A researcher has shown how malicious actors can create custom GPTs that can phish for credentials and exfiltrate them to external servers.
The post Malicious GPT Can Phish Credentials, Exfiltrate Them to External Server: Researcher appeared first on SecurityWeek.