What happened
On 16 October 2026 (IC3 PSA number PSA261016) the FBI and CISA issued a joint Public Service Announcement, 'Cyber Threat Actors Use a Variety of Techniques to Target Election and Campaign Officials', warning of phishing/smishing against election officials, candidates, and political organisations ahead of the 2026 U.S. midterms. It describes how 'super intelligence' (SI) is enabling larger, more efficient and more effective phishing campaigns — generating genuine-sounding language in native dialects — and increasingly convincing deepfake audio and video used to create new identities, impersonate real people, and bolster phishing schemes. It recommends phishing-resistant MFA, checking for content credentials/watermarking to detect SI-generated media, SPF/DKIM, least-privilege access, and external-email banners.
Why it matters
This is a Tier-1 US government (CISA/FBI) advisory that officially treats AI/'super intelligence' as an amplifying factor for social-engineering and identity attacks. It codifies practical, actionable defensive measures for AI-enabled phishing, synthetic-media impersonation, and OAuth consent phishing, and is directly relevant to critical-infrastructure and election-security work as well as any organization assessing AI-driven account-compromise risk. It also reflects the super-intelligence terminology now being used in US federal cyber guidance.
Action needed
Adopt the advisory's controls — phishing-resistant MFA on email/VPN/critical systems, content-credential checks for AI-generated media, SPF/DKIM, least-privilege access, and staff verification of sensitive requests via secondary channels; brief election/campaign and critical-infrastructure teams.