Impacts and ethics of using Artificial Intelligence (AI) by the Indian Police

Abstract

Purpose

The paper aims to examine the impacts and ethics of utilizing Artificial Intelligence (AI) in Indian policing. It explores both the positive and negative consequences of using AI, as well as the ethical considerations that have be taken into account.

Design/methodology/approach

This study is based on secondary sources of information, such as national and international reports, journal articles, and institutional websites that discuss the use of AI technology by the police in India.

Findings

AI has proven to be effective in policing, from preventing crime to identifying criminals, by detecting potential crimes in advance with fewer resources and in more areas. In India, the police use AI technology not only for facial recognition but also for crime mapping, analysis, and building blocks. However, factors such as caste, religion, language, and gender continue to cause conflict. India has shown a strong interest in using AI technology for policing, and wishes to accelerate its implementation in various policing contexts, including law and order. This paper calls for an assessment of the complexities and uncertainties brought about by new technologies in policing with ethical considerations.

Originality/value

This paper can provide valuable insights for policy-makers, academics, and practitioners engaged in discussions and debates concerning the ethical considerations associated with the adoption of AI tools in policing practices.

Citation

Rani, M. (2024), "Impacts and ethics of using Artificial Intelligence (AI) by the Indian Police", Public Administration and Policy: An Asia-Pacific Journal, Vol. 27 No. 2, pp. 182-192. https://doi.org/10.1108/PAP-06-2023-0081

Introduction

Police system in India

The Indian police forces are primarily governed by the Police Act of 1861, which grants the state government the authority to establish its own police force. Along with the Police Act, the Code of Criminal Procedure (CrPC) and other laws also oversee the functioning of the police system. The Police Act of 1861 outlines the key features of the Indian police system, which include an organized, maintained, and supervised police force at the state level. The Indian police system is structured horizontally, similar to military forces, with different cadres. Within each state, the police forces are further sub-divided into armed and unarmed branches. As per the Indian constitution, the states have exclusive authority to control and regulate the operations of the police, particularly in maintaining law and order, since policing matters fall under state subjects. Although the Constitution designates policing as a state subject, it includes a comprehensive list of related and quasi-police matters in the Union List. For instance, responsibilities such as preventive detention, arms control, ammunition, explosives, extradition, and passport issuance fall solely under the purview of the Central Government.

Artificial Intelligence: its meaning

Artificial Intelligence (AI) is a branch of computer science that involves designing computer machines with AI-generated tools to enable them to work like humans. The term was first coined by American computer scientist John McCarthy in 1955 (Lifschitz, 1991). AI is an umbrella term that encompasses various fields and methods, including machine learning, automation, and robotics, all of which contribute to the development of AI technology (Chandak, 2020).

AI can be classified into two categories: strong AI and weak AI. The concept of strong AI is controversial and has raised concerns about its potential to pose a threat to humanity. On the other hand, weak AI has already been used in numerous applications, giving rise to policy, governance, and legal challenges. These challenges include safety and privacy concerns, as well as issues related to justice and equality (Bajpai and Irshad, 2019).

In the last two decades, India has emerged as a leading hub for AI technology. The National Program for Artificial Intelligence (NPAI) was introduced by NITI Aayog (National Institution for Transforming India) in 2018, with a goal of developing AI for social good. Over the past five years, India has witnessed a rapid rise in AI innovations, with the country filing AI patents globally. According to a report released by the National Association of Software and Service Companies (NASSCOM) in June 2021, India is ranked eighth among the top ten countries in AI, but is the fourth largest producer of scholarly papers related to AI in the world (NASSCOM, 2021).

Police forces across the globe are employing various AI systems to assist in human decision-making. In 2019, NITI Aayog, the apex public policy think tank of the Indian government, created the Institutional Framework Cloud Computing Platform for Artificial Intelligence- AIRAWAT (Koshy, 2019).

According to a 2021 report by the International Data Corporation (IDC), India’s AI market is projected to grow at a compound annual growth rate of 20.2 percent, reaching $7.8 billion by 2025 from $3.1 billion in 2020 (Sachdev, 2023). Globally, the projection is a 39.4 percent compound annual growth rate, with the market expected to reach approximately $422.37 billion by 2028. The report also noted that AI startups have increased fourteen-fold since 2000.

