A new framework combining machine learning and blockchain technology could transform how engineers protect computational systems from security breaches, according to research published this month in the journal Engineering.
The study introduces Machine Learning on Blockchain (MLOB), an innovative approach that addresses a significant vulnerability in current systems that integrate artificial intelligence with blockchain networks.
While machine learning has been rapidly adopted across engineering fields to boost accuracy and efficiency, current blockchain solutions primarily focus on data security, leaving the actual computational processes vulnerable to manipulation—a gap the researchers aimed to close.
“The research objective is to develop a novel ML on blockchain framework to ensure both the data and computational process security,” write the authors, Zhiming Dong and Weisheng Lu, in their paper. “The central tenet is to place them both on the blockchain, execute them as blockchain smart contracts, and protect the execution records on-chain.”
Unlike existing systems that typically run machine learning models in off-chain environments where they remain susceptible to tampering, the MLOB framework embeds both the data and computational processes on the blockchain itself. This creates a secure, traceable environment for sensitive engineering computations.
Security Without the Performance Penalty
The researchers tested their framework against multiple attack scenarios, comparing it with conventional approaches. Their testing revealed that the MLOB framework successfully defended against all six attack scenarios designed to compromise the system, outperforming all baseline methods.
Perhaps most surprisingly, the enhanced security didn’t come with significant performance costs. The team’s evaluations showed only a negligible 0.001 difference in accuracy compared to traditional methods, with a latency increase of just 0.231 seconds per computational task.
“The key finding is MLOB can significantly enhance the computational security of engineering computing without increasing computing power demands,” the researchers note. “This finding can alleviate concerns regarding the computational resource requirements of ML–BT integration.”
For industries where computational security is paramount—such as critical infrastructure, quality assurance systems, or engineering forensics—this marginal efficiency trade-off could be well worth the substantial security improvements.
Real-World Applications
To demonstrate practical applications, the team implemented their framework in a construction industry scenario, using it to monitor indoor construction progress by comparing as-built conditions with planned models.
“The construction progress monitoring process and its results are highly relevant to progress payments and quality accountability assurance,” the researchers explain, noting that traditional monitoring is “susceptible to potential threats or interference that can compromise the accuracy of the progress estimation process.”
In this real-world test, the framework maintained high accuracy while providing verifiable security throughout the computational process.
The system works by first acquiring and training a machine learning model for a specific task. This model is then converted to a format compatible with blockchain deployment, securely loaded into the blockchain, and finally executed through a consensus-based process that ensures computational integrity.
Balancing Security, Accuracy, and Efficiency
The researchers frame their approach through what they call a “balance triangle” of security, accuracy, and efficiency—acknowledging that improvements in one area typically require trade-offs in others.
“While such framework enhances computing security, it may incur costs in terms of efficiency and accuracy due to the occupied computing resources,” they write. However, their evaluations indicate that “this rebalancing maintains satisfactory efficiency and accuracy levels.”
This framework arrives at a critical moment as engineering fields increasingly rely on artificial intelligence for complex tasks while simultaneously facing growing security threats.
Industry implications extend beyond technical improvements. The researchers suggest the framework could drive innovation in engineering practices by integrating advanced technologies, potentially leading to “more competitive engineering operations, increased productivity, and the attraction of talent interested in cutting-edge technologies.”
Future Development
The system isn’t without limitations. The current implementation offers limited support for latency-sensitive applications and lacks a user-friendly interface—both areas targeted for future development.
“To enhance the efficacy and user-friendliness of MLOB, future efforts should be directed towards expanding the platform,” write the researchers, who aim to optimize efficiency further and develop a more accessible user interface.
For organizations concerned about computational security in engineering applications, the researchers recommend starting with pilot explorations, comprehensive training programs, ongoing performance monitoring, and iterative improvements based on real-world feedback.
As machine learning continues its rapid integration into critical engineering systems, securing not just the data but the computational processes themselves will likely become increasingly crucial. The MLOB framework represents a significant step toward addressing these emerging security challenges.
If our reporting has informed or inspired you, please consider making a donation. Every contribution, no matter the size, empowers us to continue delivering accurate, engaging, and trustworthy science and medical news. Independent journalism requires time, effort, and resources—your support ensures we can keep uncovering the stories that matter most to you.
Join us in making knowledge accessible and impactful. Thank you for standing with us!