Woman Arrested for Bond Violation After Ex-Boyfriend Submitted AI-Fabricated Text
A Florida woman was arrested and jailed after her ex-boyfriend submitted an AI-generated fake text message as evidence that she violated a no-contact order. Prosecutors later dropped the bond violation charge and she was acquitted of the underlying battery charge, but the case illustrates systemic vulnerabilities in how courts verify digital evidence.
What happened
Melissa Sims, a Florida resident, was arrested in November 2024 for battery after an altercation with her boyfriend. The judge issued a no-contact order prohibiting communication between them.
Months later, her ex-boyfriend created an AI-generated text message purporting to show Sims violating the no-contact order by sending him disparaging messages. Sims was arrested for violating her bond based on this fabricated text screenshot.
No one verified the evidence before bringing charges. Sims spent two days in jail during this period.
Reported impact
Sims described her jail experience as traumatic, comparing it to depictions in popular media. She reported suicidal ideation during her ordeal.
After eight months of legal proceedings, prosecutors dropped the bond violation charge. Sims was acquitted on the original battery charge at trial.
Index analysis
The case highlights vulnerabilities in the criminal justice system's handling of AI-generated evidence. The court accepted a screenshot as evidence without verifying metadata or the text's original form.
Experts note that AI-generated images, video and audio have become increasingly difficult to detect, creating challenges for judges and law enforcement in determining authenticity.
Governance controls
Judge Herbert Dixon of the Superior Court of the District of Columbia stated that such incidents are increasing in frequency. He noted that prosecutors and police need greater due diligence before bringing charges involving potential AI-fabricated evidence.
Experts recommend reversing the evidentiary standard from 'trust unless proven otherwise' to 'distrust until verified' for digital content presented as evidence.
Classification
Sources and evidence
Sources not yet published
This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.