Synthetic media capabilities, enacted rules, and operational practices offices document.
Deepfakes and elections
A deepfake is synthetic or heavily altered media depicting a real person in words or actions that did not occur. Voice cloning can use short samples. NCSL counts 31 states with political-deepfake laws as of Oct 13, 2026, and notes ongoing constitutional litigation in several jurisdictions.[28]
Texas Election Code §255.004 makes it a Class A misdemeanor to create a deceptive deep fake video and publish it within 30 days of an election with intent to injure a candidate or influence the result.[14]
Practices offices list in readiness plans
- A named channel that confirms or denies a circulating clip.
- Retention of original recordings of public events for comparison.
- Call-back verification before acting on urgent voice requests for funds or data.
- Referral of suspected election-law violations to the appropriate authority.
Provenance technology
C2PA Content Credentials attach signed provenance metadata. Watermarking may be combined for durability. Content Authenticity Initiative materials state that credentials report origin claims; they are not a universal deepfake detector, and metadata can be stripped.[47]
Child safety
Texas SB 20 (effective Sep 1, 2025) addresses obscene AI-generated images that appear to depict children.[15] Federal TAKE IT DOWN requires covered platforms to remove validly reported NCII within 48 hours as of May 19, 2026.[16,17] Colorado’s Chatbot Safety Act adds duties for conversational AI operators beginning Jan 1, 2027.[31]
Voice interfaces
Lifelike voice agents raise the same disclosure, records, and fraud-awareness issues as text chat. Texas TRAIGA disclosure duties apply when a government AI system interacts with residents.[1]