Brave has introduced Cookiecrumbler, an open-source tool that uses advanced large language models and human oversight to accurately detect and block intrusive cookie consent banners without disrupting website functionality, marking a new milestone in online privacy protection.
Brave has officially launched its new open-source tool, Cookiecrumbler, designed to address the pervasive issue of intrusive cookie consent banners that frequently clutter web pages. This initiative aims to enhance user privacy while mitigating the risks associated with hidden tracking, a concern that many privacy advocates share.
Cookiecrumbler employs a sophisticated approach to detecting and blocking these cookie notices. Unlike traditional methods that rely on generic rules and often lead to unintended consequences, the tool utilises advanced large language models. These models have the capability to identify cookie banners across various web platforms, even accommodating different languages and regional formats. This innovation aims to significantly reduce the common problem of inadvertently blocking essential website functionalities, such as checkout buttons and navigation menus.
In conjunction with its automated capabilities, Brave emphasises the importance of human oversight in the cookie detection process. Automated crawlers are deployed to scan websites, simulating access from various locations through the use of proxies. Following this, Cookiecrumbler analyses the findings, but human reviewers are involved to ensure detections are accurate. This two-tiered model is designed to balance automation with the need for web stability, preventing errors that could disrupt user experience.
Currently, Cookiecrumbler operates within Brave’s backend infrastructure, with plans for potential integration into the browser itself, contingent on successful privacy reviews. In addition, Brave is fostering community involvement by publishing its website crawl results on GitHub. This move invites developers and users alike to contribute to the ongoing refinement of cookie blocking, promoting a collaborative effort that not only enhances Cookiecrumbler’s functionality but also strengthens the broader ad-blocking community.
The launch of Cookiecrumbler by Brave is seen as a significant step forward in the ongoing effort to improve online privacy while maintaining a seamless browsing experience, positioning it as a noteworthy development in the landscape of digital privacy tools.
Source: Noah Wire Services
- https://github.com/brave/cookiemonster – Brave’s open-source tool, CookieMonster, is designed to automatically detect and block cookie consent notices, enhancing user privacy by mitigating hidden tracking risks.
- https://arxiv.org/abs/2204.04221 – The ‘CookieEnforcer’ system utilizes machine learning to automatically discover cookie notices and enforce privacy-respecting settings, aiming to disable non-essential cookies without disrupting user experience.
- https://brave.com/did-you-know/brave-blocks-cookie-consents/ – Brave’s browser blocks cookie consent notices by default, providing users with a cleaner and more private browsing experience by eliminating intrusive cookie banners.
- https://arxiv.org/abs/2204.11836 – This study discusses the challenges and methodologies in automating the detection of dark patterns in cookie banners, highlighting the complexities in ensuring user privacy through automated systems.
- https://arxiv.org/abs/2309.06196 – The ‘Cookiescanner’ tool automates the detection and evaluation of GDPR consent notices on websites, assessing their compliance and effectiveness in protecting user privacy.
- https://zisc.ethz.ch/2022/03/31/solving-deceptive-cookie-banners-with-machine-learning/ – The CookieBlock extension uses machine learning to classify and remove privacy-threatening cookies, aiming to protect users from deceptive cookie banners and enhance online privacy.
- https://news.google.com/rss/articles/CBMixgFBVV95cUxOelJYWkc4X0JxVWJjcU15TnpyNEF1dXgzbVpuUDhlZDI3TGFnUEYyUlJSblM3RWJpdUFnVDV0T2d0dG1XOWxOUV96ak5hYkVGUkFEM213Qklha2VWaXItcXRXZUFGVkZDT1BiUUZYUkZMUW1Yd2tHeHNQTzRMdURyV1hGZHFGUWZwU25KUExna2JUMU9WdU5SUzZtSXFQLXE4bXlDNS0yZkZudHdiRFlCdmY5QkVseFJGM2MzSVFtMnJIMjNqWXc?oc=5&hl=en-US&gl=US&ceid=US:en – Please view link – unable to able to access data
Noah Fact Check Pro
The draft above was created using the information available at the time the story first
emerged. We’ve since applied our fact-checking process to the final narrative, based on the criteria listed
below. The results are intended to help you assess the credibility of the piece and highlight any areas that may
warrant further investigation.
Freshness check
Score:
8
Notes:
No outdated references detected. Mentions future integration plans, suggesting recent development. Lack of a direct press release link or third-party verification limits full confirmation.
Quotes check
Score:
9
Notes:
No direct quotes requiring attribution. Technical descriptions align with Brave’s public privacy advocacy and prior tool developments (e.g., ad-blockers).
Source reliability
Score:
7
Notes:
Narrative originates from Google News-hosted article; publisher unknown. Brave’s history of privacy tools supports plausibility, but unverified third-party corroboration lowers certainty.
Plausability check
Score:
8
Notes:
Claims align with Brave’s focus on privacy innovations (e.g., Tor integration, ad-blocking). Community GitHub collaboration and LLM use for banner detection are technically plausible but require further verification.
Overall assessment
Verdict (FAIL, OPEN, PASS): PASS
Confidence (LOW, MEDIUM, HIGH): MEDIUM
Summary:
The narrative is consistent with Brave’s operational focus and includes plausible technical details. Lack of direct press release or third-party corroboration necessitates medium confidence.
