HPE says hackers breached Aruba Central using stolen access key
FBI warns of Iranian hackers looking to buy US orgs’ stolen data
Telnyx is the latest VoIP provider hit with DDoS attacks
NUCLEUS:13 TCP security bugs impact critical healthcare devices
Gmail accounts are used in 91% of all baiting email attacks
Careful: ‘Smart TV remote’ Android app on Google Play is malware
Void Balaur hackers-for-hire sell stolen mailboxes and private data
HPE says hackers breached Aruba Central using stolen access key
Qualys BrowserCheck
Junkware Removal Tool
How to remove the PBlock+ adware browser extension
Remove the Toksearches.xyz Search Redirect
Remove the Smashapps.net Search Redirect
Remove the Smashappsearch.com Search Redirect
Remove Security Tool and SecurityTool (Uninstall Guide)
How to remove Antivirus 2009 (Uninstall Instructions)
How to Remove WinFixer / Virtumonde / Msevents / Trojan.vundo
How to remove Google Redirects or the TDSS, TDL3, or Alureon rootkit using TDSSKiller
Locky Ransomware Information, Help Guide, and FAQ
CryptoLocker Ransomware Information Guide and FAQ
CryptorBit and HowDecrypt Information Guide and FAQ
CryptoDefense and How_Decrypt Ransomware Information Guide and FAQ
How to make the Start menu full screen in Windows 10
How to install the Microsoft Visual C++ 2015 Runtime
How to open an elevated PowerShell Admin prompt in Windows 10
How to Translate a Web Page in Google Chrome
How to start Windows in Safe Mode
How to remove a Trojan, Virus, Worm, or other Malware
How to show hidden files in Windows 7
How to see hidden files in Windows
IT Certification Courses
Gear + Gadgets
A team of researchers at the Imperial College in London have presented a simple method to evade detection by image content scanning mechanisms, such as Apple’s CSAM.
CSAM (Child Sexual Abuse Material) was a controversial proposal submitted by Apple earlier this year. The proposal was eventually retracted in September, following strong backlash from customers, advocacy groups, and researchers.
Apple hasn’t abandoned CSAM but rather postponed its roll-out for 2022, promising new rounds of improvements and a more transparent approach in its development.
The main idea is to compare image hashes (IDs) of pictures shared privately between iOS users to a database of hashes provided by NCMEC and other child safety organizations.
If a match is found, Apple’s reviewers will look into the content and alert the authorities of the distribution of child abuse and pornography, all without compromising the privacy of people who share legal images (non-matches).
This theoretically sounds like a good system to prevent the dissemination of harmful material, but practically, it inevitably opens a “Pandora’s box” for mass surveillance.
However, the question that researchers at the Imperial College in London asked is, would such a detection system even work reliably in the first place?
The research presented at the recent USENIX Security Symposium by British researchers shows that neither Apple’s CSAM nor any system of this type would effectively detect illegal material.
As the researchers explain, it’s possible to fool content detection algorithms 99.9% of the time without visually changing the images.
The trick is to apply a special hashing filter on the images, making them appear different to the detection algorithm even if the processed result looks identical to the human eye.
The paper presents two white-box and one black-box attack for discrete cosine transform-based algorithms, successfully altering an image’s unique signature on a device and helping it fly under the radar.
A possible countermeasure to the evasion methods presented in the paper would be to use a larger detection threshold, leading to an increase in false positives.
Another approach would be to flag users only after image ID matches reach a certain threshold number, but this introduces probability complications.
Applying additional image transformation before computing the perceptual hash of the image is also unlikely to make detections any more reliable.
Increasing the hash size from 64 to 256 would work in some cases, but this introduces privacy concerns as longer hashes encode more information about the image.
All in all, the research demonstrates that current perceptual hashing algorithms are not nearly as robust as they should be for adoption in illegal content distribution mitigation strategies.
“Our results shed strong doubt on the robustness to adversarial black-box attacks of perceptual hashing-based client-side scanning as currently proposed. The detection thresholds necessary to make the attack harder are likely to be very large, probably requiring more than one billion images to be wrongly flagged daily, raising strong privacy concerns.” – concludes the paper.
This is a significant finding coming at a time when governments are considering hash-based invasive surveillance mechanisms.
The paper shows that for illegal image detection systems to work reliably in their current form, people will have to give up their privacy, and there’s no technical way around this at this time.
Tor Browser 11 removes V2 Onion URL support, adds new UI
New Gummy Browsers attack lets hackers spoof tracking profiles
Credit card PINs can be guessed even when covering the ATM pad
Researcher drops three iOS zero-days that Apple refused to fix
Microsoft patches Excel zero-day used in attacks, asks Mac users to wait
Not a member yet? Register Now
Microsoft urges Exchange admins to patch bug exploited in the wild
Microsoft November 2021 Patch Tuesday fixes 6 zero-days, 55 flaws
To receive periodic updates and news from BleepingComputer, please use the form below.
Terms of Use Privacy PolicyEthics Statement
Copyright @ 2003 – 2021 Bleeping Computer® LLC – All Rights Reserved
Not a member yet? Register Now
Read our posting guidelinese to learn what content is prohibited.


You May Also Like

Phishing attacks use QR codes to steal banking credentials

New zero-day exploit for Log4j Java library is an enterprise nightmareALPHV BlackCat…

XE Group exposed for eight years of hacking, credit card theft

Emotet now drops Cobalt Strike, fast forwards ransomware attacksSonicWall ‘strongly urges’ customers…

Phishing actors start exploiting the Omicron COVID-19 variant

Former Ubiquiti dev charged for trying to extort his employerNew malware hides…

Microsoft Intune bug forces Samsung devices into non-compliant state

AMD fixes dozens of Windows 10 graphics driver security bugsVoid Balaur hackers-for-hire…