Use of AI in policing

The Indian government has acknowledged the benefits of AI in various sectors, including finance, healthcare, insurance, and transportation. With the cost of computer processing significantly decreasing, the government is now turning its attention towards utilizing AI in law enforcement to combat criminal activities and terrorism. As of January 1, 2022, the Bureau of Police Research and Development (BPR&D) reports that the sanctioned ratio of police personnel to per lakh people is 196.23, but in reality, it is only 152.80 per lakh population. This makes it challenging for police personnel to address the various issues faced by the public (Mudaliar, 2023). The adoption of AI-based strategies is already underway, with police departments worldwide embracing AI software to predict crimes and identify potential suspects in their daily investigations. By running algorithms on extensive datasets, the software not only assists but also modifies police operations. Due to the complexity of the data involved, it is impractical for humans alone to handle it. The implementation of AI not only enhances the efficiency of police work but also extracts valuable insights from the data, aiding in crime prevention and the maintenance of law and order.

Several initiatives in India demonstrate the active utilization of AI by state police forces. For instance, Uttar Pradesh has implemented an AI-enabled video analytics platform called ‘Jarvis’ in 70 prisons to monitor inmates (Ahaskar, 2019). The Delhi Police has deployed the Intelligence Traffic Management System (ITMS), which utilizes AI technology. Within this system, the 3D radar-based Red-Light Violation Detection Camera (RLVD) system is employed at 24 junctions across Delhi to monitor red-light violations, while a gantry-mounted radar-based system detects instances of overspeeding (Chand, 2022). In December 2017, the Telangana Police launched a Smart RoboCop equipped with cameras, GPS, and various sensors including ultrasonic readers, proximity sensors, and temperature sensors. This robot is designed to assist the police in maintaining law and order as well as managing traffic. The robot can handle security at specific locations such as malls, airports, and other public places. It is capable of recognizing individuals, accepting complaints, detecting bombs, identifying suspects, interacting with people, and answering their queries (Janyala, 2017).

There are databases that facilitate the sharing of crime-related information among various police departments. The Crime and Criminal Tracking Networks and Systems (CCTNS) is a component of the Indian Digital Police Initiative. Furthermore, software has been developed to predict crimes, such as the Compstat software employed by the New York City Police Department in the United States. Facial recognition technology is also being used, as seen in the CCTV surveillance system deployed in Pembrokeshire, United Kingdom. Notably, the ABHEDA app has emerged as a result of this growth, which operates in collaboration with a private company in Gurugram and significantly aids police in apprehending criminals. Trinetra, within the ABHEDA app, contains a database of 500,000 criminals, encompassing their photographs, addresses, and criminal histories. This information is obtained through contributions from various departments, including the government railway police. Trinetra not only identifies individual criminals but also provides data regarding their active associates in different parts of the state. However, it is important to acknowledge that these accomplishments sometimes raise concerns regarding the violation of human rights, a price that society may not be willing to pay.

The purpose of this paper is to examine the various AI-based technologies utilized in policing within the Indian context, investigate the perceptions of police officials regarding the implementation of AI in policing, identify the issues and challenges faced by police in incorporating AI technologies, and discuss the ethical considerations associated with the use of AI technologies in policing.

Research methodology

This study primarily relies on secondary sources. Data and information have been collected from various national and international reports issued by different state governments and agencies on AI. Additionally, to gain a comprehensive understanding of the topic, valuable information has been sourced from credible websites such as the European Crime Prevention Network (EUCPN), as well as major journals such as Policing and Society, International Journal of Public Administration, Journal of Scandinavian Studies in Criminology and Crime Prevention, Police Practice and Research, and Justice Quarterly.

Major AI tools used by the Indian Police

• 1.

  Face Recognition Technology (FRT)

Facial recognition technology (FRT) is the primary AI technique utilized in the Indian police system. This technology is crucial for police departments as it is more accurate than humans in matching faces, thus saving officers time. Police officers use image data to identify the faces of criminals and missing persons. FRT falls under the domain of pattern recognition research and technology and employs statistical methods to detect and analyze patterns. It has enabled the automated identification of individuals by comparing digital images of their faces. FRT technology compares video footage captured from various sources such as drone cameras and CCTV with a database of facial images. Several governments across the globe have implemented or are in the process of installing multiple cameras in public areas to identify and analyze potential criminals. Additionally, FRT serves as a valuable tool in locating missing persons. Furthermore, many law enforcement agencies are currently experimenting with live facial recognition, which allows for real-time detection and identification of individuals of interest.

