AI Case Study
DoNotPay has succesfully appealed £2 million in parking tickets using a chatbot to contest tickets based on user inputs
The DoNotPay website and Messeger chatbot uses NLP to appeal parking tickets automatically, with an initial 40% claim success rate. The site's bot is now being implemented for other legal uses, including for refugees legal claims and making complaints against landlords.
From Business Insider: "Users fill in some details and choose their defence from a list of 12 options to generate an automated appeal to the council issuing the parking ticket." However, while DoNotPay's bot "performs very well in narrow Q&A fields, it can sometimes be prone to ‘communication breakdown’ when a user gives responses to the bot that have not been planned for, or are not recognised. This is often due to a limitation of the NLP program, which then tends to stop the flow of the process. It’s also the major challenge faced by all other bots, in the legal space or elsewhere... with continued machine learning in terms of the decision trees for each legal query, as well as improvements in NLP use, DoNotPay will improve the more people use it and the data set expands."
According to Business Insider, in its first four months "more than 86,000 people have used it to appeal against council parking fines. Nearly 40 per cent of them were successful, according to a poll of the site's users."
Artificial Lawyer reports that "the tech used involves IBM Watson for some of its natural language processing capabilities. DoNotPay has also been using Facebook Messenger to host the bot."
DoNotPay is website where a chatbot asks users a series of questions to provide legal advice, and initially to automate the appeal of parking tickets.