Google Unveils New AI Updates to Help Gmail Users Fight Spam

Google has substantially invested in artificial intelligence (AI) throughout the year. From launching its AI chatbot, Bard, in February to offering new AI-powered features in Google search, the tech giant has been deeply immersed in emerging technology. The new Google AI update for Gmail has been aiming to combat spam.

This update is undoubtedly essential, considering spam remains a significant issue for several Gmail users. Many users have been in situations where their storage space was exhausted due to an influx of spam emails.

According to a report by IANS, Google has recently introduced a robust upgrade to its spam detection system. This move is being praised as a significant enhancement in defense capabilities. The innovation, named RETVec (Resilient and Efficient Text Vectorizer), signifies a notable advancement in text classification technology, specifically crafted to counteract “adversarial text manipulations.”

In simpler terms, this implies that Google has fortified its capacity to recognize and block tricky spam tactics, including emails with special characters, emojis, and typos that might otherwise evade Gmail’s defenses. The company underscores this upgrade is one of the “most substantial” in recent years.

The New Google AI Update Features

The cornerstone of Google’s AI update lies in RETVec, an advanced text classification system. Google elucidates that RETVec aims to enhance the robustness and efficiency of text classifiers, achieving top-notch classification performance while substantially decreasing the computational resources required.

Prominent Google services such as Gmail, YouTube, and Google Play heavily depend on text classification models to identify harmful content, from phishing attacks to inappropriate comments and scams. These models encounter difficulties classifying specific texts because malicious actors actively employ techniques like homoglyphs, invisible characters, and keyword stuffing to avoid detection.

A notable characteristic of RETVec is its innovative architecture, enabling it to function seamlessly across all languages and characters without extensive text preprocessing. This adaptability makes RETVec suitable for various applications, ranging from on-device deployments to large-scale text classification on the web.

Google underscores that models trained with RETVec deliver improved accuracy and exhibit faster inference speed owing to their compact representation. Smaller models contribute to reduced computational costs and decreased latency, essential factors for large-scale applications and on-device models.

Furthermore, Google has designed RETVec as an open-source tool, enabling developers to harness its capabilities to construct resilient and efficient text classifiers for server-side and on-device applications. Google notes explicitly that the Gmail spam filter already incorporates RETVec to fortify its defenses against malicious emails.

WRITTEN BY

Team Eela

TechEela, the Bedrock of MarTech and Innovation, is a Digital Media Publication Website. We see a lot around us that needs to be told, shared, and experienced, and that is exactly what we offer to you as shots. As we like to say, “Here’s to everything you ever thought you knew. To everything, you never thought you knew”