many published studies lack statistical power to detect meaningful effects, leading to systematically inflated effect sizes in the literature
Research transparency is inconsistent, with fewer than half of published papers in some fields having fully reproducible code and data
Technology transfer processes between universities and industry can be bureaucratic and inefficient, hindering the practical application of academic discoveries
Government
Government plays both a direct and indirect role in the innovation landscape as a performer and funder of R&D. Public investment often addresses research areas where market incentives may be insufficient but societal benefit is substantial. Government agencies can identify strategic priorities, maintain research infrastructure, coordinate large-scale initiatives and create the regulatory framework within which innovation occurs.
Scale
Globally, governments directly perform around 10-15% of total R&D activities, but their influence extends further through funding provided to universities, research institutes and businesses. In the US, the federal government performed 8% of R&D while funding 18% across all sectors. The role varies significantly between countries, in some nations, government institutions perform the majority of research activities.
Examples
Government research has yielded numerous innovations with far-reaching impact. The Internet originated from ARPANET, a project of the U.S. Department of Defense's Advanced Research Projects Agency (DARPA). GPS technology, has become essential to navigation and countless location-based services. The Human Genome Project, led by the National Institutes of Health and Department of Energy, established the foundation for modern genomics and personalised medicine.
Challenges
Political cycles can create funding instability and shift priorities
Bureaucratic processes may slow innovation
Government agencies sometimes struggle with mechanisms for technology transfer to industry
Risk-averse culture may favour safe, incremental projects over transformative but uncertain research
Private Sector
The private sector constitutes the largest component of the R&D landscape in most advanced economies, focusing primarily on applied research and development activities with commercial potential. Companies invest in innovation to develop new products and services, improve efficiency and maintain competitive advantage in global markets.
Business R&D typically excels at translating research into marketable products and services. Private sector innovation is more responsive to market demands and customer needs, with clearer metrics for success and resource allocation.
This sector generally tends to emphasise shorter time horizons and clearer paths to market applications than academic or government research, as seen by the pharmaceutical industry's focus on late-stage drug development rather than basic disease mechanisms, and technology firms' prioritisation of product improvements over theoretical computing advances.
Notable exceptions to the short time horizons include Microsoft Research's decades-long commitment to fundamental computer science and Bell Labs' historical breakthroughs in transistors and information theory.
Scale
Businesses perform approximately 70% of global R&D, with the private sector's role particularly dominant in richer economies. The concentration varies significantly by country, from over 90% of total R&D to less than 20% in some LMICS . In the US, businesses perform 78% of total R&D, and globally, the pharmaceutical, automotive, and technology sectors represent the largest corporate R&D investors.
R&D spend by Amazon, Alphabet, Microsoft, Apple, Meta in financial year 23-24 was nearly $229 billion collectively.
Examples
Bell Labs, the research arm of AT&T, developed the transistor, the laser and information theory fundamentals
Xerox PARC pioneered the graphical user interface, Ethernet networking and laser printing technologies
Challenges
Intellectual property protections can sometimes impede knowledge sharing
Market incentives may result in underinvestment in areas with significant social benefits but limited commercial potential
If a market is dominated by a monopoly they may have less incentive to innovate
Companies may attempt to influence regulators through lobbying, payments to experts, or selective data disclosure to favour their commercial interests (when the research isn’t for their own internal benefit)
Commercial incentives may lead to selective reporting of results, rushed timelines, or research designed primarily to support marketing rather than scientific advancement
Philanthropy
Philanthropic organisations can provide funding that supports novel, high-risk research areas that may fall between the priorities of government agencies and commercial enterprises. Foundations can operate with greater flexibility than public institutions and longer time horizons than businesses, allowing them to seed emerging fields and support unconventional approaches.
Scale
While representing less than 5% of global R&D funding, philanthropic contributions play an outsized role in particular fields, especially biomedical research, climate science and emerging technologies. In the US, philanthropic sources account for approximately 3% of total R&D funding but contribute over 10% of funding for basic research at universities. Major scientific foundations direct billions toward research priorities they identify, often catalysing additional investment from public and private sources.
