Decentralized Science (DeSci): The Renaissance of Independent Scientists

Authors: Yi Zhang (PhD, Codatta, @drtwo101), Diana (BNBChain, @dianabnb), Eva (AuraSci, @1vayou), Andrea (OGV, @Andrea_Chang), Lucy (@BoboLucyWisdom)

Editor: Tess Li (Codatta, @li_tess)

Introduction

Decentralized Science (DeSci) is fundamentally changing our approach to scientific research by addressing the limitations of traditional centralized systems. Historically, great discoveries often stemmed from independent scientists working without institutional priorities or corporate funding constraints. Today, research is heavily reliant on centralized funding sources, which often prioritize commercially beneficial outcomes or reinforce institutional biases. Decentralized science creates a more transparent and inclusive research environment by leveraging blockchain and Web3 technologies to decentralize funding, execution, and dissemination.

This article explores how decentralized science empowers independent scientists and communities to regain control over scientific exploration. By examining decentralized funding platforms, data collaboration tools, and community-driven governance models, it highlights the transformative potential of this movement. Through decentralized mechanisms, researchers can gain support for high-risk, unconventional ideas, promote transparent decision-making, and disseminate research findings openly. With the rise of artificial intelligence, collaborative tools, and Web3, decentralized science provides a blueprint for democratizing innovation and accelerating the pursuit of knowledge for societal advancement.

Why Choose Decentralized Science?

Scientific discovery is the process of systematically acquiring new knowledge through iterative hypothesis testing and experimentation. Inductive reasoning enables researchers to generalize scientific conclusions from specific observations and develop principles that can confidently predict outcomes.

Scientific research can be decentralized. Decentralization must begin with funding, as control over financial resources fundamentally determines the direction of scientific exploration. Historically, many great scientists conducted independent research funded either by individuals or patrons, allowing them the freedom to explore without the influence of centralized authorities or corporate interests. Figures like Galileo Galilei (supported by the Medici family), Isaac Newton (who worked primarily independently), and Charles Darwin (who self-funded his evolutionary research) exemplify the impact of decentralized research. Their independence led to groundbreaking discoveries that shaped scientific progress, free from institutional constraints.

Scientific research should be decentralized. In contrast, today's scientific research is highly centralized. It primarily relies on government grants, partially on corporate funding, and is subject to institutional oversight, often determined by a few "gatekeepers" who dictate research topics and limit scientists' autonomy. This centralized funding model introduces significant biases—corporate funding tends to favor commercially beneficial outcomes, undermining objectivity (BMJ, 2014). For example, studies funded by the food industry are 3.2 times more likely to report favorable results (Springer, 2021). While government grants are less susceptible to commercial-related biases, they still often prioritize established institutions and well-known research groups over truly novel or unconventional ideas. Even agencies like the NIH, aimed at reducing reputational biases, cannot fully eliminate these issues. Political and commercial influences continue to shape research priorities, marginalizing high-risk, innovative ideas from emerging researchers.

Scientific research will be decentralized again. Decentralized funding has gained momentum, with initiatives like BIO Protocol and VitaDAO enabling scientists to receive funding directly from the community. This community-supported model provides a viable alternative to traditional funding. Web3 technologies also enhance the liquidity of scientific outcomes and reduce financial risks for independent researchers, allowing them to pursue innovative ideas more freely. Decentralized participation and governance are interconnected aspects of decentralized science. Platforms like Codatta facilitate collaborative data sources, allowing individuals to contribute knowledge in the form of cutting-edge data while sharing risks and rewards. The existence of decentralized governance is essential for providing necessary oversight and ensuring research integrity. It ensures balanced, community-driven decision-making, reducing biases typically present in centralized systems. These aspects collectively foster a more transparent and inclusive research environment. Decentralized dissemination is also crucial to decentralized science. Platforms like ResearchHub help address inherent issues in centralized scientific publishing channels, such as high costs, gatekeeping, and lengthy publication delays, by enabling transparent, community-led publishing and peer review.

The mission of decentralized science is to empower collaborative knowledge creation through community-driven efforts, blockchain, and open collaboration, making research more accessible and unbiased.

• Discover more truths about the universe, free from inherent or systemic biases.
• Lower barriers to entry, allowing talented individuals from unconventional backgrounds to contribute.
• Encourage exploration of suppressed or overlooked scientific directions.

Decentralized science will begin with decentralized funding, but it will not stop there. Distributed contribution credits, transparent funding processes, open-source methodologies, broad community participation, and community-led publishing are crucial for fostering collaboration and inclusivity throughout the research process.