In India, where 240,000 children went missing between 2012 and 2017, AI has become a crucial tool in the effort to locate these missing children and reunite them with their families through facial recognition. Saru Breirle, an India-born Australian businessman and writer, is a notable example of the power of AI in this regard. Breirle was adopted by an Australian couple and separated from his biological family at the age of five in 2012. With the aid of a Google pin, he was able to locate his biological mother using AI technology (Tiwari, 2017).

The Indian police have made use of a new facial recognition app to successfully reunite numerous lost and trafficked children with their families. As part of Operation Smile, Telangana Police has employed face recognition equipment, leading to the successful reunion of over 8 out of 3000 children with their families as of February 2020 (News18, 2020). In addition, the government has utilized face recognition technology to identify the culprits involved in the protest at Delhi’s Red Fort on 26th January 2021 during the farmers’ movement (The Economic Times, 2021).

In March 2020, the Ministry of Home Affairs, Government of India approved the National Automated Face Recognition System (NASRS). This system is aimed at developing a national-level platform that utilizes face recognition technology to investigate crimes and identify criminals, even if they are wearing a face mask or makeup. The NASRS system will be linked with the National Crime Records Bureau (NCRB) data to facilitate the tracing of criminals, including those who have undergone plastic surgeries. Additionally, face recognition technology, including video, has been allowed to be installed at railway stations in Maharashtra and Gujarat since August 2021 (The Hindu, 2021).

In December, India’s Civil Aviation Minister VK Singh informed the Lok Sabha that starting from March 2022, four airports in the country, namely Varanasi, Pune, Kolkata, and Vijayawada, will implement a biometric system using face recognition technology. This system will scan passengers' faces for identification purposes (Chandra, 2021). The plan is to gradually extend this technology to other airports across the country in a phased manner. The government has named this initiative Digi-Yatra and established a company called Digi-Yatra Foundation under the Companies Act 2013 in February 2019. As part of this initiative, a pilot project for a stand-alone Biometric Boarding System has been launched in airports throughout India, with the ultimate goal of expanding it nationwide. The next step involves the proposal of a Digi-Yatra Kendra ecosystem, which will facilitate passenger registration within the system and enable biometric verification of their faces using their Aadhaar images.

• 2.

  Predictive policing

In simple terms, predictive policing involves preventing future crimes by predicting them. AI techniques have proven to be extremely useful in this forecasting and have become important tools for police departments to carry out this task with the available resources.

Short-staffed and budget-strapped police departments have turned to AI systems to aid in crime prediction. For instance, the Reading Police Department in Pennsylvania has implemented a major software that collects and processes historical crime data to calculate and predict where crimes are most likely to occur (Jenkins and Purves, 2020). This technique is believed to be capable of preventing crimes before they happen. These predictive programs collect and calculate crime data, categorizing future suspects as low, medium, or high risk, such as the HART (VIT) system used by police departments in the United Kingdom. These programs base their predictions on evidence that crimes are likely to reoccur in certain geographic areas or be committed by the same criminals. This greatly enhances the efficiency of police in dealing with crimes within their jurisdiction. In early 2015, the Delhi Police utilized software based on CMAPS (Crime Mapping Analytics and Predictive System) to identify criminal hotspots in the city (Singh, 2017). This software looks for patterns and correlations in past crime data. The system is also being used by the Gurugram and Kolkata Police. In India, where the police-citizen ratio is inadequate, AI provides remarkable and unprecedented solutions.

• 3.

  Pre-trial release and parole

Pre-trial release and parole are additional areas where AI is being utilized in various countries. These systems evaluate the risk associated with releasing an accused individual and, by analyzing complex datasets, determine whether the offender should be granted parole. These datasets are generated by collecting historical crime data and personal information about the offender. In the United States criminal justice system, for instance, the WDC (Wisconsin Data Collaborative) is employed to conduct a baseline risk assessment that helps determine the duration of parole to be granted to a convicted individual.

• 4.

  Crime scene analysis

AI can also be utilized to analyze crime scene data, such as fingerprints and other evidence, to assist analysts in finding traces or identifying suspects. Automated crime detection through AI can increase the speed at which crimes are solved, including theft, robbery, and the identification of dangerous situations. This technology enhances the accuracy of crime detection by improving the entire process.

