Singapore – Cloudflare has introduced Adaptive Intelligence, a continuous detection engine built into its Bot Management platform, aimed at tackling the growing wave of automated cyberattacks.
The announcement was made in Singapore on 1 September 2026, positioning the technology as a direct response to how accessible AI-powered attack tools have become.
Unlike static security systems, Adaptive Intelligence draws on real-time data from more than a trillion web requests processed daily across Cloudflare’s global network.
The engine autonomously generates short-lived rules based on live traffic patterns, with the aim of making sustained automated attacks costlier and less viable for threat actors to run.
Cloudflare said the tool addresses a structural imbalance in modern cybersecurity, whereby attackers can launch cheap, repeated attempts, while defenders must block every threat instantly without disrupting genuine users.
According to the company, launching an automated attack once demanded advanced technical expertise, but the spread of AI tools and low-cost online services has considerably lowered that barrier.
Cloudflare noted that attackers can now rent networks of compromised devices, disguise their origin using genuine residential IP addresses, and deploy free software capable of mimicking human behaviour online.
Meanwhile, traditional security tools remain reliant on scheduled updates that can take weeks or months to roll out, widening the gap between attacker agility and defender response times.
“Building taller walls fails when the cost of scaling an attack is effectively zero. To stop modern bot threats, you have to flip the economics on the attackers. Traditional defenses offer a static target that threat actors can systematically solve”, said Dane Knecht, CTO at Cloudflare.
Cloudflare’s Adaptive Intelligence creates a moving target—rendering an attacker’s engineering efforts obsolete before their operation can ever achieve scale”, Knecht added.
In contrast to conventional bot mitigation tools that depend on fixed rule updates, Adaptive Intelligence operates as a continuously learning system, retraining itself on live traffic to identify emerging threats as they surface.
Cloudflare outlined several capabilities the platform now brings to customers, beginning with what it describes as “always-on defence.”
The machine learning model retrains continuously on live traffic, folding new bot frameworks and bypass techniques into its detection engine in real time, rather than waiting for manual updates or scheduled releases.
The system also generates short-lived, hyper-targeted rules aimed at specific threats, with rotating rules designed to stop attackers studying the system, mapping its defences, or making lasting progress against it.
Furthermore, Adaptive Intelligence analyses behaviour across multiple timeframes simultaneously, a feature built to catch stealthy, low-and-slow attacks such as slow credential stuffing and scraping campaigns that typically slip past conventional defences.
To distinguish malicious activity from genuine human behaviour, the platform combines browser-level session signals from Cloudflare Precursor with global edge telemetry, moving beyond simple binary detection tests.
Security updates, meanwhile, are tested automatically against live traffic before deployment, a process intended to verify accuracy and reduce false positives without causing downtime.
Cloudflare positions Adaptive Intelligence within its broader Bot Management suite, reflecting the company’s ongoing efforts to keep pace with the increasing sophistication—and accessibility—of automated cyber threats.
