AI helped drive a 300% increase in bot traffic over the past year, particularly affecting the commerce sector, according to Akamai’s State of the Internet report published on 22 September 2026. The report draws on threat intelligence from Akamai’s global security and network infrastructure. It also recorded a 113% rise in daily API attacks between 2024 and 2025. Meanwhile, 87% of surveyed organisations said they experienced an API-related security incident in 2025, compared with 76% in 2022. Akamai said AI is helping make APIs the dominant attack surface for modern enterprises.
The report highlighted additional risks from AI browser extensions, chatbots and autonomous agents. Forty per cent of enterprise users had installed AI browser extensions, while a quarter had changed their permissions within the previous 12 months. Akamai said 6% of chatbot conversations contained sensitive corporate information, and that 47% of AI conversations on enterprise devices used personal identities or accounts, limiting IT visibility.
It also warned that Model Context Protocol (MCP) deployments could allow malicious servers to influence large language models through prompt injection or cross-server attacks. Akamai recommended adaptive edge and API protections, browser visibility and behavioural controls, and limiting agent autonomy according to how easily actions can be checked or reversed, with human approval for high-risk tasks.