Examples
The Rockefeller Foundation played a central role in the Green Revolution
The Howard Hughes Medical Institute has supported breakthrough research in genetics, neuroscience and structural biology through its investigator programme
Challenges
Philanthropic funding can be unpredictable and subject to donor priorities rather than systematic assessments of societal needs
The sector has limited resources compared to government or industry
Foundation priorities may not always align with broader scientific consensus
Some cross sector examples
There are also many examples of innovations (including most of those mentioned above) that have had funding from multiple sources and were discovered in a mix of universities, industry labs and government departments over many years.
CRISPR-Cas9 Gene Editing
Initial observations of clustered repeats in bacterial DNA (later named CRISPR) were made by researchers at Osaka University in 1987 and University of Alicante in 1995
A breakthrough came from private industry when food scientists at Danisco (a Danish company) demonstrated in 2007 that bacteria acquire CRISPR-based immunity against viruses, leading to early commercialisation for "vaccinating" bacterial cultures
The technology's adaptation for gene editing in human cells involved academic researchers at institutions including UC Berkeley, the Broad Institute and the University of Vienna, with Jennifer Doudna and Emmanuelle Charpentier receiving the 2020 Nobel Prize in Chemistry for their contributions
mRNA Vaccines
The development of mRNA vaccine technology that enabled rapid response to COVID-19 resulted from decades-long collaboration across sectors
The breakthrough came from academic researchers Katalin Karikó and Drew Weissman at the University of Pennsylvania, who discovered how to modify mRNA to avoid triggering inflammatory immune responses
Their 2005 paper, initially rejected by top journals, laid the groundwork for mRNA therapeutics
The technology's development into viable vaccines required private sector investment from companies like BioNTech (where Karikó later worked) and Moderna, along with government funding through agencies like DARPA and BARDA
Metascience
This field draws from multiple disciplines including statistics, sociology of science, research methodology and science policy to address challenges like reproducibility, publication bias and incentive misalignment in the research enterprise.
Scale
As a relatively new field, metascience receives a small fraction of global research funding, primarily through philanthropic sources and targeted programmes at public funding agencies. Estimated annual funding specifically dedicated to metascience research and infrastructure development is in the range of tens to hundreds of millions of dollars globally.
Organisations & Links
Events
Metascience Annual Conference - London in 2025
International Conference on the Science of Science and Innovation - Denmark in 2025
Organisations
Center for Open Science - Key organiser in the space
Metascience Working Group - Policy-focused science funding experiments
J-PAL - Science for Progress Initiative
Association for Interdisciplinary Meta-Research and Open Science
Accelerator For Innovation & Research Funding Experimentation
Science/Biology Initiatives
Convergent Research - Incubates focused research organizations (FROs)
Astera Institute - Supports promising innovators during their earliest stages of exploration
New Science - aiming to enable researchers who would’ve been working in traditional academia to work on problems they could not work on in academia
Speculative Technologies - nonprofit industrial research lab for public good
Future House - non-profit building AI agents to automate research in biology and other complex sciences
Research Infrastructure & Tools
Some Are Useful - a living literature review on how ideas from ML/AI are influencing scientific and technological progress
Blogs
Metascience 101 is a set of interviews on the debates and ideas driving the metascience movement
Metascience Episode 9 - How to get involved with metascience
The Good Science Blog by Stuart Buck
Scientific Discovery by Saloni Dattani
What’s New Under The Sun by Matt Clancy
Academia
Tilburg - Meta-Research Center
Zurich - Center for Reproducible Science
Wikipedia also has a list of metascience research organisations
Funders
Advanced Research + Invention Agency (ARIA) - UK R&D funding agency built to unlock scientific and technological breakthroughs
Novo Nordisk Foundation - Life science ecosystem support
UK Government
UK Research and Innovation (UKRI)
Department for Science, Innovation and Technology (DSIT)
Progress Studies
Emerged as an intellectual movement focused on understanding the drivers of technological, economic and social advancement and developing frameworks to accelerate beneficial change. The field was formally proposed in 2019 by economist Tyler Cowen and Stripe CEO Patrick Collison in their Atlantic article "We Need a New Science of Progress", though it draws on earlier work in economic history, innovation studies and the history of science.
Scale
As a recently established field it’s relatively small in terms of dedicated funding and institutional support compared to established disciplines. The movement primarily operates through think tanks, online publications and informal networks of researchers rather than through traditional academic departments.