• 5.

  Sentencing analytics

AI can also be employed to analyze data on previous sentences for similar offenses and offer recommendations on the appropriate sentence for a defendant. This application of AI, known as sentencing analytics, provides valuable insights to support the decision-making process in determining suitable sentences.

• 6.

  Transcription

AI-powered transcription technology is also useful in transcribing audio and video evidence, thereby saving time and resources for law enforcement. This technology provides an accurate and efficient way of transcribing recordings, reducing the need for manual transcription and freeing up resources for other important tasks.

Ethical considerations of using AI technology

It is evident that the use of AI has increased in India and worldwide, proving highly effective in maintaining law and order and apprehending criminals. However, there are limitations to this technology that raise questions about its significance. Ethical concerns raised by human rights advocates further add to the debate. In a country like India, which lacks robust data privacy laws, these concerns become even more significant. In situations where the algorithm incorrectly identifies a match, leading to a false arrest, the responsibility for this error falls into question. Who should be held accountable for the actions of the matching machine? Another concern is with predictive policing, where past data is used to estimate crime areas and types. In India, where the police have faced accusations of bias towards marginalized social groups and communities, the use of biased historical data raises concerns. The 2020 Palghar lynching incident was widely reported in the media, whereby the police remained inactive while the mob killed innocent people on the pretext of child-kidnapping (Singh, 2020). Similarly, some reports have accused Uttar Pradesh police of detaining 41 children and even torturing some of them during the protests against the Citizenship Amendment Act (Suresh and Ali, 2020). The most recent case of police bias and brutality was reported during the 2023 Manipur ethnic conflict, i.e., the state police was accused to be a mute spectator until the pressure mounted and the government was forced to bring in the Central Bureau of Investigation (CBI) to investigate the incident (The Indian Express, 2023).

In conclusion, although AI technology has demonstrated potential in law enforcement, it is essential to acknowledge and address its limitations, ethical concerns, and potential biases to ensure its responsible and fair utilization in upholding law and order.

There are several ways in which the Indian police can fulfill their human rights obligations and be more ethical while implementing AI. These include: (a) ensuring that clear lines of accountability are established for AI-related actions and outcomes, (b) implementing mechanisms to detect and mitigate bias in AI algorithms to prevent discriminatory outcomes, particularly in areas like predictive policing, (c) establishing secure data storage and access controls by creating independent oversight bodies to monitor and evaluate AI implementation, (d) ensuring compliance with ethical standards and collaborating with civil society organizations, human rights advocates, and technology experts to gain diverse perspectives on AI implementation, and (e) establishing channels for citizens to report concerns or complaints related to AI usage, with transparent procedures for addressing them.

If the training dataset primarily consists of certain types of faces, such as women, children, and ethnic minorities, it will result in a biased performance. The rapid implementation of such technology has led to concerns about a surveillance state, particularly among minority communities. For instance, a report by the Vidhi Center for Legal Policy highlighted that police stations in Delhi were disproportionately concentrated in areas with under-represented Muslim populations (Bopanna, 2020). Consequently, these areas experienced heightened policing and data collection. This imbalance implies that areas with a higher density of CCTV cameras are subjected to excessive surveillance, over-policing, and consequently, a higher likelihood of errors compared to other areas. While face recognition techniques may demonstrate better performance in detecting the over-represented faces in their training databases, it is important to note that this technology is rarely flawless and frequently leads to misidentification of individuals. In other words, innocent people are more susceptible to being wrongly identified as criminals or suspects.

The need for responsible AI

Despite the impressive advancements in the competency and capabilities of governments resulting from the integration of AI, there is a growing demand for responsible management of such techniques. Policy makers are well aware of the potential shortcomings, biases, and prejudices associated with AI, and thus, responsible AI is emerging as an approach that prioritizes ethical and legal considerations. This approach aims to minimize the risks of bias, exclusivity, and inequality, and instead promotes equity and ethical behavior in the use of AI techniques. Responsible AI advocates for greater transparency and accountability in the use of these techniques.

