What is a Digital Fingerprint?
Digital Fingerprints are a unique digital signature or digital identifier much like a person’s fingerprint. Every computer user and internet surfer has a digital fingerprint that contains data that identifies the web browser used, some plugins being used, IP address, devise information, CSS information, clock information and even preferred fonts that are used . These digital fingerprints are unique to every user on the internet. Now with Artificial Intelligence software identifying and finding these unique digital fingerprints is becoming much easier.
The Internet is vast space with many dark corners. From these dark corners individuals are using bots and fake profiles while trying to hide behind their VPN and private browsing to find young and vulnerable victims to human traffic or use in child pornography. A 2020 report from a cyber security firm found that more than one quarter (27%) of the profiles on the internet were fake. These users simply engage the incognito mode on their web browser and mask their IP address to find and lure young victims.
However, how their fake profiles are written, how they engage their private browsing, to the VPN services they use, or the code they write to mask themselves creates a unique data signature that amounts to an even more unique digital fingerprint that can be tracked by AI. Everything from the metadata on their posted photos, to the cookies they accept, the websites they visit, to their keystroke dynamics, to the device they use creates a unique fingerprint.
The intelligence community in Canada and the United States has created an artificial intelligence program that can identify individuals behind misinformation campaigns. The same AI powered technology is being used to detect and track individuals that are actively involved in human trafficking and the distribution of child pornography. The AI software collects millions of data points from the browser information, to the operating system, IP addresses that the VPN uses, to the browser plugins, fonts and other settings and codes to create digital fingerprints.
Data and Information Contained in a Digital Fingerprinting
The data points and information collected are then processed by the AI software that then calculates to provide a unique digital fingerprints for the various users. Some of the data points that are collected and processed are:
- IP address
- Clock information
- Device MAC address
- User agent string
- Keystroke dynamics
- Web browser configuration – its own fingerprinting
- Web browser plugins
- Fonts installed on device
- CSS information
- Device Screen resolution
- Device operating system and language settings
- Flash data
The data collected to create the unique digital fingerprint actually contains unique identifiers that can reveal the device fingerprint, the browser fingerprint, IP and VPN data.
AI Digital Fingerprint is not Prefect… Yet
Technology is still being perfected, but tracking the activities of individual and groups based on the text they write in their profiles. This new technique will reportedly work similar to other ways forensics experts determine someone’s identity based on their handwriting. Just as humans have tiny little individual differences and idiosyncrasies in the way they write a word, online authors similarly have tells when crafting sentences online that are unique only for them which can be used as an easy way of confirming who is really behind any given piece of content or blog post you might come across while browsing social media platforms like Facebook without having anything concrete proving this person existed at all.
Digital fingerprints can be used to determine the authenticity of digital content. With this technology, government agencies could potentially identify whether an online actor is trying to impersonate someone else or tell if something posing as human is just a bot spouting disinformation.
We’re living in a world where the only thing that can’t be used in the electronic frontier to identify someone is their name. A recent study found fingerprints and facial recognition systems, but there’s also ways for robots or algorithms recognize people by voice-the way they walk; what kind of shoes they have on ; even feces could hold clues about who might be nearby!
Digital Fingerprinting to Identify Child Pornography Makers
Persons involved in human trafficking and the making of child pornography take extraordinary measures to hide their identity and real location. They will use every digital security and privacy method to hide their IP address, to mask their browser fingerprint and their device fingerprint while online. The internet has been a safe place for them to recruit and find victims to lure into their trade and also a place to distribute their products.
Human traffickers have a number of different scripts and methods to lure their victims. Often online they will find young females that they can lure with promises of better lives and money. Usually involving activities that appear legal or that are on the fringe of the sex trade industry. Much of the initial luring and communication is conducted online and using methods to hide identity and tools that are blocking IP addresses of the users.
Once they have their victims under their control the traffickers again turn to the internet to traffick them. This allows investigators another opportunity to find the unique fingerprint of the traffickers.
Examples of Law Enforcement using AI
The use of AI powered technology is not new, but the rate at which AI is being adopted and expand is growing exponentially. Law enforcement in Canada has been using facial recognition technology (FRT) and other biometric technology to identify suspects since the 1990s. Currently the RCMP and the Toronto Police Service as part of their biometric authentication are using facial recognition supplied by Clearview AI. But, facial recognition tech is not the only data driven AI that is being used by police. Voice Recognition will soon have a larger role in identity verification along with other types of biometric authentication that is AI powered. There will be a time that people attend to renew their driver’s licence in front of a AI system kiosk and that with record their voice, and use face recognition, voice recognition and the uses of recognition iris software.
Predictive Policing is also heavily driven by tracking data points to determine the best use and allocation of police services and assets. Not only to determine the best use of police officers scheduling, but tracking of the frequency of crimes and locations. Information collected from arrests, police calls, crime reports generates schedules that allows the most efficient use of police officers and other resources. But, predictive police also includes AI that is able to perform online tracking digital fingerprints to monitor the patterns and behaviour of suspects involved in human trafficking and child pornography.
In 2017 the Dubai Police Service introduced AI equipped police robots to interact with the public in crime reporting. Robots are able to take police reports for crimes committed in six different languages. The robots can be used in high-risk situations in place of officers, patrol high crime areas and monitor traffick. Further, these robots are always collecting data through their biometic systems such as facial recognition and voice recognition to identify wanted people in the area. Dubai expects that a quarter of its human officers will be replaced by robots in the next 10 years all with biometic technology that will be gathering data to further enhance their biometic authentication and biometric security data systems.
Privacy Concerns When Arrested?
Although the use cases of AI in law enforcement can be seen as a positive step forward in the fight against crime, there are concerns about how this technology will be used and the implications it has for our privacy. If you have been charged with a criminal offence that involves police use of AI, please contact me immediately. I am experienced in dealing with these types of AI use cases and I can help you get the best outcome possible.