Cloudflare announced Defensive AI, a personalised approach to securing organisations against the new wave of risks presented by emerging technology. Threat actors have begun to successfully test the limits of AI-enhanced attacks, using the power of AI to launch sophisticated phishing scams, help write malicious code and supercharge attacks on critical business functions. Leveraging the unique traffic patterns of an organisation, Cloudflare’s Defensive AI keeps defenders a step ahead of threat actors, by providing tailored mitigations that enable protection of critical applications and entire networks.
The crown jewels and trade secrets of organisations vary, and to protect them, organisations require highly agile security strategies. While AI holds tremendous potential for organisations, it also allows threat actors to rapidly increase their effectiveness and renders one-size-fits-all security offerings obsolete. Adopting a personalised approach to security – looking at an organisation’s own traffic patterns – has the power to shift the cybersecurity balance back in favour of defenders.
“Fighting AI with AI is now a non-negotiable. And a personalised approach to protect data and defend against complex threats unique to an organisation’s attack surface, at-speed, and scale, is paramount,” said Matthew Prince, CEO and co-founder at Cloudflare. “By understanding ‘normal baselines’ in a customer’s environment and mitigating the threats that will move the needle towards increased resilience, Defensive AI is the crucial edge defenders need to stay ahead of today’s adversaries.”
With Defensive AI, Cloudflare’s AI models look at a specific customer traffic patterns, providing that organisation with a tailored defence strategy unique to their environment, allowing customers to:
* Protect the Modern Web: APIs comprise 57% of all dynamic traffic on the web and underpin the most popular apps and services for businesses. Cloudflare is developing API Anomaly Detection, which will help prevent attacks designed to damage applications, take over accounts or exfiltrate data. It will leverage AI to learn the behaviour of an application and build a model of what a sequence of good requests over time looks like. The resulting traffic model is used to identify attacks that deviate from the “normal” behaviour – acting as a guardrail to help stop potentially malicious activity.
* Secure the Number One Threat Vector: Email: Nine out of every ten cyber attacks begin with a phishing scam, so reducing the risk presented by email is pivotal to upholding cyber resilience. Cloudflare’s Cloud Email Security solution stays ahead of threat actors by training AI models to identify different parts of a message and flag suspicious content. The rise of AI-enhanced attacks are making traditional email security providers obsolete, as threat actors can now craft phishing emails with little to no language errors. But while trends like Generative AI continue to evolve, Cloudflare’s models analyse all parts of a phishing attack – most of which are harder to fake.
* Mitigate threats posed by employees: whether accidental or on purpose: Almost half of insider threat incidents involve an employee with privileged access to company assets – underscoring the importance of a Zero Trust approach. Cloudflare Gateway will allow customers to create a baseline of their organisation’s user behaviour and resources being accessed, to flag or filter what presents as risky or unauthorised. This will provide both a score for users themselves and resources, both those that are managed internally and external resources an organisation reaches out to.
Cloudflare has reimagined security to leverage AI in the fight against AI. As emerging technology evolves and new tools are developed, Cloudflare is positioned to allow organisations to embrace these productivity enhancements without exposing themselves to malicious use cases.