TMCnet News
Push Security Wins Pinnacle Award for Best AI-Powered Threat DetectionPush Security, the most powerful AI-native security tool in the browser, today announced it has been named a winner in the 2026 Pinnacle Award for Technology, being selected as a Diamond honoree in the Best AI-Powered Threat Detection category. The award recognizes Push's development of an end-to-end agentic threat hunting and detection engineering pipeline, a system that uses AI agents as a force multiplier for Push's human threat research team, tripling the number of new detections shipped to customers in the first half of this year. "Adversaries don't iterate according to a quarterly roadmap, they evolve continuously, shipping new attack techniques faster than any human team can manually respond to them," said Adam Bateman, CEO of Push Security. "We built our agentic pipeline to match that pace, and winning this award is a signal that the industry recognizes what's at stake when detection speed is measured in minutes."
Putting AI to work against AI-enabled attacks
In a recent demonstration of the system's capability, Push researchers and AI agents analyzed trillions of browser events to uncover an in-the-wild InstallFix attack, a sophisticated phishing campaign using a Cloudflare Pages-hosted kit with a WebAssembly C2 connector, designed to trick users into installing malware via a convincing fake AI platform page. The detection was engineered and deployed to all Push customers within minutes of discovery. The same foundational capability that detects attacks also governs how employees use AI tools. Push's browser-native platform provides real-time visibility and control over AI application usage, seeing what users paste into AI prompts, what data moves to unapproved tools, and which AI apps are operating without security oversight, all enforced at the session layer where the data actually flows.
About Push Security
About the Pinnacle Awards
View source version on businesswire.com: https://www.businesswire.com/news/home/20260702125841/en/ |

