The vulnerability in the AiBot2Controller allows an authenticated low‑privilege user to supply a chatid belonging to another user without verifying ownership. This bypass permits the attacker to read, delete, exfiltrate, or interrupt the target’s AI assistant conversation, effectively compromising confidentiality and disrupting service. The flaw is a classic authorization bypass (CWE‑639).

Rebuild versions through 4.5.0 built by get flaw resides in the AiBot2Controller’s /aibot2 endpoints.

The CVSS base score of 6 signifies medium severity. The EPSS score of < 1% indicates a very low probability of exploitation. It is inferred that the vulnerability can be exploited via the /aibot2 endpoints exposed over the network. An attacker must be an authenticated low‑privilege user and can trigger the exploitation if they possess legitimate credentials. The vulnerability is not currently listed in CISA KEV, indicating no confirmed active exploitation.