There Is An App That Scans All Your Tweets To See How Problematic They Are And I Tried It Out

    The app called Vanilla allows you to scan old tweets that you may have forgotten about.

    Over the past few years, people have had problematic old tweets resurface, and they’ve faced some permanent consequences.

    I have made the choice to step down from hosting this year's Oscar's....this is because I do not want to be a distraction on a night that should be celebrated by so many amazing talented artists. I sincerely apologize to the LGBTQ community for my insensitive words from my past.

    I’ve seen a few of my old tweets from 7/8 years ago floating around (which I have now deleted) using words like “chav” “skank” and other words I wouldn’t use now as part of my language and lot of them were taken out of context referring to TV shows but I would never say those

    Samuel Wood, a 24-year-old software engineer from London, has made an app called Vanilla that allows you to scan your tweets for anything problematic.


    Wood told BuzzFeed News: “The idea for Vanilla came about when I saw a young musician who had overcome the odds by breaking through racial and gender glass ceilings being characterised by a silly tweet he posted years ago while he was still a teen.

    “It seemed like an attack on an individual whose work to date and personality represented something much more important and positive than that singular tweet. I knew there must be a way to help people in such experiences.”

    Wood says the timing for his app has never been better.

    Imagine your kids looking at your old tweets and saying your cancelled as their father

    He said: “We’re living in a time where consumers want to own their data. People have been reminded that they will be held accountable in the court of public opinion. There have been constant reminders that what you post online matters and can be costly.”

    He said that over 1,000,000 tweets have already been scanned through the app, so I decided to scan my Twitter archive.

    Screenshot / Vanilla

    About a day later, all 38K of my tweets had been scanned by Vanilla.

    Screenshot / Vanilla

    The software flagged some of my tweets with a high “toxicity score”, but they were basically me screaming at TV shows.

    Screenshot / Vanilla

    Wood said that the toxicity scoring model was put together by asking internet users to rate internet comments on a scale from “very toxic” to “very healthy”.

    He said: “Toxic is defined as a rude, disrespectful, or unreasonable comment that is likely to make you leave a discussion.

    “The data has been tested and validated by external researchers. We leverage this model to rank social posts by these scores.

    “The more training data the model consumes the closer it gets to the average perceived toxicity for any given post.

    “Over time, at Vanilla, we will be regularly checking our training models and ensuring that we limit the impact of bias in our data sets in detecting and identifying what is ‘toxic’.”

    In a lot of the tweets the app flagged, if the words were to be used in a different context...they could be problematic.

    Screenshot / Vanilla

    The software also scans retweets, and flagged this funny tweet by 50 Cent.

    Screenshot / Vanilla

    And this one by Ed Balls tweeting his own name.

    Screenshot / Vanilla

    And me saying the word “poop”.

    Screenshot / Vanilla

    Ikran Dahir is a reporter for BuzzFeed News and is based in London.

    Contact Ikran Dahir at

    Got a confidential tip? Submit it here