How AI will change the future of mass surveillance

The most worrying aspect of using AI by governments appears in their endeavor to exploit the capabilities of AI to monitor people – unlawfully – on a large scale.

Artificial intelligence (AI) technology is spreading rapidly worldwide. Prior to the release of ChatGPT, the public was unaware of the robust capabilities of AI systems for handling massive volumes of data and making complex decisions better than humans.

Nowadays, businesses are integrating AI technology in almost all work areas to improve efficacy, reduce operational costs, and achieve competitive advantages. Governments also work to harness the power of AI in different arenas to serve the public.

For instance, they utilize it to monitor traffic congestion, analyze vast volumes of data to make better-informed decisions and leverage it across different use cases in transportation, security, and the environment.

Yet, the most alarming facet of governments deploying AI lies in their pursuit to harness its capabilities for the unlawful mass surveillance of people.

This article will discuss how governments and other interested parties, such as social media platforms and giant tech companies, can leverage AI tools to enhance mass surveillance programs.

However, before I begin, it is worth examining how the usage of AI technologies will result in a significant boost in mass monitoring compared with traditional spying techniques.

Comparison with traditional spying methods

The traditional spying methods that rely on human efforts are limited in scope and ability. For example, conventional spying methods involve tapping target phone to monitor conversations and following target persons physically from one place to another to log their movements.

However, with the advance of the internet and the increased reliance on it to perform many of our daily activities, governments and other commercial companies can monitor everything we do online and connect it – using different ways – to our real identity, creating a complete profile for every internet user. For example, our online purchases, search queries we submit to search engines, social connections, geographical locations we visit, and anything we do online can be recorded and traced back to our real identity.

AI surveillance will allow interested parties to exploit the power of AI to collect and analyze the tremendous volume of data available online and connect it to each person.

However, it is worth noting that surveillance is not limited to people's online activities; we mean what people do on the internet. For instance, surveillance cameras powered by AI technology can recognize individuals' faces quickly and track them across the city.

Most smartphone manufacturers provide the ability to unlock your device using facial recognition or fingerprint, which entitles recording your biometric data and storing it in the device. Google, Amazon, and Apple have AI-powered assistants (Google – Google Assistant; Amazon – Alexa; Apple – Siri) designed to understand verbal instructions and complete requested tasks to aid and enhance the user experience on their respective platforms.

Digital assistants will always listen to the microphones of people's computing devices. It’s not always clear whether they’re recording anything or even sending collected data somewhere.

AI's role in automating surveillance

No doubt, the proliferation of AI technology will provide significant aid to improving the capabilities of governments when conducting mass surveillance across different use cases and not only enhance the monitoring of internet communications.

Data management

AI-powered systems will change the landscape of data management in radical ways. For instance, by 2025, there will be more than 180 zettabytes of data. This massive amount of digital data generated from the rapid acceleration of digital transformation will provide numerous opportunities for mass surveillance and internet user profiling activities. AI tools will significantly facilitate collecting and analyzing massive amounts of data via the following:

• Data collection: AI-powered web scrapers can gather online data from various sources without human intervention. In addition, these scrapers can collect unstructured data easily compared to traditional web scrapers tools. For instance, the web contains many unstructured data, such as textual, multimedia, metadata, and raw data, to name a few. Textual data include social media posts and forum threads, such as Reddit. Multimedia content includes images (e.g., visual content including infographics and memes), videos (such as those generated by TikTok and YouTube users), and audio content (e.g., podcasts and audio recordings). Using traditional tools to get information from content that does not conform to a specific data model or is not organized in a predefined manner is no longer applicable. Natural Language Processing (NLP) allows AI tools to adapt quickly to collect unstructured data without needing a predefined template for each data model on which to base their collection work.
• Data mapping: After gathering the data, we must find patterns and entity relationships. AI tools and Machine Learning (ML) algorithms can do a great job in mapping data automatically from sources to their target destination. For instance, by using ML algorithms, we can discover relationships between a specific organization and its digital interactions with other entities, whether people or organizations, without needing to discover these relationships manually, which can take a tremendous amount of time.
• Data quality: Collecting and mapping data is not the final phase when gathering and analyzing a mass volume of data. For instance, we still need to discover any errors or inconsistencies in the collected datasets before using them. AI tools can help in this direction by discovering anomalies in datasets and filling missing data with estimated values without sacrificing their accuracy.

Speech analysis

Another area in which AI technology can help when conducting mass surveillance activities is speech recognition. For instance, AI technology can aid in this direction:

• AI can identify individuals worldwide by inspecting their unique vocal characteristics. For example, this allows sifting across mass volumes of audio recordings to find all those related to a specific person.
• Natural Language Processing (NLP) algorithms can be adapted to extract specific keywords from voice conversations. For instance, finding particular keywords can trigger an action to shift the conversation to be analyzed by a human agent.
• AI can understand the emotional states of the people involved in the voice communication (e.g., stress, anger, excitement, or other emotional states).
• AI tools can create a text transcript of intercepted voice communications automatically. This allows Intelligence services to index tapped voice communications and store them for later analysis.
• The automatic translation of foreign languages is a significant advantage of using AI tools in mass surveillance operations. For instance, AI can automatically translate intercepted digital communications from any spoken language and save them as text or voice recordings.

Simplify tracking objects in public places

Aside from internet communications monitoring, video surveillance systems powered with AI technology can quickly identify and track people in the streets and other public places through facial recognition. This advanced tracking allows governments to have a record of all people's movements when they are outside their homes. While this behavior is suitable for monitoring suspicious people, conducting this on a mass will be against people's privacy.

On the other hand, AI-enabled surveillance cameras can automatically read vehicle license plate information. This enables tracking of cars as they move through cities in real-time. Further cross-referencing these vehicle movement patterns with databases tracking individuals' web browsing histories, online purchases, and physical movement patterns (derived from facial recognition systems) can provide governments with a comprehensive profile of each citizen.

These integrated AI-powered monitoring networks would allow governments to maintain a detailed record of where a person has been, everything they have bought or searched for online, and pinpoint records of their real-world movements, whether traveling by foot or using their private car.

Currently, AI plays an increasingly significant role in mass intelligence gathering. In the future, we can expect to see more involvement of AI tools in collecting, analyzing, and organizing the masses of online data and those gathered by intercepting electronic communications. This will boost governments' and giant tech companies' abilities to track and profile people on a large scale.

The problematic thing is that individuals can’t do anything to stop this invasion.