QuillBot launches new grammar checker tool

QuillBot, a revolutionary AI-based writing platform that uses language processing software to summarize and paraphrase text, has announced the beta release of its long-awaited grammar checker tool, a new feature designed to improve textual clarity and fluency.

In addition to identifying grammatical errors, the grammar checker is capable of flagging spelling and punctuation mistakes and generating suggested edits, allowing users to easily make corrections to the text.

QuillBot was founded in 2017 by Rohan Gupta, Anil Jason and David Silin, a trio of computer science students with a background in natural language processing. In 2020, the company received more than $4 million in seed funding from a fundraising initiative led by venture capital firms GSV Ventures and Sierra Ventures, with contributions from Service Provider Capital.

Since Quillbot's launch four years ago, its user base has expanded significantly, with more than 5 million people worldwide using the company's tools each month, according to a company news release.

QuillBot's signature paraphraser tool, which rewrites text with new sentence structures and vocabulary, comes in several modes designed to enhance aspects of writing such as fluency, creativity, formality, conciseness and level of detail. Users can further customize their writing with a word flipper that controls the number of words replaced with synonyms in the outputted text, and additional features allow for keeping quotations intact, including or avoiding contractions and adhering to American English, British English or Australian English standards.

The company later added a summarizer tool that compiles key information into a summarized paragraph or list of bulletin points, with further options to adjust the length of the generated text. 

According to Quillbot Business Development Specialist Emily Perry, Quillbot's tools are powered by natural language processing models that have been trained to perform tasks by learning from examples. The AI first undergoes pre-training on a generic language processing task — for instance, predicting the next sentence in a large piece of text such as a news article — before learning how to perform a more specific operation such as summarizing or paraphrasing. 

Training the paraphraser tool involves feeding it examples of paraphrased text and corrections of anonymous user inputs, which teaches the AI how to address common writing issues such as unclear statements and sentence fragments. Similarly, the summarizer has been trained on how to identify and extract an article's main points, while the grammar checker has been taught grammar, spelling and punctuation concepts.

"Because the AI keeps learning over time as we train it further, our tools are continuously improving, helping our users learn to become better, more efficient writers," Perry said.