Mar. 25 at 10:52 AM
$CRWD $PANW $ZS
This guy is delusional ⬇️
-Claude can flag suspicious behavior within its own API interactions — prompt injection attempts, anomalous usage patterns, potential misuse. That’s self-monitoring of an AI model. That is not: a SIEM, an EDR, a network security platform, identity threat detection, cloud workload protection, etc, Calling it “intrusion detection” is like saying your smoke detector replaced your fire dept. It doesn’t call the fire truck, route it, contain the blaze, do forensics, file the insurance claim, or harden the building for the next event.
$CRWD,
$PANW isolate endpoints, kills processes, maps lateral movement, remediate, & feeds threat intl back into a global network trained on trillions of events.
- That’s a platform moat.
-
$CRWD ,
$PANW $ZS have combined ARR north of
$15B, deep federal contracts, with strict compliance mandates baked into regulated industries.
-A blog post about an AI model’s self-monitoring capability is not a disruption event.
