It’s no secret that crypto-centric data breaches have skyrocketed recently, and this trend is likely to increase for the foreseeable future, especially as cybercriminals continue to employ more sophisticated techniques to facilitate their attacks.

Up to this point, the losses emanating from various cryptocurrency hacks increased by approx. 60% during the first seven months of the year, driven, in large part, by the theft of funds from various decentralized finance (DeFi) protocols.

An AI response

During October 2022 alone, a record $718 million was stolen from DeFi protocols in 11 different hacks, sending cumulative hack-related losses over the $3B mark. Now many experts believe that artificial intelligence (AI) and machine learning (ML), the latter being a subset of the former, could help alleviate many of today’s most pressing cybersecurity problems.

An essential piece of the puzzle?

ML-based privacy systems are designed to learn and calculate a project’s regular network activity and subsequently detect and identify suspicious movement. There are two types of ML systems that can be used: supervised ones that can learn to generalize from past attacks, and unsupervised ones that can detect unknown attacks, alerting cybersecurity personnel to any deviations from the rule.

In fact, ML-ready technologies should become a crucial component of threat detection and defense in the burgeoning web3 industry, keeping bad actors at bay in an automated fashion.

The total capitalization of the AI ​​cybersecurity market (of which ML is a major component) is projected to grow at a compound growth rate (CAGR) of 23.6% over the next five years, reaching a cumulative total of $46.3 billion for 2028.

From a technical perspective, ML systems allow security experts to quickly identify problems, use more data sets than is possible with simple human accounting, and allow them to design systems that are free from innate bias. In other words, they can augment older heuristic processes, making them more efficient and error free.

As a result, it becomes easier for platforms to respond to hacking incidents long before the problem escalates. In fact, when ML platforms detect and identify malicious activity within a web3 system, they can automatically block a malicious entity from exploiting a protocol. Forta, for example, is a decentralized monitoring network capable of detecting threats and anomalies in DeFi, NFTs, governance, bridges, and other web3 systems in real time.

existing challenges

Most ML platforms are driven by data scientists, and herein lies one of the key challenges when it comes to implementing this technology within the world of cybersecurity. While web3 has attracted a lot of developers, it has not been able to attract a lot of data scientists so far.

This is unfortunate, as there is so much data available for analysis, opening the door to many research opportunities to solve real-world problems. In this regard, the industry needs to make web3 more attractive to data scientists, which can be done by educating that cohort about the underlying technology and providing incentives to make this niche more attractive.

Crypto Sleuth ZachXBT exposed cheating for a grateful DeFi community

Deeply vetted surveys of questionable projects proved ‘invaluable’ in Mad Year

The vast majority of data science engagement in the cybersecurity ecosystem revolves around identifying attacks and suspicious activity on the chain. While these models encompass important elements such as anomaly detection, time series analysis, and supervised classifiers, there are still more opportunities to be developed that extend beyond monitoring.

There are many ways that ML can make today’s cybersecurity systems more secure and reliable. For example, it can be used to detect third-party threats and anomalies, identify irregular patterns, kill bots, orchestrate existing platform security protocols, and behavior analysis.

Here are some of the main impacts that the aforementioned technologies have on current cybersecurity frameworks:

Efficient Vulnerability Management

Most cryptographic protocols cannot keep up with vulnerabilities that emerge on a daily basis. While conventional vulnerability management techniques are designed to respond to incidents after hackers have exploited a particular loophole, machine learning systems can automatically identify vulnerabilities.

ML-powered behavioral analytics tools can analyze the behavior of digital asset users across various transactions, allowing them to detect anomalies that point in the direction of an unknown attack. As a result, protocols can safeguard their stock even before a problem is reported and fixed.

Over time, ML-enabled technology may even be applied in the context of platform auditing and monitoring, and the technology is used for the development of graph-based algorithms, embedded deep learning systems, and mechanisms. of reinforcement learning.

Faster detection of external threats

Most traditional security systems use attack signature-based indicators to identify individual threats. While this method is very effective at highlighting previously discovered problems, it is not very effective at eliminating problems that have not yet been found.

That being said, when traditional attack signature indicators are linked with ML, detection of potential threats can be significantly increased while minimizing false positives.

SOL is up 36% on the week as Bonk Memecoin makes waves

NFT Marketplace Magic Eden generated record volume in December

Machine learning is known for providing users with excellent forecasting capabilities and efficient data analysis methods, which are essential for optimizing blockchain mechanisms. Not only that, these properties are even more useful when it comes to improving a blockchain’s native data verification procedures, malicious attack detection, and faster identification of fraudulent transactions.

Looking to the future

As cyber attacks become more sophisticated, machine learning can help make projects more prepared for external threats. Using the right systems, organizations can not only detect and respond to hacking attempts in real time, but also take corrective action before a threat becomes serious.

Still, AI/ML technology is not a panacea for today’s cybersecurity problems. Rather, the technologies should sit alongside the expert systems, making the ecosystem more secure. As we move towards a more decentralized future, it will be interesting to see how these new technological paradigms will evolve.

Christian Seifert, A former Microsoft web security specialist, he is a security researcher in the Forta community.