From airports to local supermarkets and mobile phone applications, facial recognition technology (FRT) has become increasingly commonplace. That doesn't mean it's accepted as completely benign however, with the pros and cons of facial recognition technology under almost constant discussion.
On the one hand, it can make life easier and processes smoother; many smartphones have the option to login using facial recognition, such as Apple's FaceID on iPhone. It generally speeds up security checks and improves the user experience.
However, campaign groups, such as Liberty and Big Brother Watch in the UK, argue that this technology is dangerous and could lead to the unfair profiling of individuals who have not committed a crime.
How does facial recognition technology work?
Facial recognition technology uses computer vision technology to extract useful information from still images or videos, which is then analyzed by an algorithm to estimate the degree of similarity between two faces.
Facial recognition for bears (and other ways to use the technology for good)
To do this, the algorithm takes into account facial expressions and face geometry. It looks for a number of data points including the distance between the eyes, between the nose and mouth, cheekbone shape, as well as the overall length of the face between forehead and chin.
This will then be transformed into a 'faceprint' – a unique set of biometric data similar to a fingerprint. The facial recognition system can then be used for a variety of use cases. Let's now consider the pros and cons of facial recognition technology.
Pros of facial recognition technology
Improving security systems and identifying criminals are often cited when arguing in favour of facial recognition, as well as getting rid of unnecessary labour or human interaction. However, there are also plenty of other examples.
1. Finding missing people and identifying perpetrators
Facial recognition technology is used by law enforcement agencies to find missing people or identify criminals by using camera feeds to compare faces with those on watch lists.
The technology has also been used to locate missing children. Sometimes it is combined with advanced aging software to predict what a child might look like based on photos taken when they disappeared. Law enforcement agencies often use facial recognition with live alerts to help track potential matches.
2. Protecting businesses against theft
Facial recognition is increasingly being deployed as a means of identifying known individuals before they commit crimes like theft or public affray. It's common to see CCTV in shops and places of work, and by using facial recognition software it's possible to create tools like automatic cross-referencing to match individuals to a database of known suspects.
The technology has the dual purpose of helping to prevent crime before it happens, and also – some would argue – a deterrent effect for would-be offenders.
If something is stolen from the business, the software can also be used to catalogue the thieves for future reference.
3. Better security measures in banks and airports
Facial recognition has also come to be used as a preventative security measure in sensitive locations such as banks and airports. Similar to identifying criminals that come into shops, the software has helped identify criminals and passengers that pose a potential risk to airlines and passengers.
Border checks have also been sped up at some airports through the use of facial recognition cameras at passport-check gates.
Institutions like banks use the software in the same way to prevent fraud, identifying those previously charged with crimes and alerting the bank to watch specific individuals more carefully.
5. Drastically reduces human touchpoints
Facial recognition requires fewer human resources than other types of security measures, such as fingerprinting. It also doesn’t require physical contact or direct human interaction. Instead, it uses artificial intelligence (AI) to make it an automatic and seamless process.
It also limits touchpoints when unlocking doors and smartphones, getting cash from the ATM or performing any other task that generally requires a PIN, password or key.
5. Better tools for organising photos
Facial recognition can also be used to tag photos in your cloud storage through iCloud or Google Photos. Users who wish can enable facial recognition in their respective photo app’s settings, resulting in named folders for regular photo subjects. Facebook also used facial recognition to suggest people to tag within a photo.
6. Better medical treatment
One surprising use of facial recognition technology is the detection of genetic disorders.
By examining subtle facial traits, facial recognition software can, in some cases, determine how specific genetic mutations caused a particular syndrome. The technology may be faster and less expensive than traditional genetic testing.
Cons of facial recognition
As with any technology, there are drawbacks to using facial recognition, such as the violation of rights and personal freedoms that it presents, potential data theft and the risk of overreliance on inaccurate systems.
1. Greater threat to individual and societal privacy
The threat of technology intruding on an individual's rights to privacy is perhaps the greatest threat created by extensive use of facial recognition.
Privacy is now a critical issue, so much so that in some cities across the likes of California and Massachusetts, law enforcement agencies are banned from using real-time facial recognition tools. Police instead are forced to rely on recorded video from the likes of body-worn cameras.
In 2021, then UK Information Commissioner Elizabeth Denham described the use of live facial recognition (LFR) cameras in public spaces as "deeply concerning".
2. Infringement on personal freedoms
It's just not personal privacy that is potentially at risk with mass use of facial recognition – the simple act of being recorded or scanned by the technology could discourage individuals from moving freely around their local neighbourhood or city.
The argument here is that people simply do not want to feel like they are overtly being watched, judged, or recorded.
The basic premise of facial recognition is to match everyone to a database of known suspects, essentially treating you as if you are a criminal suspect without probable cause. It's a concept that some people find inherently dangerous to public freedoms.
