google has come under heavy fire ever since it unveiled a new photo-scanning feature for Android phones. google has come under fire from users who claim that the company installed a new spy app on Android smartphones without their knowledge or permission. Serious questions concerning privacy and control over personal data have been raised by this action.
Google's Statement and Reassurance
According to forbes, google promised users that the technology would not begin scanning images or other content without their consent when the new function was first unveiled. The company claims that SafetyCore is a framework that aids in the private and secure classification of content on the device. "SafetyCore enables on-device infrastructure to help users detect unwanted content," google said, stressing that it will only categorize certain content upon request from apps via an optional feature. Additionally, the business assured customers that the technology would only function on the device and would not transmit any information to Google.
Google's 3 billion Android, email, and other users will need to decide where they draw their lines and how much AI scanning, monitoring, and analysis they can tolerate. Many of the latest versions lack the same privacy protections, even though certain functionality is on-device.
The reality: google messages start scanning sensitive content
The moment has arrived for the feature to start working, though, as forbes now reports, and google Messages will be the first to implement the new features. As reported by 9to5Google, "Google Messages is now issuing sensitive content warnings that blur nude images on Android devices." This not only makes the visuals blurry but also serves as a warning that the content can be dangerous. Users have the choice to either block the numbers linked to such information or view the image in its clearest form.
On-device AI scanning: Google's assurance
Google has emphasized that no data is transmitted back to the firm because the scanning takes place locally on the device. GrapheneOS, an Android hardening project, backed up this claim by confirming that SafetyCore doesn't send any information back to google or any other organization. SafetyCore "provides client-side scanning for use in experiments that classify content as spam, scams, malware, etc." according to GrapheneOS. This implies that apps can issue warnings and verify content locally on the device without exchanging any data with third parties.
GrapheneOS voiced concerns regarding the new system's lack of transparency despite these assurances. The Android Open Source team and the underlying machine learning models are not publicly available, and the team bemoaned the fact that SafetyCore is not open-source. The absence of open-source access raises concerns about the possibility of misuse or lack of user control, even if GrapheneOS has no problems with local neural network functionality.
In conclusion, although Google's new photo-scanning technology provides a certain amount of security and privacy by locally scanning information on Android devices, many people are doubting the technology's genuine nature due to its lack of open-source openness. The conflict between user privacy and the growing usage of AI for content moderation is brought to light by the ongoing discussion.
Google's Statement and Reassurance
According to forbes, google promised users that the technology would not begin scanning images or other content without their consent when the new function was first unveiled. The company claims that SafetyCore is a framework that aids in the private and secure classification of content on the device. "SafetyCore enables on-device infrastructure to help users detect unwanted content," google said, stressing that it will only categorize certain content upon request from apps via an optional feature. Additionally, the business assured customers that the technology would only function on the device and would not transmit any information to Google.
Google's 3 billion Android, email, and other users will need to decide where they draw their lines and how much AI scanning, monitoring, and analysis they can tolerate. Many of the latest versions lack the same privacy protections, even though certain functionality is on-device.
The reality: google messages start scanning sensitive content
The moment has arrived for the feature to start working, though, as forbes now reports, and google Messages will be the first to implement the new features. As reported by 9to5Google, "Google Messages is now issuing sensitive content warnings that blur nude images on Android devices." This not only makes the visuals blurry but also serves as a warning that the content can be dangerous. Users have the choice to either block the numbers linked to such information or view the image in its clearest form.
On-device AI scanning: Google's assurance
Google has emphasized that no data is transmitted back to the firm because the scanning takes place locally on the device. GrapheneOS, an Android hardening project, backed up this claim by confirming that SafetyCore doesn't send any information back to google or any other organization. SafetyCore "provides client-side scanning for use in experiments that classify content as spam, scams, malware, etc." according to GrapheneOS. This implies that apps can issue warnings and verify content locally on the device without exchanging any data with third parties.
GrapheneOS voiced concerns regarding the new system's lack of transparency despite these assurances. The Android Open Source team and the underlying machine learning models are not publicly available, and the team bemoaned the fact that SafetyCore is not open-source. The absence of open-source access raises concerns about the possibility of misuse or lack of user control, even if GrapheneOS has no problems with local neural network functionality.
In conclusion, although Google's new photo-scanning technology provides a certain amount of security and privacy by locally scanning information on Android devices, many people are doubting the technology's genuine nature due to its lack of open-source openness. The conflict between user privacy and the growing usage of AI for content moderation is brought to light by the ongoing discussion.
click and follow Indiaherald WhatsApp channel