Former Central Intelligence Agency chief technology officer Bob Flores has warned that artificial intelligence must be developed with robust security measures from the outset to avoid repeating the mistakes made during the creation of the Internet. Speaking at a cybersecurity conference in Tel Aviv, Flores emphasized that the lack of early security protocols in the Internet's development has led to ongoing vulnerabilities, which should serve as a lesson for current AI innovation.

Flores outlined a range of emerging threats posed by artificial intelligence, including the rapid creation of malware, infiltration of financial and security institutions, and vulnerabilities caused by data poisoning and supply chain tampering. He stressed the importance of integrating security frameworks and validation mechanisms into AI systems during their initial development stages, rather than attempting to retrofit protections years later.

While acknowledging that AI can also be harnessed to improve digital security, Flores called for the adoption of modern governance frameworks and standards to ensure consistent and effective defense capabilities. Looking ahead, he identified quantum computing as a future challenge that will require new approaches to AI security. Flores concluded by urging developers to prioritize security and data integrity from the very beginning, warning that neglecting these aspects could have serious consequences.