PROJECT BRIEF

BACKGROUND

The proliferation of AI-generated deepfakes and misinformation through high traffic media channels and phishing schemes are a significant threat to civic life across America. Particularly, deceptive content can erode resident-government relations, trust in democratic processes, and the safety of public and private settings across the United States. 

AIMS

The City of Fort Collins is aiming for a practical, feasible, and city-ready strategy to (1) better inform citizens about the dangers of AI deepfakes, (2) integrate effective mitigation techniques in the most vulnerable channels, and (3) and create effective policy that aligns with City goals and will remain sustainable as generative AI evolves. 

METHODOLOGY

The team focused the current state of AI deepfakes along four main fronts: (1) current regulations and policies surrounding AI technology in public input, (2) AI-driven risks in public input, (3) current technological solutions and detection tools for AI-generated content, and (4) AI literacy and training. The team also conducted stakeholder interviews with Fort Collins groups and external experts, including forensic analysts, cybersecurity fellows, and social media researchers to support research and outline future concerns and areas of AI deepfakes.

DELIVERABLES

The team developed a 3-pillar framework for the City of Fort Collins to implement and integrate:

(1) a public awareness campaign tailored to vulnerable Fort Collins channels, (2) technology solutions and implementation plan for existing public and private sector organizations, and (3) recommended policy measures for provisions surrounding deepfakes. 

IMPACT AND FUTURE WORK

This three-pillar approach covers an exhaustive framework that offers Fort Collins a practical roadmap for strengthening their resilience toward AI deepfake threats, ultimately protecting citizens, public officials, and organizations from misinformation and scams. The multi-lateral and multi-layered framework applies to any mid-sized city looking to combine education, technology, and governance to proactively address the dangers that deepfakes present. By emphasizing scalability, partnerships, and data-driven iteration on technical and non-technical fronts, this approach provides an infrastructure that remains agile and adaptable to the evolving nature of deepfake technologies.