For example, the aforementioned example of facial recognition being used to catalogue potential shoplifters has led to problems for companies such as Southern Co-operative, which in 2022 faced a legal complaint for its widespread use of FRT CCTV in its shops.
3. Violation of personal rights
When used for identification purposes, facial recognition data is considered as part of the ‘special category’ of personal data under the UK's implementation of the GDPR. This also extends to racial or ethnic origin, and some facial recognition CCTV companies have been accused of.
In July 2022, a cross-party group of 67 MPs called for surveillance equipment from Chinese firms Hikvision and Dahua to be banned from use in the UK, citing concerns over ethics and security. These were informed by stories such as a report by the LA Times alleging that Dahua developed software to allow its cameras to detect Uighur minorities and issue law enforcement users with a warning upon successful detection.
4. Creates data vulnerabilities
Facial recognition also creates a data protection and cyber security headache. The large volume of personally identifiable information (PII) being collected and stored is an attractive target for cyber criminals, and there are already examples of hackers gaining access to such systems.
This data is particularly sensitive given that many online services, such as banking, are increasingly utilising biometric data as part of their multi-factor authentication. A threat actor with access to a database of facial data could have the tools to bypass such checks, and access even more sensitive information.
5. Provides opportunities for fraud and other crimes
Lawbreakers can use facial recognition technology to perpetrate crimes against innocent victims too. They can collect individuals’ personal information, including imagery and video collected from facial scans and stored in databases, to commit identity fraud.
With this information, a thief could take out credit cards and other debt or open bank accounts in the victim’s name. In consideration of the aforementioned use of facial recognition to place shoplifters on criminal databases, threat actors could even place individuals on a criminal record.
Beyond fraud, bad actors can harass or stalk victims using facial recognition technology.
For example, stalkers could perform reverse image searches on a picture taken in a public place to gather information about their victims, to better persecute them.
Facial recognition law has lagged behind potential use by bad actors in recent years, which has prompted calls from rights groups for stricter biometrics regulations, to extend to technologies such as live facial recognition.
6. The technology is imperfect
Facial recognition is far from perfect, and cannot be relied upon to produce accurate results in place of human judgement.
The technology depends upon algorithms to make facial matches. Those algorithms are more effective for some groups, such as white men than other groups such as women and people of colour due to lack of representation within the data set on which the algorithm was trained. This creates unintentional biases in the algorithms, which could in turn translate to biases in whatever action the technology is informing, such as arrests.
In 2018, civil liberties organisation Big Brother Watch published evidence that facial recognition technology utilised by the Metropolitan Police Service (MPS) was incorrectly identifying innocent people as criminals 98% of the time.
7. Innocent people could be charged
Following on from the imperfection of facial recognition, there are inherent dangers in false positives. Facial recognition software could improperly identify someone as a criminal, resulting in an arrest, or otherwise cause them reputational damage if they were to be included on, for example, a list of shoplifters.
8. Technology can be fooled
Other factors can affect the technology’s ability to recognize people’s faces, including camera angles, lighting levels and image or video quality. Mild alterations of facial data, such as a false moustache, can trick weaker facial recognition systems, while especially poor facial recognition technology could simply be tricked with a photo of a face it recognises.
As facial recognition technology improves, its flaws and the risks associated with it could be reduced. Other technology is also likely to be used in tandem with facial recognition technology to improve overall accuracy, such as gait-recognition software.
For the time being, though, the technology’s inadequacies and people’s reliance on it means facial recognition still has much room to grow and improve.
What are examples of facial recognition software or apps?
Although you might not know it, there’s plenty of examples of facial recognition software available on the market today. This ranges from options provided by tech giants, to software created and fine tuned by smaller companies. Here’s a selection of a few that are available on the market today, some with free options available too.
Microsoft Azure AI Face
This allows you to embed facial recognition technology into any apps you create. The good news is that you don’t need any machine learning knowledge, you just plug in the API and you’re good to go. It contains face detection and can identify a person by matching the face to a private database or through photo ID.
It has a free tier, with 30,000 transactions free per month, or the standard tier which starts at $1 (£0.81) per 1,000 transactions up to a maximum of 1 million transactions per month.
Amazon Rekognition
Amazon Rekognition is the tech giant’s computer vision APIs that you can add to your apps without needing to spend time building machine learning models. It claims to be able to analyse millions of images or videos in seconds. Some of the features include face compare and search, text detection, and video segment detection.
It has a free tier which lasts for 12 months, where you can analyse 5,000 images per month and store 1,000 face metadata objects per month for free. Its paid tier varies depending on how many images you plan to analyse per month.
Kairos
This company provides another API that developers and businesses can use to easily integrate into their software or applications. Its features include gender detection, age detection, multi-face detection, and face verification.
Pricing starts at $19 per month for the Student Cloud, while developers will pay $99 and businesses $249 per month. Each tier supports a different amount of transactions per minute, and these are priced at $0.002 per transaction.
