Why Long-Horizon AI Conversations Break Traditional Safety Models
The lawsuit against OpenAI following the Florida State University [...]
ObjectSecurity Presents “Your Model Remembers More Than You Think”
Pretrained models power today's AI systems, but they also [...]
Mythos vs. Economics
On April 7, 2026, Anthropic’s Mythos announcement indicates [...]
Mythos and the Rise of Autonomous AI: Why Control Matters More Than Capability
Most of the attention on Anthropic's Mythos focuses [...]
ObjectSecurity Showcases BinLens and FortiLayer at the 8th Annual SSC Cyber Expo 2026
Space and defense mission systems increasingly rely on complex [...]
Mythos vs. Determinism
On April 7, 2026, Anthropic’s Mythos announcement points [...]
PRESS RELEASE: ObjectSecurity investigates stand alone CLI option for BinLens
San Diego, CA – November 19, 2025 – ObjectSecurity, [...]
FortiLayer for LLMs
FortiLayer extends its internal model analysis to LLMs, providing [...]
ObjectSecurity Presents “Reverse Engineering Embedded AI Models in Firmware and Binaries”
ObjectSecurity will present at HackTheBay on the challenge of [...]
ObjectSecurity Presenting “Increasing the Analysis Surface of Large Language Models”
LLM safety tooling still leans heavily on black box [...]