AI-Powered Science: A Major Boost for Independent Scientists

Figure 2: Illustration of the integration of AI research practices in the scientific field (Source: https://ai4sciencecommunity.github.io/)

Artificial intelligence is revolutionizing scientific research, fundamentally transforming the way scientific discoveries are made and workflows are conducted (Toner-Rodgers, 2024). Top scientists worldwide report significant productivity increases through AI integration, including a 44% increase in new material discoveries and a 39% increase in patent applications (Toner-Rodgers, 2024). These early achievements demonstrate how AI enhances efficiency, particularly in data-intensive and time-consuming fields such as materials science, drug discovery, and biology (Nature, 2023).

Figure 3: Scientific research process

AI significantly amplifies individual capabilities, enhancing productivity across the entire scientific workflow. During the ideation phase, AI analyzes vast datasets to uncover patterns and ideas that surpass human cognition (AI4Science, 2023). In the hypothesis formation process, AI optimizes research questions and highlights promising research directions. In experimental design, AI optimizes experimental setups, simulates outcomes, and assists in decision-making. AI-driven robots automate laboratory tasks, bridging the gap between design and execution, while virtual simulations allow hypothesis testing before conducting physical experiments (MIT, 2023). Finally, AI aids in data interpretation, refining results and iterating conclusions for faster, more accurate insights (Nature, 2023).

Figure 4: Collaboration between human scientists and AI (Tony Stark and JARVIS - Marvel movie "Avengers: Age of Ultron")

Human researchers play a crucial role in providing creativity, ethical judgment, and intuition—qualities that AI lacks. While AI excels at data processing and optimization, human researchers can interpret these findings in a broader context, ensuring scientific rigor and ethical standards are maintained. The partnership between AI and human researchers forms a complementary relationship that pushes the boundaries of science. In this collaboration, AI handles complex data tasks while humans provide strategic oversight, creativity, and ethical guidance, making the entire research process more efficient and innovative.

The synergistic effect of human-machine collaboration is reshaping scientific research, accelerating productivity and innovation at an unprecedented pace. Notably, the developers of AlphaFold (a groundbreaking technology for protein structure prediction) recently received the Nobel Prize, highlighting the transformative impact of human-machine collaboration. Human scientists excel at assessing the potential of candidate ideas, effectively filtering out less viable directions and ensuring time and resources are used efficiently. Their heuristic approaches and methodologies can be documented as domain-specific knowledge and enrich the capabilities of AI agents through post-training techniques (such as RAG, prompt engineering, and fine-tuning).

Scientific workflows also involve the use of complex tools, often requiring multiple specialized software tools. The logical workflows defined by scientists—covering inputs, outputs, and objectives for each interaction—can be encoded into expert knowledge segments within AI agents. Projects like TXYZ.ai aim to create universal AI-assisted research tools that integrate these workflows into AI systems, making them more efficient and effective.

As AI continues to accumulate domain-specific knowledge, it will enhance underlying models, enabling relevant systems to process the growing data more effectively. This iterative collaboration between humans and machines forms a self-reinforcing cycle, accelerating research progress and continually pushing the boundaries of human knowledge.

Decentralized Science Landscape: Lightweight Survey

Decentralized science is reshaping the entire research process from funding to dissemination by leveraging blockchain and Web3 technologies. This model decentralizes key aspects of research: funding, execution, and dissemination. The accompanying diagram visualizes this process, highlighting the participants and contributions at each stage.

Figure 5: Decentralized scientific research process

The process begins with fundraising, where independent scientists propose research projects, breaking free from traditionally centralized funding sources that often favor established institutions. In the decentralized science model, research proposals are funded through decentralized supporters, with community-driven contributions playing a significant role. Supporters, guided by community-driven decision-making, review these proposals and allocate resources. This decentralized funding mechanism ensures that even high-risk or unconventional ideas can receive support, bypassing institutional gatekeepers.

Once funding is secured, the next phase is the research process, which includes multiple steps—ideation, hypothesis formation, experimental design, data collection, and analysis. Unlike traditional processes strictly controlled by centralized institutions, decentralized science introduces a more collaborative and transparent workflow. Independent scientists (as illustrated) engage in ideation and hypothesis formation. During the data collection phase, external data creators can contribute to the research, providing incentives to reward high-quality data contributions. Data analysis follows, where analytical results are used for hypothesis testing, forming an iterative approach that continuously refines and evaluates results until meaningful conclusions are reached.

Governance and oversight are another key component. Decentralized supporters oversee and guide projects, providing funding support and ensuring research integrity aligns with community values. This decentralized governance model ensures power is distributed, and all contributions—whether data or expertise—are fairly recognized, as shown in the "Fair Acknowledgment and Contribution" phase in the diagram.

Finally, there is dissemination and impact. The traditional publishing model, often restricted by paywalls, is replaced by community-driven platforms that ensure research findings are publicly accessible. Publications and the resulting intellectual property or outcomes will flow back to decentralized science supporters and the broader community, which can be used to generate further impact and receive appropriate economic returns or credits. This cycle helps acknowledge contributions and create incentives, further fostering a collaborative environment for scientific advancement.

