vff
News

Recall's Security Gaps Persist Despite Microsoft Overhaul

Read original
Share
Recall's Security Gaps Persist Despite Microsoft Overhaul

A new tool called TotalRecall Reloaded has reportedly found a way to access Windows 11's Recall database through an alternative method, circumventing Microsoft's security measures. Recall, Microsoft's screenshot-based activity tracking feature for Copilot+ PCs, was originally deployed with minimal encryption and no authentication requirements, exposing months of sensitive user data to anyone with device access. After security researchers exposed these flaws in 2024, Microsoft delayed the rollout by nearly a year and implemented substantial security upgrades including encryption, Windows Hello authentication, and improved detection of sensitive information. The emergence of TotalRecall Reloaded suggests that even these revised protections may have exploitable gaps.

A new tool called TotalRecall Reloaded has reportedly found a way to access Windows 11's Recall database through an alternative method, circumventing Microsoft's security measures. Recall, Microsoft's screenshot-based activity tracking feature for Copilot+ PCs, was originally deployed with minimal encryption and no authentication requirements, exposing months of sensitive user data to anyone with device access. After security researchers exposed these flaws in 2024, Microsoft delayed the rollout by nearly a year and implemented substantial security upgrades including encryption, Windows Hello authentication, and improved detection of sensitive information. The emergence of TotalRecall Reloaded suggests that even these revised protections may have exploitable gaps.

  • TotalRecall Reloaded tool reportedly bypasses security measures in Windows 11's Recall feature by finding an alternative access point to the activity database
  • Recall originally stored unencrypted screenshots and activity logs on disk, making it trivial for attackers to extract weeks or months of sensitive user data
  • Microsoft's 2024 overhaul added encryption, Windows Hello authentication, improved sensitive data filtering, and made Recall opt-in rather than default
  • The new tool's emergence suggests that Recall's revised security architecture may still contain exploitable vulnerabilities

Recall represents a fundamental tension in on-device AI: the promise of local processing and privacy versus the reality of storing comprehensive activity logs that become high-value targets for attackers. The repeated discovery of security gaps in Recall, even after a major redesign, highlights how difficult it is to implement privacy-preserving features at scale and raises questions about whether local AI features can be secured adequately without fundamentally limiting their functionality.

  • Microsoft's security-by-redesign approach to Recall may require additional iterations, delaying broader rollout and adoption of the feature
  • The existence of side-channel access methods suggests that Recall's architecture may have fundamental design flaws that cannot be fully patched without rearchitecting the feature
  • Enterprise customers and security-conscious users may need to disable Recall entirely or implement additional endpoint controls, limiting the feature's practical reach
  • The incident reinforces the broader principle that comprehensive activity logging systems are inherently difficult to secure and may require stronger threat modeling before deployment
Share

Our Briefing

Weekly signal. No noise. Built for founders, operators, and AI-curious professionals.

No spam. Unsubscribe any time.

Related stories

AI Discovers Security Flaws Faster Than Humans Can Patch Them

AI Discovers Security Flaws Faster Than Humans Can Patch Them

Recent high-profile breaches at startups like Mercor and Vercel, combined with Anthropic's disclosure that its Mythos AI model identified thousands of previously unknown cybersecurity vulnerabilities, underscore growing demand for AI-powered security solutions. The article argues that cybersecurity vendors CrowdStrike and Palo Alto Networks, which are integrating AI into their threat detection and response capabilities, represent undervalued investment opportunities as enterprises face mounting pressure to defend against both conventional and AI-discovered attack vectors.

21 days ago· The Information
AWS Launches G7e GPU Instances for Cheaper Large Model Inference
TrendingModel Release

AWS Launches G7e GPU Instances for Cheaper Large Model Inference

AWS has launched G7e instances on Amazon SageMaker AI, powered by NVIDIA RTX PRO 6000 Blackwell GPUs with 96 GB of GDDR7 memory per GPU. The instances deliver up to 2.3x inference performance compared to previous-generation G6e instances and support configurations from 1 to 8 GPUs, enabling deployment of large language models up to 300B parameters on the largest 8-GPU node. This represents a significant upgrade in memory bandwidth, networking throughput, and model capacity for generative AI inference workloads.

29 days ago· AWS Machine Learning Blog
Anthropic Launches Claude Design for Non-Designers
Model Release

Anthropic Launches Claude Design for Non-Designers

Anthropic has launched Claude Design, a new product aimed at helping non-designers like founders and product managers create visuals quickly to communicate their ideas. The tool addresses a gap for early-stage teams and individuals who need to share concepts visually but lack design expertise or resources. Claude Design integrates with Anthropic's Claude AI platform, leveraging its capabilities to streamline the visual creation process. The launch reflects growing demand for AI-powered design tools that lower barriers to entry for non-technical users.

about 1 month ago· TechCrunch AI
Google Splits TPUs Into Training and Inference Chips

Google Splits TPUs Into Training and Inference Chips

Google is splitting its eighth-generation tensor processing units into separate chips optimized for AI training and inference, a shift the company says reflects the rise of AI agents and their distinct computational needs. The training chip delivers 2.8 times the performance of its predecessor at the same price, while the inference processor (TPU 8i) achieves 80% better performance and includes triple the SRAM of the prior generation. Both chips will launch later this year as Google continues its effort to compete with Nvidia in custom AI silicon, though the company is not directly benchmarking against Nvidia's offerings.

28 days ago· Direct