JAKARTA - According to WhatsApp's monthly report, Meta's messaging platform banned more than 1.67 million accounts in India in April. Of these, 122,000 accounts were banned due to mandatory actions based on user complaints, while most, about 1.6 million accounts, were prohibited from preventing malicious activity in the app.
"We are very focused on prevention because we believe that it is better to prevent harmful activity from happening than detecting it after losses occur," the monthly report said, quoted by Tech2.
According to the WhatsApp framework, the app will ban accounts if it believes that users operating them are rude on the platform. "Our goal is to identify and stop abusive accounts as quickly as possible, that's why identifying accounts manually is unrealistic. Instead, we have a sophisticated machine learning system that takes action to ban accounts, 24 hours a day, 7 days a week," the report said.
The company says accounts will be violated in some cases, including when the account gets negative feedback, such as when other users report or block accounts.
The report also states that the WhatsApp system evaluates user accounts and behavior, and takes appropriate action after a number of negative feedbacks were reported.
These instant messaging companies use machine learning and other analytical tools to detect "very motivated users" and ban them from platforms.
Previously, WhatsApp banned nearly 1.8 million accounts in March and 1.4 million accounts in February. In January 2022, 1.8 million accounts in India were banned. WhatsApp has also banned a number of accounts in 2021. A total of 2. Million Indian accounts were banned by WhatsApp in December 2021 and 1.7 million in November that year.
The new IT rules, which took effect last year, require digital platforms with more than 5 million users to issue compliance reports every month, which include details of the complaints they receive and actions that have been taken to solve the problem.
Previously, WhatsApp emphasized that as end-to-end encrypted platforms, they have no visibility of message content. To overcome this, they rely on a number of behavioral signals from abusive accounts, and how often people report them.
In addition to behavioral signals from accounts, WhatsApp AI also relies on unencrypted information available, including user reports, profile photos, group photos and descriptions as well as several other advanced AI tools and resources to detect abuse on its platform.
The English, Chinese, Japanese, Arabic, and French versions are automatically generated by the AI. So there may still be inaccuracies in translating, please always see Indonesian as our main language. (system supported by DigitalSiber.id)