This workflow significantly improves traditional scientific processes by democratizing funding, encouraging interdisciplinary collaboration, and achieving seamless data sharing. Decentralized oversight minimizes bureaucratic inefficiencies, while credit and reward systems incentivize contributors at all stages of research. Ultimately, this approach not only accelerates innovation but also ensures fair recognition and tangible returns for all stakeholders, establishing a sustainable and impactful model for scientific progress.

Survey of Decentralized Science Subfields

Figure 7

This figure illustrates the vibrant and diverse ecosystem of decentralized science, highlighting key subfields and innovative participants reshaping the scientific landscape. Notable projects include BIO Protocol, supported by Binance Labs, and ResearchHub, co-founded by Coinbase's Brian Armstrong, both dedicated to democratizing research funding and publishing. Another prominent project is Pump.Science, whose URO and RIF initiatives have gained momentum.

In the decentralized data collection and collaboration subfield, Codatta stands out as a key player, committed to connecting, collaborating, and co-developing future general artificial intelligence (AGI). Platforms like Data Lake and Ocean Protocol also contribute to collaboration and trust in decentralized data sharing. Additionally, Codatta is an important part of the AI/decentralized physical infrastructure network scientific applications subfield, uniting communities to provide data, samples, and labels (including reasoning) for training AI models. These efforts collectively demonstrate how decentralized science is transforming science into a more transparent, collaborative, and equitable ecosystem for the future.

Overall, decentralized science is reforming the research and licensing landscape, expected to fundamentally change how human civilization reveals truths about the surrounding world, the inner world, and even beyond the current world. However, like the broader Web3 industry, decentralized science is still in its early stages. While decentralized funding is gaining traction and collaborative research shows promise, adoption remains a challenge. The traditional academic system still holds significant influence, necessitating further work to build trust and scale these new approaches.

The overall maturity of decentralized science heavily depends on the progress of the Web3 ecosystem. There is immense potential here, but it requires ongoing technological development, cultural change, and broader acceptance. As decentralized science and Web3 grow, we can anticipate a more open, collaborative, and efficient scientific research landscape.

The Renaissance of Independent Scientists

Figure 7: Pioneers of independent science: Nikola Tesla (left) and Albert Einstein (right)

History shows that many groundbreaking discoveries were made by scientists working outside institutional systems. Innovators like Nikola Tesla, Albert Einstein, and Marie Curie, especially early in their careers, pursued bold ideas with limited institutional support. For instance, Nikola Tesla relied primarily on his income and personal investors for support when he began researching alternating current, rather than formal institutions. Albert Einstein proposed the theory of relativity while working at the Swiss patent office, largely isolated from academic institutions. Marie Curie worked tirelessly with extremely limited resources early in her career, often relying on personal perseverance and donations to advance her pioneering research in radioactivity. These pioneers demonstrate how innovation can thrive without institutional constraints. Over time, scientific discovery became more centralized due to the need for more resources, but today's tools are reversing this trend, reigniting the renaissance of independent science.

Figure 8: Super individuals empowered by AI and Web3, gaining strength through community support (original image from the anime "Naruto Shippuden," depicting Naruto Uzumaki in Nine-Tails Chakra Mode)

Modern technology is empowering individuals to rediscover their roles. Artificial intelligence democratizes data analysis, open-source platforms facilitate collaboration, and Web3 enables decentralized funding through community-driven networks. Decentralized Autonomous Organizations (DAOs) provide financial and technical support for independent projects, bypassing traditional gatekeepers. Combined with accessible research tools, these advancements are creating a new class of "super individuals" capable of independently tackling bold challenges.

Figure 9

This movement is not only pushing traditional boundaries but also opening doors for fields lacking mainstream support that may provide significant insights. For example, research on Unidentified Aerial Phenomena (UAP) has been marginalized but is now gaining legitimacy through decentralized communities that crowdsource resources and data. Similarly, questions about the connection between gravity and electromagnetism are being reexamined free from institutional biases. With community support and cutting-edge tools, independent scientists are ready to explore these unknown territories.

The rise of decentralized science is redefining the way discoveries are made, combining technological empowerment with collective action. Individuals and communities now have the tools and opportunities to democratize the future of innovation. Now is the time to embrace this movement and unleash the full potential of independent research.

References

• BMJ (2014). Bias in industry-funded research. Available from: https://www.bmj.com/industry-bias
• Springer (2021). Industry-funded studies in food sector more likely to report favorable results. Available from: https://www.springer.com/industry-food-bias
• Toner-Rodgers, A. (2024). Artificial Intelligence, Scientific Discovery, and Product Innovation. MIT Press.
• AI4Science (2023). AI's Role in Advancing Scientific Research. Available from: https://ai4sciencecommunity.github.io/
• Nature (2023). Scientific Discovery in the Age of Artificial Intelligence. Nature Publishing Group.
• MIT (2023). AI's Impact on Research Workflows. Available from: https://mitpress.mit.edu/ai-research