Funding comes predominantly from technology entrepreneurs and innovation-focused philanthropies, with estimated annual support in the tens of millions of dollars.
Organisations
Open Philanthropy - Abundance & Growth Focus Area
Asimov Press - focused on the science and technologies that promote flourishing
ARC Institute - gives scientists no-strings-attached, multi-year funding, mainly in biomedical science and technology
The Up Wing - covering progress related news
Further Resources
Ryan Briggs - Can We Trust Social Science Yet? - “Everyone likes the idea of evidence-based policy, but it’s hard to realise it when our most reputable social science journals are still publishing poor quality research”
Nature - Is science’s dominant funding model broken?
Bottlenecks for Evidence Adoption (in government)
80,000 Hours podcast - Matt Clancy on whether science is good
The University of Chicago - The Market Shaping Accelerator
The Innovation in Government Initiative helps LMICs implement evidence-based policies, and provides assistance for effective policy scale-up initiatives to improve the lives of people living in extreme poverty
Jano Costard - Can we implement learnings from Fast Grants in public institutions that face different rules?
Nature - Scientists rise up against statistical significance
Sophie Gulliver and Tom Drake - The $1 Trillion Paradox: Why Reforming Research Publishing Should Be a Global Priority
Stuart Buck - Why Are We Screwing Over Researchers Who Make Innovative Discoveries? - the University of Pennsylvania has made $1.2 billion in royalties from patents on Katalin Karikó’s work after forcing her out
Adam Mastroianni - The rise and fall of peer review
Saloni Dattani, Rachel Glennerster & Siddhartha Haria - Why we didn’t get a malaria vaccine sooner
Ulkar Aghayeva - What groundbreaking discoveries might have already been made, and how can we uncover them faster?
Jason Crawford - Accelerating science through evolvable institutions
Matt Clancy - Boosting innovation by teaching people to be tech entrepreneurs?
Questions
Overall Reflection
Was there anything surprising from this week's post?
Innovation is presented as a key driver of development alongside direct interventions and economic growth. How do you weigh these different pathways for impact?
How can you balance supporting directly-targeted innovation (malaria vaccines) versus general advancement (smartphones) that later finds development applications?
What are the risks of over-emphasising innovation versus improving delivery of existing solutions?
What would a development ecosystem optimised for breakthrough innovations look like?
How could innovation benefits reach those who need them most rather than just those who can afford them?
Innovation
What role could countries play in their own innovation systems versus technology transfers from wealthier nations?
How can we better identify which innovations have transformative potential before they're widely adopted?
When evaluating innovation opportunities, how should we weigh solutions that help millions moderately versus those that transform lives completely for fewer people?
Who should decide research priorities? Scientists, governments, companies, universities, funders or the communities that interventions are supposed to help?
How should we handle innovations that have both positive impacts and significant negative externalities?
How do we evaluate whether an innovation is truly beneficial when its full impacts may not be visible for decades?
Should development funders completely pivot away from direct interventions and just fund basic science instead?
Oral rehydration therapy (salt, sugar, water) saves millions but took decades to spread globally. What does this say about our innovation distribution systems, and what should we do differently
Are we over rating or under rating university research compared to private sector innovation?
Is the goal of innovation to solve problems or to create new possibilities? Does this distinction matter for how we fund research?
Should innovation be culturally neutral, or should solutions be designed specifically for different cultural contexts?
Metascience & Research
Which metascience interventions seem most promising?
Should we be more sceptical of interventions that cite scientific evidence?
When is it better to focus on improving research quality versus simply doing more research?
If we study how to do science better, are we just creating a new academic field that will have its own problems? What are they likely to be?
Should science funders completely pivot away from basic science and just fund metascience instead?
Careers & Future Impact
Which innovation-related career paths seem most impactful?
Would you rather work on fundamental research with uncertain applications or applied research addressing specific development challenges?
What skills seem most valuable for contributing to innovation or metascience over the next decade?
How might emerging technologies change innovation processes?
What's one innovation you think will likely transform development in the next 20 years that others are sleeping on?
How could AI impact the metascience field? And the rest of science?
Also highlighted by Stuart Buck in his post
Taken from the Stuart Buck post