Perplexity has launched "Teammates," work-focused AI agents that integrate with Perplexity Computer to handle multiple workspace roles.
1 editorial report · 1 verified social mention. The most authoritative report leads while later evidence completes the story.
Cybercriminals Are Now Impersonating AI Agents to Steal Credentials A new campaign highlights an emerging cybersecurity risk: attackers impersonating legitimate AI bots to bypass security controls and hunt for exposed credentials. According to research from CQ Prime Research (Cequence Security), a threat actor impersonated 32 different AI bots, including agents associated with OpenAI, Anthropic, Meta, and Perplexity. Key Findings: * Credential Hunting: Automated scans targeted exposed passwords, API keys, cloud credentials, configuration files, and private SSH keys. * Identity Spoofing: Attackers rotated between 32 different AI bot identities to appear legitimate. * Broad Targeting: Organizations across financial services, retail, telecommunications, energy, and software were targeted. * W
Open mention