Security
AI Now Powers Over Half of Cyberattacks in Africa
Interpol's 2026 assessment reveals criminals operate at machine speed while government defenses lag behind.
Key takeaways
- AI now enables more than 50% of cyberattacks across Africa, automating every stage from reconnaissance to evasion
- Ransomware attacks account for 28% of African cybercrime, with South Africa and Nigeria dominating business email compromise operations
- Criminals breached a North African government database using AI tools, extracting 300,000 identity records and data on 500,000+ companies
- Africa's 1.1 billion mobile subscribers face growing risk as digital expansion outpaces regulatory frameworks and law enforcement AI readiness
- 45 African governments have adopted e-government platforms, expanding both service delivery and vulnerability to AI-powered attacks
Artificial intelligence has crossed a threshold in Africa's cybersecurity landscape: It now enables more than half of all cyberattacks on the continent, according to Interpol's 2026 Africa cybercrime assessment.
The shift represents a fundamental change in how cybercrime operates. AI has evolved from a supporting tool into the primary engine driving attacks, allowing criminals to automate every phase of an operation — from initial reconnaissance through phishing, extortion, and evasion tactics.
"Criminals are now operating at machine speed, with AI enabling attacks that are faster, more scalable, and more difficult to attribute," Interpol researchers wrote in the report.
Why it matters
With over 1.1 billion mobile internet subscribers across Africa as of 2025, the continent's digital expansion is outpacing both regulatory frameworks and digital literacy programs. This gap leaves a massive population vulnerable to AI-generated scams and influence operations at precisely the moment when 45 African governments are adopting e-government platforms to improve service delivery. The collision of rapid digitization with industrialized cybercrime creates systemic risk for both citizens and institutions.
How AI transforms cybercrime tactics
The technology's impact appears across multiple attack vectors. In Central Africa during 2025, criminals increasingly deployed AI-generated synthetic identities to bypass "know your customer" verification systems used by banks and businesses. These same tools proved capable of fooling biometric security systems.
Business email compromise attacks show particularly sophisticated AI integration. South Africa originated 70% of these attacks, with Nigeria accounting for another 29%, according to Interpol data. AI now generates correspondence that mimics executive communication styles, internal terminology, and signature patterns with near-perfect accuracy.
These deceptive messages trick recipients into opening malicious attachments or links. Ransomware attacks — frequently the end result of successful BEC campaigns — represented 28% of all African cyberattacks in 2025, making them the single largest attack category.
Real-world breach demonstrates scale
AI company Anthropic disclosed that cybercriminals used its Claude agent to breach an unidentified North African government database. The attackers extracted 300,000 national identity records and data from more than 500,000 companies operating in that country.
Anthropic's separate report confirmed that criminals are leveraging its technology for writing malicious code, conducting surveillance, and executing online influence operations at scales that exceed government detection capabilities.
The enforcement gap
"Cybercriminality has evolved from isolated incidents into an industrialized, borderless ecosystem," Interpol stated. "However, cybercrime legislation is fragmented and AI readiness in law enforcement agencies remains alarmingly low."
Ghana, Mauritius, South Africa, and Tunisia rank among African nations with the highest adoption of e-government functions. While these platforms improve efficiency and citizen access, they simultaneously expand the attack surface for AI-driven cybercrime without corresponding security investments.
Researcher Karen Allen of the Institute for Security Studies noted that as more countries implement digital government services, "exposure to AI risks will increase unless safety controls are ramped up."
Interpol's assessment emphasizes that the human element remains the weakest defensive link. "As AI continues to refine the realism of impersonation, the human element remains the weakest link — and the most difficult to defend," the report concluded.
These findings were first reported by Africa Defense Forum, drawing on Interpol's 2026 cybercrime assessment and related security research.