Responsible AI development in India involves the practice of creating and implementing AI in a manner that aligns with ethical and societal values. As a leader in AI development, India recognizes that this technology has the potential to bring both positive and negative impacts to society. Consequently, several organizations in India are actively working towards promoting responsible AI. The Ministry of Electronics and Information Technology (MeitY) has established a National Strategy for Artificial Intelligence, which serves as a roadmap for AI development in India (Press Information Bureau, 2022). Additionally, various think tanks, industry bodies, and academic institutions in India are working to create ethical frameworks for AI development. For instance, the Centre for Internet and Society (CIS) has developed the “AI and Ethics in India” manifesto, which outlines a set of principles for responsible AI.

In India, significant efforts have been made towards the creation of Responsible AI for All, as evidenced by the publication of a policy document titled “Principles for Responsible AI” by NITI Aayog in February 2021 (NITI Aayog, 2021). Moreover, the Government of India collaborated with industry and academia to organize the Responsible AI for Social Empowerment - RAISE 2020 summit. The summit brought together experts from around the world to exchange ideas and establish a roadmap for utilizing AI in various sectors, including healthcare, agriculture, education, and smart mobility. The primary objective of the summit was to leverage AI as a tool for promoting social empowerment, inclusivity, and transformative impact.

When addressing responsible AI, it is essential to adhere to overarching principles and ensure that their implementation creates a favourable environment for the adoption of a responsible AI ecosystem within the country. Moreover, collaboration among the government, academia, and industry stakeholders is crucial in developing AI technologies that meet societal needs while effectively addressing concerns related to transparency, accountability, and privacy. Responsible AI development holds significant importance in India, given the country’s distinct social, economic, and cultural landscape. It is imperative to design and deploy AI technologies in a manner that prioritizes the public good, including the empowerment of marginalized communities, rather than perpetuating societal biases and inequalities.

Suggestions for effective implementation of AI in policing

It is evident that AI systems are not entirely free from fault, which is why it is essential to take necessary steps before implementing them on a large scale. at a rapid pace. Here are a few key steps that need to be considered:

• 1.

  Defining AI legally. The regulation of AI is an ongoing debate in the United States. Countries like China, Japan, and Korea have already established rules for automatic or self-driving cars (West, 2016), but India has yet to take appropriate measures in this regard. It is crucial for every nation to establish a legal framework that safeguards human rights when utilizing AI technologies.

• 2.

  Fixation of responsibility is a crucial requirement when it comes to AI systems. The question of who holds the ultimate responsibility for decisions made by machines is paramount. It is important to recognize that machines cannot replicate human feelings and emotions, making it inappropriate for them to possess decision-making power in such matters. AI systems should only be operated under human supervision. Additionally, for AI-based products and equipment, strict responsibilities should be assigned to their producers and manufacturers. Every nation should establish a legal framework that evaluates the impact on human rights before implementing any AI-based system. One of the major drawbacks of AI techniques is their reliance on prior data. In countries like India and the United States, where instances of police bias exist (Times of India, 2020), the uncritical implementation of AI technology in policing poses the risk of perpetuating biases. To prevent prejudice and bias in AI-based policing, it is crucial to ensure that the training data used in algorithms is as unbiased as possible. This requires collecting data from diverse sources and conducting rigorous reviews to identify and address any existing bias.

• 3.

  To ensure the success of AI systems, it is imperative to prevent discrimination stemming from biases and guarantee diversity in the data used. Privacy is a fundamental right, as affirmed by the Supreme Court in the case of KS Puttaswamy vs. Union of India (Supreme Court Observer, 2017). The Court recognized the right to privacy as a fundamental right under Article 21 of Part III of the Constitution. It is essential to use this form of intelligence with utmost care to avoid violating the right to privacy. The first step in avoiding any biases in AI-based policing is to ensure that the training data fed into the algorithm is as unbiased as possible. This requires collecting data from a diverse pool of sources and subjecting it to rigorous review to identify any existing biases.

• 4.

  Advanced algorithms such as random forests, decision trees, and neural networks can aid the Indian police in making more accurate decisions based on the analysis of big data. These algorithms have the capability to modify any implicit biases in the data. The use of Explainable AI systems such as SHAP and LIME can authenticate the output of these advanced algorithms.

• 5.

  AI based surveillance systems can identify any signs of prejudice and bias in police activity. These systems can detect patterns of discrimination, excessive use of force, or other biased behaviors among police personnel.

• 6.

  AI-enabled tools can assist police departments in crime prediction, criminal identification, and forensic analysis to prevent any erroneous or biased outcomes.

