Attackers are beginning to use AI agents to automate parts of the hacking process itself.
A recent campaign against online retailers provides a useful real-world example. Researchers reported that one operator used three AI-agent frameworks to support vulnerability discovery, exploitation and attack orchestration.
During one five-day period in September 2026, 105 attack projects were launched and at least 27 organizations were compromised to varying degrees. More than 600,000 unexpired payment-card records were stolen from two victims.
For anyone learning cybersecurity, this case is important because it shows how the skills required for defence are evolving.
What Is an AI Agent in Cybersecurity?
A traditional chatbot mainly responds to prompts. An AI agent can go further.
Depending on how it is configured, an agent can use tools, run commands, interpret results, choose its next step and continue working toward an objective.
In cybersecurity, legitimate teams can use agents for tasks such as investigation, testing and vulnerability analysis.
Attackers can misuse the same concept.
An agent might scan a target, identify an exposed service, analyse a vulnerability, attempt exploitation and collect useful information before a human operator becomes directly involved again.
Why Automation Changes the Threat Model
Cybercriminals have used automation for decades.
What makes modern AI agents different is their ability to interpret changing information.
Traditional automation normally follows predefined rules. An AI-enabled workflow can potentially analyse output and decide what to try next.
That makes attacks more adaptable.
For defenders, this means internet-facing systems may be tested more frequently and weaknesses may be exploited sooner after discovery.
Students and junior security professionals should therefore stop thinking of vulnerability management as simply “running a scanner.”
The real challenge is understanding which vulnerabilities are exploitable and how quickly the organization can respond.
Skills That Matter in the AI-Attack Era
The rise of AI-powered cyberattacks does not reduce the importance of cybersecurity fundamentals. It increases it.
Start with networking. You need to understand how systems communicate before you can identify abnormal behavior.
Learn Linux and Windows administration. Attack activity often becomes clearer when you understand normal operating-system behavior.
Study web application security. The recent card-theft campaign makes concepts such as authentication, APIs, sessions, cloud configuration and client-side scripts particularly relevant.
Develop log-analysis skills. Automated attacks can generate large volumes of activity, and defenders need to recognize patterns across endpoint, identity, web and cloud telemetry.
Learn identity security. Stolen credentials and excessive privileges can turn a small compromise into a major incident.
Learn some scripting too. Python, PowerShell or shell scripting can help automate defensive tasks and explain how attackers automate theirs.
Do Not Become Dependent on AI
Cybersecurity learners should use AI tools, but carefully.
Ask AI to explain a log entry, create a practice scenario, compare technologies or help troubleshoot a lab. Then verify the answer.
If you allow AI to complete every technical task, you may learn how to prompt without learning how the underlying technology works.
That is dangerous in security, where incorrect assumptions can have real consequences.
The Opportunity for Defenders
AI agents may give attackers greater speed, but defenders can use automation too.
Security operations teams can use AI to summarize alerts, enrich investigations and correlate events. Application security teams can improve testing and triage. Vulnerability teams can prioritize findings more intelligently.
The professionals who benefit most will not choose between “traditional cybersecurity” and AI.
They will understand both.
AI-powered attacks are already moving from theory into real incidents. Learn the fundamentals deeply, understand how automation changes attack workflows and become comfortable using AI without surrendering your judgment to it.
The tools will change.
Your ability to understand systems, evaluate evidence and make sound security decisions will remain essential.
To know more about Anand Shinde and his work in cybersecurity, awareness, and books:
https://anandshinde.com/
If your business needs cybersecurity services, AI security guidance, cloud security review, or protection against modern cyber threats, visit DevOM Consulting:
https://devomconsulting.com/
