How AI Is Used in Cybersecurity—and the Risks It Introduces
Summary
An overview of AI-enabled cyber threats and defensive uses, including phishing, generative AI risks, and how different AI methods can support layered security. Darktrace promotes its own platform and vendor-selection resources.
Key points
- The article says AI can help threat actors scale spear-phishing, reconnaissance, deepfakes, and other attacks, while lowering the technical barrier for less experienced attackers.
- Darktrace reports a 135% rise in novel social-engineering attacks across its email customers in 2023 and says 62% of 17.8 million phishing emails detected from December 2023 to July 2024 passed DMARC checks.
- Generative AI can summarize security data and support training simulations, but may produce false positives and create risks such as prompt injection and sensitive-data exposure.
- The article argues generative AI alone is insufficient and describes supervised and unsupervised machine learning, large language models, and natural language processing as complementary approaches.
- Darktrace presents its own AI security platform and offers guidance for evaluating AI-based security tools.
Article Details
- Topic
- The use of artificial intelligence in cybersecurity, including AI-enabled threats and defensive applications.
Vendors
Products
ChatGPTand simulate human-like interactions to deceive victims more effectively. Generative AI tools, such as ChatGPT, are already being used by adversaries to craft these sophisticated phishing emails, which can more aptlyDarktrace ActiveAI Security Platformunique risks, providing extended visibility across their entire digital estates with the Darktrace ActiveAI Security Platform™.Microsoft CopilotGenerative AI (e.g., ChatGPT or Microsoft Copilot) is a type of AI that creates new or original content. It has the capability to generate images, videos, or text based on information it learns from large datasets.