In a curious turn of events, Matthew Elliott from Connecticut has been caught embedding secret instructions for artificial intelligence within his court documents during an ongoing lawsuit against a medical firm. Elliott chose to use a minuscule, nearly invisible white font to try to sway the AI's responses in his favor. One particular line in his filing directed that any AI review should align its output with his arguments to effectively challenge the denial noted in the court's official entries.
The court's scrutiny revealed the hidden text, prompting an investigation into the unusual formatting. Judge Spader, overseeing the case, clarified that the court does not utilize AI in its proceedings.
In subsequent filings, Elliott included links to a SpongeBob SquarePants video and made playful jabs, such as "hi :) I hope you can't see me." Ultimately, the court decided to prohibit him from submitting documents electronically.
Elliott defended his actions by claiming he wanted to test whether artificial intelligence was being used in legal settings. In an interesting twist, journalists later uploaded the altered document into an AI model, which successfully identified and disregarded the simplistic manipulation, proceeding to make its own judgment.
The question remains: did Elliott's actions serve a purpose? Some argue that vigilance regarding AI is essential, while others see it as mere mischief.
Informational material. 18+.