Copperhelm today emerged from stealth to launch a platform that aggregates cloud security data to enable its artificial intelligence (AI) agents to autonomously monitor cloud environments, investigate threats and automatically remediate issues in real-time.
Fresh off raising $7 million in funding, Copperhelm CEO Shimon Tolts said the company has developed a Context Lake that normalizes cloud security data in a way that enables AI agents to perform those tasks. The Copperhelm platform includes specialized AI agents that perform network analysis, analyze system behavior, simulate attacks and automatically mitigate issues. The Copperhelm agents connect directly to live workloads, inspect active processes and container images, map cloud network topology and deploy, for example, a web application firewall (WAF) if needed, without any downtime being required.
In general, cloud computing environments are highly complex and fragmented, making it difficult for AI tools to access and understand the context needed to ensure security is maintained. In organizations that have hundreds of cloud accounts, there needs to be a context engine that organizes all the metadata and configuration information that enables an AI agent to perform specific security tasks, said Tolts.
Armed with those insights, it then becomes possible to deploy a series of AI agents that collaboratively perform security functions spanning discovery to remediation, noted Tolts. That closed-loop approach makes it possible to manage cloud security at the level of scale that will be required to cope with the tsunami of vulnerabilities that will be discovered in the AI era, he added.
That tsunami is being driven first by AI coding tools that are generating more vulnerabilities faster than ever and more advanced AI models that are capable of discovering what are likely to become thousands of new zero-day vulnerabilities in existing legacy systems. Once discovered, it now takes less than a day for cybercriminals using AI tools to create an exploit, noted Tolts.
While humans will still be needed to supervise AI agents, it’s not going to be feasible for cybersecurity teams to respond to issues that are occurring with greater frequency at machine speed. In effect, cybersecurity teams are now caught up in an AI arms race they can only win by investing more in AI to thwart cyberattacks that, thanks to AI, are only going to increase in volume and sophistication, said Tolts.
The challenge, of course, is explaining to business and IT leaders why the bulk of previous cybersecurity investments are rapidly becoming obsolete. While the total cost of cybersecurity might decline in the age of AI as more functions are automated, there is still going to be a need for an initial investment in new tools and platforms.
Hopefully, AI will benefit defenders more than attackers, but in the meantime, cybersecurity is in a state of flux. Unfortunately, it may yet require a few high-profile cyberattacks enabled by AI to occur before business leaders fully appreciate how the scope of threats facing the organization has fundamentally been forever changed.
Recent Articles By Author