COVID-19 Misinformation: Twitter’s AI Misfires

Maybe you are interested

Automated labeling technology employed by Twitter to combat coronavirus misinformation has stumbled upon a series of errors. This raises concerns about the platform’s reliance on artificial intelligence for content moderation. The labeling system aims to identify tweets spreading harmful conspiracy theories, such as the false claim that 5G technology causes COVID-19. While some tweets containing the conspiracy theory are appropriately labeled, the technology has mistakenly flagged numerous tweets that actually refute the theory or provide accurate information. This includes tweets linking to respected news outlets like Reuters, BBC, and Wired. Experts warn that these mislabeled tweets may mislead users, especially if they don’t click on the label. Twitter’s failure to notify users about labeled tweets or provide an appeal process further complicates the issue.Array## Fooling Twitter’s Automated System

You are watching: More harm than good? Twitter struggles to label misleading COVID-19 tweets

Twitter has implemented an automated system to label tweets containing incorrect information about the coronavirus 5G conspiracy theory. However, some users are finding ways to circumvent the system.

Testing the System

YouTubers and other users are testing the system by tweeting words such as “5G” and “coronavirus,” flooding the site with incorrectly labeled tweets. One user, Ian Alexander, discovered the label on a tweet unrelated to the conspiracy theory. He then tested the system by tweeting the trigger words and the label appeared automatically.

Concerns about the Labels

Alexander believes that the labeling system may be counterproductive. Users may see the notice on their timeline without clicking through to the full content, potentially reinforcing misleading information.

Other Examples of Misleading Tweets

Despite the labeling system, misleading coronavirus information continues to slip through the cracks. Prominent individuals, such as actress Fran Drescher, have tweeted unsubstantiated claims about 5G and its role in the virus. Another tweet featured remarks from Judy Mikovits, who promotes conspiracy theories about 5G and COVID-19. These tweets did not receive labels.

Success of Warning Labels on Other Platforms

Other social media platforms have reportedly had success with warning labels on false content. Facebook, for example, displayed labels on approximately 40 million COVID-19 posts, reducing the frequency of people viewing the inaccurate content by 95%.

Steps to Take

See more : MagSafe Battery Pack: Hands-on with Apple's new iPhone 12 accessory

To avoid spreading misinformation, it’s important to:

  • Be skeptical of claims made on social media.
  • Check the source of information before sharing it.
  • Report misleading tweets to Twitter.
  • Consult credible sources for accurate information about COVID-19.## FAQs

  1. What is Twitter doing to combat coronavirus misinformation?

    • Twitter is using automated technology to label tweets that spread false information about the coronavirus.

  2. What type of tweets are being labeled?

    • Tweets that spread the conspiracy theory about 5G causing the coronavirus.

  3. What do the labels say?

    • “Get the facts about COVID-19” and link to a page with curated tweets debunking the 5G coronavirus conspiracy theory.

  4. Why have some accurate tweets been mislabeled?

    • Twitter’s technology is new and prone to errors, especially when tweets contain words like “5G,” “coronavirus,” or “#5Gcoronavirus.”

  5. How many tweets have been mislabeled?

    • Twitter has not disclosed this information.

  6. Is Twitter taking steps to correct the errors?

    • Yes, Twitter says its automated systems are improving over time.

  7. See more : T-Mobile announces first Netbook for 3G network

    How is Twitter’s approach to coronavirus misinformation different from Facebook’s?

    • Twitter is using automated labeling, while Facebook relies more on human reviewers.

  8. Why is Twitter using an automated system and not human reviewers?

    • Automated tools help workers review reports more efficiently by surfacing content that’s most likely to cause harm.

  9. Can users appeal a labeling decision?

    • No, Twitter does not currently offer a way for users to contest the labeling of their tweets.

  10. Is Twitter’s labeling system effective?

    • While it may reduce the spread of misinformation, it can also lead to confusion and distrust.

  11. What are other ways Twitter can combat coronavirus misinformation?

    • Take action against users who spread misinformation with a large following, partner with third-party fact-checkers, and promote accurate information.

  12. How can users spot fake news?

    • Check the source, look for multiple perspectives, consider the author’s motivations, and be aware of cognitive biases.

Summary

Automated technology used by Twitter to label tweets containing coronavirus misinformation has raised concerns due to its high error rate. Experts warn that mislabeled tweets can confuse users, especially if they don’t click on the label. While Twitter’s labeling approach is similar to Facebook’s efforts to combat misinformation, it relies solely on automation, leading to the mislabeling of accurate tweets. Twitter plans to improve its system over time, but challenges remain in accurately distinguishing between false and true information.

To address these concerns, Twitter must take steps to reduce errors and provide users with an opportunity to contest labeling decisions. It should also consider partnering with third-party fact-checkers and promoting accurate information. Individuals can help combat misinformation by evaluating news sources, checking multiple perspectives, and being mindful of cognitive biases.

Source: https://anytimesoftcare.com
Category: Laptop