Barbadian cybersecurity strategist Niel Harper has raised an alarm internationally about the ongoing threats posed by the professionalisation of cybercrime and the emerging security risks introduced by autonomous artificial intelligence (agentic AI).
He framed the current threat landscape not as a technical problem, but as a market failure, where the profitability of ransomware as a Service (RaaS) groups continues to outpace enterprise defense spending.
Harper, who has worked in executive roles in global organisations such as Bermuda Commercial Bank, Canonical, INTERPOL, and the United Nations, was speaking recently at Black Hat Middle East & Africa MEA conference
Black Hat MEA, the largest and most significant cybersecurity and hacking conference in the Middle East and Africa region, is held annually in Riyadh, Saudi Arabia.
As a VIP speaker at the conference, Harper, managing director and digital trust practice leader, Octave Cyber Security Group, delivered a number of key addresses at the event, issuing several warnings to businesses.
He focused extensively on the financial due diligence model employed by RaaS affiliates, urging corporations and law enforcement to enhance their cooperation.
“The RaaS threat needs to be addressed through robust disaster recovery, enhanced incident response, and improved methods for freezing and seizing funds,” he stated, emphasising the need for mandated security controls to disrupt the cyber insurance payout mechanism.
In a specialised “deep dive” session, Harper addressed the security challenge of Agentic AI. He detailed why traditional security layers fail against autonomous agents, calling for robust, layered controls to manage their high velocity and capacity for adaptive, independent decision making.
“When you empower AI with autonomy, you simultaneously multiply its potential blast radius,” he cautioned.
Harper recommended that organisations immediately implement zero trust principles for all AI agents, requiring just-in-time access and frequent authentication as well as layered guardrails to prevent malicious or misaligned AI agents from inflicting harm on critical financial systems or IT infrastructure.
He underscored the necessity for global enterprises to adopt proactive governance models that integrate security, compliance, and product go-to-market strategies at the highest executive levels. (SC/PR)