• 7.

  Adequate publicity should be given to AI technology and equipment, and people should be educated about AI. In the absence of sufficient knowledge of this technology, its implementation can lead to the violation of human rights.

In a country like India, where the majority of the population has not been able to achieve literacy in the seven decades since independence, it is not an easy task to keep them informed about AI and its effects (India Today, 2023). Therefore, it is important to adopt effective laws to ensure that human rights are not infringed upon. Unfortunately, there is currently no such law in India. Clause 12 of the proposed Personal Data Protection Act excludes the government and its agencies from its purview. This view is also supported by the Joint Parliamentary Committee’s report on the Act in December 2021, which makes it clear that the Act only protects the right to privacy against private agencies, whereas the Constitution provides protection against the state as well (Mustafa and Leo, 2021). This effectively upholds the government’s authority to oversee and leaves citizens’ confidential information at the mercy of both companies and the government.

Potential AI applications in policing

AI has the potential to revolutionize the operations of law enforcement, offering a wide array of capabilities and opportunities to enhance existing systems. However, as AI is implemented, it is crucial to address concerns regarding privacy, bias, and ethical considerations. It is of utmost importance to use AI responsibly and ethically, ensuring that it does not contribute to discrimination or harm against specific individuals or communities.

• a)

  Deploy AI-powered surveillance systems: The government can implement advanced AI-powered surveillance systems in public areas and on roads to monitor movements and identify potentially suspicious activities.

• b)

  Predictive policing: AI algorithms can be utilized to identify high-risk areas for crime, enabling law enforcement agencies to allocate resources and personnel effectively.

• c)

  Facial recognition: AI-powered facial recognition software can be implemented to identify criminals in public spaces, such as on the streets, in public transportation, or in crowded areas.

• d)

  Natural Language Processing: AI-powered chatbots and voice assistants can be put in place to improve communication between citizens and law enforcement officials, streamlining the reporting of crimes and queries.

• e)

  Crime pattern analysis: AI-based algorithms are becoming increasingly crucial for analyzing data from different sources, such as social media posts and surveillance footage, to detect crime patterns and identify potential signals of criminal activities.

• f)

  Cybersecurity: AI technology is being deployed to track and detect the growing instances of cybercrime activities in India and countries worldwide.

• g)

  Intelligent evidence analysis: AI-tools are now being utilized to analyze forensic evidence, such as DNA, fingerprints, and criminal records, in a more efficient and effective manner.

• h)

  Risk assessment and pre-emptive measures: AI-based systems are important in identifying potential threats and risks in various contexts, such as crowd control measures, enabling proactive measures to be taken to prevent incidents from occurring.

• i)

  24/7 Monitoring: AI is ideally suited for continuous 24/7 monitoring, which is crucial in law enforcement. It reduces the reliance on human operators for round-the-clock supervision.

• j)

  Training and simulations: AI-powered technologies can be used to create realistic simulations for training sessions, enhancing the preparedness of law enforcement agencies.

Conclusion

The paper delves into the growing use of AI by Indian police for crime prevention and detection, highlighting the associated risks and contentious nature of this issue. It also explores various practical applications of AI technology in government agencies and police departments in India, including AI-enabled databases, platforms, and initiatives. The paper acknowledges the achievements of police organizations in maintaining law and order, identifying criminals, and managing prisons. However, it also outlines ways in which Indian police can become more ethical in fulfilling their human rights obligations and provides critical recommendations for the effective implementation of AI in Indian policing.

The increasing significance of AI systems and associated tools is a prominent trend. In a densely populated country like India, where there is a relative scarcity of police personnel, AI plays a crucial role in maintaining law and order, anticipating crimes, and apprehending criminals. AI tools have proven beneficial in connecting families, selecting suitable beneficiaries for various schemes, and identifying unclaimed deceased individuals. However, it is important to acknowledge the limitations of these systems. Criticisms regarding privacy infringement and human rights violations highlight that AI cannot be the sole solution to policing challenges. Its effective and optimal use is only possible with human supervision, robust laws, and diverse datasets. Therefore, further research is warranted, particularly focusing on AI models capable of reducing biases, enhancing facial recognition technology, and developing tools for early crime detection. This necessitates strong collaboration between AI experts, law enforcement agencies, non-governmental organizations, and policymakers.