National cybersecurity requires proactive strategies that integrate defensive AI and continuous monitoring to protect critical infrastructure. Beyond the technical impact on companies, the automation of attacks increases the volume of fraud targeting common users. Prevention becomes critical, demanding not only identity-based security controls but also greater employee training to recognize the signs of algorithm-driven attacks.
A troubling precedent occurred at the end of 2025, when the first global attack was documented where AI executed between 80% and 90% of the tactical operations with no substantial human intervention, affecting 30 international organizations in the finance and government sectors. This automation drastically reduces costs and execution times for attackers, allowing them to target critical sectors on a massive scale.
For Panama, where digitalization is accelerating in areas such as health, commerce, and education, this landscape represents a high risk. Experts emphasize that criminals can now accurately imitate communications from financial institutions and government authorities. Ramón García, an executive at Palo Alto Networks for the region, emphasized that the emergence of models designed for malicious purposes lowers the barriers to entry for criminals.
Researchers point out that tools like WormGPT or FraudGPT are marketed on dark web forums under monthly subscription models that include customer support. Cybercriminals use unrestricted AI models to scale massive attacks. The autonomy of AI in cybercrime has moved from a hypothesis to a reality after detecting global attacks that operate up to 90% without human intervention.
