Meta has officially rolled out its AI-powered support assistant across Facebook and Instagram, aiming to deliver faster, more reliable help while strengthening content enforcement through advanced artificial intelligence systems.
The newly launched Meta AI support assistant is designed to provide 24/7 assistance for a wide range of user issues, from account recovery to privacy settings. Available globally on iOS, Android, and desktop Help Centers, the tool marks a major shift in how users interact with support services on Meta’s platforms.
Meta AI support assistant brings instant help to users
The Meta AI support assistant focuses on delivering quick, action-driven solutions rather than just suggestions. Users can now resolve common account-related problems directly within the apps, without navigating complex help pages.
The assistant can handle several key tasks, including reporting scams or impersonation accounts, explaining why content was removed, assisting with appeals, managing privacy settings, resetting passwords, and updating profile details.
Support requests are typically processed in under five seconds, significantly reducing wait times compared to traditional help systems. The tool is integrated directly into Facebook and Instagram, making assistance easily accessible at any time.
Meta has also started testing login support features powered by AI for users facing account access issues, initially in the US and Canada, with plans to expand globally.
Advanced AI strengthens content enforcement
Alongside the support assistant, Meta is introducing more advanced AI systems to enhance content moderation and platform safety. These systems are designed to detect serious violations such as scams, fraud, impersonation, and harmful content more accurately and efficiently.
Early results from testing show significant improvements. The AI systems have been able to identify thousands of scam attempts daily that were previously undetected. They have also reduced reports of impersonation involving high-profile individuals by over 80%.
Additionally, the technology has doubled the detection rate of violating adult content while cutting enforcement errors by more than 60%. AI has also proven effective in identifying suspicious account activity, such as unusual login patterns or profile changes, helping prevent account takeovers.
In another example, the system detected fake websites mimicking legitimate brands by analysing inconsistencies such as unusually low prices and suspicious URLs, leading to a reduction in scam-related ads.
Broader language coverage and cultural understanding
One of the key advancements in Meta’s AI systems is their ability to operate across a much wider range of languages. The new systems support languages spoken by 98% of internet users, a significant increase from the earlier 80% coverage.
These systems are also designed to quickly adapt to cultural nuances, regional slang, evolving code words, and even emoji meanings, enabling more accurate moderation across diverse communities.
A balanced approach between AI and human oversight
Meta emphasises that while AI will play a larger role in scaling operations, human oversight will remain essential. Experts will continue to train, evaluate, and oversee AI systems, particularly in high-risk or complex cases such as account appeals and legal matters.
The company plans to gradually deploy these advanced AI systems over the coming years, reducing reliance on third-party moderation while strengthening internal capabilities.
Importantly, Meta has clarified that its Community Standards will remain unchanged. Instead, the focus is on improving how those standards are enforced, with fewer mistakes and faster response times.
Comments are closed.