OpenAI says its AI can hack Linux on a large scale: what does this mean for crypto?

OpenAI, the company behind ChatGPT, has discovered hundreds of security vulnerabilities in Linux using artificial intelligence. In addition, the AI system was able to develop working attacks that can completely take over computers in certain cases.

The discovery also raises questions about the security of Bitcoin and Ethereum, as many systems within the crypto sector run on Linux.

OpenAI announced that its AI system has found no fewer than 948 vulnerabilities in the core of Linux. This operating system is used worldwide for servers, data centers, and other critical digital infrastructure.

For every vulnerability found, the AI ​​developed a working attack in a controlled test environment. This allowed researchers to demonstrate how security flaws can be exploited. In certain cases, someone with limited access to a computer can thereby gain full administrator privileges.

The costs are particularly noticeable. The full investigation cost OpenAI about $1.2 million. That amounts to an average of only $1,279 per vulnerability for which a working attack was developed.

According to OpenAI, similar discoveries at security competitions could theoretically yield between $18.9 million and $66.3 million.

The company emphasizes that the research is intended to uncover security issues before hackers can exploit them.

That is why the AI also developed proposals to remedy the vulnerabilities found. Human security researchers check the results before sharing them with Linux developers.

The discovery is also relevant to the crypto sector. Many computers that keep blockchain networks running use Linux. This applies, among other things, to a large proportion of the computers that check transactions on the Ethereum network.

A vulnerability in Linux can therefore also pose risks to crypto platforms and other services that depend on the operating system.

This does not mean that Bitcoin (BTC) or Ethereum (ETH) itself is vulnerable. An attack on an individual computer is different from an attack on the entire blockchain network.

Mudit Gupta, technical director of Polygon Labs, points out that decentralized networks can better withstand these types of attacks. They consist of large numbers of separate computers. When one system is affected, the rest of the network can continue to function.

Meanwhile, OpenAI warns that AI makes discovering and exploiting security vulnerabilities significantly cheaper. This may also enable malicious parties to detect vulnerabilities more quickly.

The research results come from OpenAI itself and have not yet been fully independently verified. Also, the AI model used is not yet publicly available.