https://theneuraltide.com/ Evidence-first research on AI reality, risk, society, preparedness, opportunity and human contribution. en-us Mon, 28 Sep 2026 19:43:00 GMT https://theneuraltide.com/posts/nvidia-agent-safety-platform.html https://theneuraltide.com/posts/nvidia-agent-safety-platform.html Mon, 28 Sep 2026 19:43:00 GMT NVIDIA is moving AI agent safety beyond model behavior into runtime and hardware controls. What is known, what remains uncertain, and what to watch next. https://theneuraltide.com/posts/ai-agent-time-horizon.html https://theneuraltide.com/posts/ai-agent-time-horizon.html Mon, 28 Sep 2026 17:00:00 GMT What AI agent time horizons actually measure, what they do not prove, three plausible work scenarios, and how to prepare. https://theneuraltide.com/posts/ai-agents-vulnerability-discovery.html https://theneuraltide.com/posts/ai-agents-vulnerability-discovery.html Mon, 21 Sep 2026 17:00:00 GMT Real capability, benchmark limits, and the attacker-defender race. https://theneuraltide.com/posts/day-02.html https://theneuraltide.com/posts/day-02.html Thu, 17 Sep 2026 17:00:00 GMT Evidence on exposure, entry-level work, productivity and adaptation. https://theneuraltide.com/posts/day-01.html https://theneuraltide.com/posts/day-01.html Thu, 17 Sep 2026 17:00:00 GMT What expert assessments and current evidence can—and cannot—tell us.