From lab to ledger: Human keys secure scientific integrity | Opinion

by

As a researcher with a PhD in Neuroscience and a background in crypto, I find myself at the crossroads of two rapidly evolving fields – science and technology. The advent of AI, while promising in many ways, poses significant challenges to the integrity of data, identity, and reputation verification. If left unchecked, it could lead to an era of misinformation, stalling scientific innovation and potentially even degrading human creativity and innovation.

The rapid advancement of AI surpassing regulatory measures could potentially jeopardize data integrity, personal identity verification, and reputations, and if not addressed promptly, it might escalate the spread of misinformation and hinder the pace of scientific advancements. The drive towards highly intelligent AI by its ardent advocates is likened to a quest for a new golden age in science. Yet, this pursuit may lead to an existential risk, where our society could stagnate on a technological plateau due to widespread adoption of immature AI technology, eventually diminishing human creativity and innovation over time.

This is a contradictory take to most accelerationists. AI is supposed to increase our ability to do work faster and synthesize greater amounts of information. However, AI cannot replace inductive reasoning or the experimental process. Today, anyone can use AI to manufacture a scientific hypothesis and use that as input to generate a scientific paper. The results of products like Aithor often appear authoritative on the surface and may even pass peer review. This is a big problem because AI-generated texts are already being curated as legitimate scientific findings and often include bogus fabricated data to support their claims. There is a major incentive for young researchers to utilize whatever means they have available to compete for a limited number of academic jobs and funding opportunities. The current incentive system in academia rewards those who can publish the most papers, whether or not those papers describe legitimate findings—they just need to pass peer review and obtain enough citations. 

Trustworthy academic content is crucial for industries that rely heavily on basic science for their research and development, as this work underpins our society’s functionality and enhances the quality of life for an expanding global population. Unfortunately, when the authorship of such academic content is questionable, it can create significant challenges. Consequently, well-resourced research and development (R&D) organizations can only rely on studies they can conduct and replicate themselves, thereby boosting the importance of trade secrets and potentially hindering open science and access to valuable information.

It’s not just about costly measures against misinformation; the issue at hand is a widespread erosion of trust in knowledge itself. Unverified assertions and unclear attributions are undermining scientific progress, posing a risk to the scientific community. What we urgently require now is a system that can authenticate content and data accurately, creating a truth-based information ecosystem.

AI systems are as powerful as the data they are trained on 

Large language models are excellent tools for generating convincing content; however, they are only as informative as the data on which they are trained. Their ability to extrapolate outside of the training set still remains limited. The role of science isn’t just to synthesize existing knowledge but to create new informative artefacts that increase the entropy of the collective corpus of knowledge amassed by humanity. Over time, as more people use AI to generate content and fewer people generate original content, we’ll face a “low-entropy bloat” that does not introduce new information to the world but rather just recombines past knowledge. Primary sources will become lost as new “knowledge” is based on self-referential AI-generated content unless we build a resilient provenance and verified attribution layer into AI tools used for serious research.

“Simplifying the complexity of human knowledge through what’s analogous to ‘lobotomization’ could have long-term effects on various fields, including medicine, economics, academia, and arts. Misleading data can distort findings, causing significant issues in policy and technology development that undermine the credibility of scientific research. The potential hazards associated with AI-driven science are numerous. Routine scientific work may grind to a halt due to disputes over authorship, accusations of plagiarism, and flawed peer review. We’ll likely have to invest more resources to address the repercussions of deteriorating quality and reliability in research.

Artificial Intelligence (AI) serves as a valuable resource for sparking ideas, organizing thoughts, and streamlining repetitive duties; it should function as an addition to content created by humans rather than a substitute. It’s essential that AI isn’t used to write scientific papers presenting new discoveries without the necessary work but instead as a tool to enhance the productivity and precision of human-led initiatives. For instance, AI could be beneficial in executing simulations on existing data using established methods, automating this process to aid in identifying potential research avenues. Nonetheless, the experimental designs and innovative thinking essential for scientific exploration can’t be effortlessly replaced.

Building a truth-based economy

In a truth-centric economic system, we create a reliable structure using technologies and guidelines to uphold data’s truthfulness, security, openness, and trackability. By addressing the necessity for trustworthiness and verifiability across our digital world, this system enables people and institutions to have confidence in the correctness of the information they share. The worth is derived from the credibility of assertions and the authenticity of evidence and initial sources. This truth-centric economy aims to make digital knowledge as dependable as Bitcoin made traditional currency, which is the vision of the decentralized science movement.

How can we reach our destination? Let’s begin by focusing on the cornerstone of the scientific community: the researcher and their work. Presently, web standards for establishing a scientist’s identity are inadequate when it comes to confirming claims about identity and the authenticity of work. As it stands, it’s relatively simple to create a profile with a respectable standing; peer reviews are also vulnerable due to potential bias and collusion. Without verifying the data accompanying a scientific claim, we can’t establish a trust-based economy for science.

Enhancements in academic identification norms could kick off by implementing a universal login system across platforms, backed by secure identity validation tech that prioritizes privacy. This way, users can log into any platform with the same credentials, confirm their authenticity, and choose what information about their reputation, data, or interactions with other users they wish to share.

At the core of DeSci lies an identity structure grounded in a researcher’s verifiable reputation. This foundation aims to establish a comprehensive, on-chain scientific economy open to both public and anonymous contributors. By facilitating massive online collaboration for research activities, this economy encourages participation from all corners.

Safeguarding the future of human progress

To build a reliable foundation for truth, we must ensure information is transparent and thoroughly validated to maintain trust in our research communities. The longevity of our collective growth, with future breakthroughs in fields like materials science, biotechnology, neuroscience, and complexity science, depends on the preservation of high-quality research and accurate data. This distinction between us and pre-enlightenment societies will define whether we progress or stagnate, potentially leading to intellectual decline for our species. It remains uncertain if DeSci can save us, but time is running out to make things right.

Shady El Damaty

Shady El Damaty, who hails from the Holonym Foundation, is on a mission to find a universal approach to personhood and secure digital entryways using a decentralized identity method that leverages the power of zero-knowledge proofs. In the year 2020, he established OpSci, the initial decentralized science, often abbreviated as DeSci, organization. Before diving into the world of cryptocurrency, Shady obtained his PhD in neuroscience from Georgetown University, Washington D.C., United States.

Read More

2024-12-22 16:14