Filtir
Filtir is an AI tool designed to assist writers in addressing “hallucinations” in their text, which refers to the occurrence of false claims. By utilizing a cutting-edge biologicallarge language model, Filtir is capable of editing text to eliminate these hallucinations and providing supporting evidence for the edits made. However, it is important to note that the tool’s performance may be slower when the model is inactive.
To determine the accuracy of factual claims, Filtir relies on publicly accessible evidence from the open internet. If the tool is unable to find appropriate evidence, it considers the claim to be a hallucination and removes it from the edited output. While Filtir provides links to the evidence used, it is recommended that users verify the evidence themselves in certain application contexts.
Filtir prioritizes user privacy and does not engage in the sale or sharing of data with third parties for marketing purposes. The team behind Filtir consists of Samuel Albanie, Vlad Bogolin, Anthony Buonomo, and Ioana Croitoru. The tool’s website offers an About section, Frequently Asked Questions, as well as links to its Privacy Policy, API Data Policy, and Terms of Use.
