Tech

Chrome’s AI features may be hogging 4GB of your computer storage

Google Chrome's AI features are quietly devouring 4GB of user storage, courtesy of a massive on-device AI model file that automatically downloads to the browser's system folders when certain AI-powered tools like scam detection and writing assistance are enabled, leaving users with unexplained storage drops. The culprit is a 4GB weights.bin file tied to Google's Gemini Nano AI model. This stealthy storage hog is a growing concern for users with limited desktop storage.

{ "headline": "Chrome's AI Features Consume User Storage", "synthesis": Google Chrome's AI features are using a significant amount of user storage, due to a large on-device AI model file that automatically downloads to the browser's system folders. The file in question is a 4GB weights.bin file connected to Google's Gemini Nano AI model, which powers Chrome AI tools like scam detection, writing assistance, autofill, and suggestion features.

Overview

The Gemini Nano model is designed to run locally, using training parameters stored on the user's device rather than pulling information from cloud-based models. This provides some privacy benefits, but it can be a problem for users with limited desktop storage. The 4GB weights.bin file is installed inside the browser directory when certain AI features are enabled, and users may not be clearly notified about the file size requirements.

What it does

The Gemini Nano AI model powers several Chrome AI tools, including scam detection, writing assistance, autofill, and suggestion features. These tools are designed to improve the user experience, but they come at a cost in terms of storage space. If you have certain Gemini AI features enabled on Chrome, it's likely that the 4GB file has already been downloaded to your system.

Tradeoffs

To free up space, you can't simply delete the weights.bin file, as Chrome may re-download it again in the future if you still have AI features enabled. To remove the file and prevent it from coming back, you need to head to Settings > System and toggle off the On-Device AI option. This will disable the AI features and prevent the file from being re-downloaded. Google specifies that the exact size of the Gemini Nano model may vary as the browser updates the model, but this information is not clearly presented to users when enabling the AI features.

In practical terms, users with limited storage space may need to weigh the benefits of the AI features against the cost in terms of storage space. If you need to free up space, disabling the On-Device AI option may be the best solution. It's worth noting that Google could have avoided this confusion by making the storage requirements clearer to users or providing an option to power Chrome AI features with cloud-based models.

Similar Articles

More articles like this

Tech 1 min

Did Lamb Weston Holdings, Inc. Insiders Breach their Fiduciary Duties to Shareholders?

A confidential trading probe into Lamb Weston Holdings, Inc. has raised questions about whether company insiders exploited non-public information to profit from stock trades, potentially breaching their fiduciary duties to shareholders. The investigation centers on suspicious stock activity around a major acquisition announcement in 2022. A review of trading records and insider communications is underway to determine the extent of any wrongdoing.

Tech 1 min

Motorcycle Market Accelerates Global Mobility Shift through EV and Premiumization Trends | US$ 250.2 billion by 2033

The global motorcycle industry is hitting the gas on electrification and premiumization, with a projected $250.2 billion valuation by 2033—fueled by a 7.5% CAGR as OEMs pivot from ICE platforms to high-margin EVs like Harley-Davidson’s LiveWire and Ducati’s V21L. Emerging markets are adopting 48V mild-hybrid retrofits, while lithium-ion battery swaps and silicon carbide inverters slash urban fleet TCO, accelerating the shift from two-wheeled mobility to last-mile logistics and ride-hail networks.

Tech 1 min

ELLKAY Appoints Lucky Singh as Chief Technology Officer

Healthcare data orchestration pioneer ELLKAY taps seasoned engineering executive Lucky Singh as Chief Technology Officer, signaling a strategic pivot towards AI-driven interoperability and next-generation data exchange standards. Singh's appointment underscores the company's commitment to harnessing artificial intelligence and machine learning to streamline healthcare data workflows. The move is expected to accelerate ELLKAY's push into the burgeoning healthcare data exchange market.

Tech 1 min

GrowthLoop Unveils 2026 AI and Marketing Performance Index, Highlighting that Data Issues Significantly Slow Marketing Cycles, Experimentation, and Personalization

Marketing cycles are being crippled by data quality issues, hindering AI-driven personalization and experimentation at scale, with 40% of marketers experiencing slow cycles and 75% citing failed experiments due to inadequate data. The problem stems from data inconsistencies, incomplete records, and poor data governance, which are exacerbated by the increasing reliance on AI platforms. This data decay is costing marketers valuable time and resources.

Tech 1 min

Hybrid Software Group: Result of the Annual General Meeting

Hybrid Software Group's AGM outcome sets stage for accelerated M&A strategy, with 99.9% of shareholders backing a £10 million share buyback plan and a 10% increase in dividend payout, as the company seeks to capitalize on its position in the document automation and enterprise content management markets. The move follows a 25% surge in revenue for FY 2025, driven by growth in its PDF and document management software offerings.

Tech 1 min

BasedAI Emerges from Stealth to Launch Hirebase, the instant AI Workforce Platform for Businesses

A stealthy AI upstart emerges with Hirebase, a platform that instantaneously deploys and manages a customized workforce of open-source AI models, agents, and automation tools, leveraging a proprietary multi-agent orchestration stack acquired from Warden App. This vertical stack promises to bridge the gap between AI research and enterprise adoption. Initial funding has been secured for the venture.