Crime in the Era of Artificial Intelligence

The development of Artificial Intelligence (AI) has had a profound impact on various sectors of modern life. From industrial automation to large-scale data analytics, AI promises efficiency and precision. However, this same technology also introduces new risks, particularly in the realm of crime and cybersecurity. Two recent academic publications — AI Based Crime Rate Prediction (Nishad et al., 2024) and AI and Serious Online Crime (Burton et al., 2025) — offer contrasting perspectives on how AI plays a dual role: as a predictive tool for crime prevention and as a weapon for digital criminals.

AI for Crime Prediction and Prevention

According to the research by Vishal J. Nishad and colleagues (2024), artificial intelligence possesses an extraordinary capability to predict potential crimes before they occur. By leveraging machine learning and deep learning algorithms, AI systems can identify criminal patterns based on historical, social, economic, and environmental data.

The study proposes a predictive model that combines multiple technologies, such as:

• Long Short-Term Memory (LSTM) networks to analyze crime trends over time.

• Geographic Information Systems (GIS) to map crime locations and identify hotspots.

• Random Forest and Gradient Boosting algorithms to evaluate socioeconomic variables such as education, income, and population density.

This approach enables law enforcement agencies and policymakers to allocate resources more efficiently and design data-driven crime prevention strategies. Furthermore, the use of Explainable AI (XAI) in the model enhances transparency and accountability, allowing officers to understand the rationale behind each prediction (Nishad et al., 2024).

However, the study also highlights major challenges, including limited access to criminal datasets, algorithmic bias, and the need for ethical oversight to ensure that AI systems do not discriminate against specific communities. Therefore, developing fair and ethical AI systems is crucial to ensuring that technology genuinely contributes to a safer society.

AI as a New Weapon in Cybercrime

While AI can help prevent crime, the report AI and Serious Online Crime by CETaS (Burton et al., 2025) warns about the darker side of technological advancement. According to the report, AI has become a key instrument in various forms of online crime — including financial fraud, ransomware attacks, identity manipulation (deepfakes), and child sexual abuse material (CSAM) generation.

Some of the report’s key findings include:

1. A rapid surge in AI-enabled crime.
   AI is being used to automate and scale crimes such as phishing, DDoS attacks, and CSAM production. Automation has made such crimes faster, more extensive, and more difficult to trace.

2. Deepfake-driven financial fraud.
   In 2024, a Hong Kong-based company lost approximately £20 million after an employee was deceived by a fake video call that mimicked the company’s chief financial officer — entirely generated by AI (Burton et al., 2025).

3. AI in online romance scams.
   Criminals now exploit Large Language Models (LLMs) to craft persuasive messages and build emotional relationships with victims. Some even use deepfake-generated faces and voices to enhance their deception.

4. Unethical exploitation of AI systems.
   The CETaS report also exposes AI models such as WormGPT and FraudGPT — custom-built systems stripped of ethical guardrails — used to assist in credit card fraud, phishing, and business email compromise.

This phenomenon illustrates how AI has evolved beyond being a mere criminal tool — it has become a collaborator in cybercrime, helping offenders automate and scale their operations more effectively.

Challenges for Law and Global Security

Burton and colleagues (2025) emphasize that law enforcement agencies worldwide are not yet fully prepared to deal with AI-driven criminal threats. They propose several strategic responses, including:

• Establishing an AI Crime Taskforce within national law enforcement bodies to monitor, analyze, and counter AI-related crimes.

• Implementing AI counter-crime systems, where AI is used to fight AI by detecting and neutralizing malicious algorithms automatically.

• Strengthening international cooperation in cyber law enforcement, given that most AI-based crimes are transnational.

• Providing intensive AI literacy and ethics training for law enforcement personnel to understand the risks, biases, and misuse potential of AI systems.

In this context, AI represents a new “cyber battlefield” where criminals and law enforcement engage in an arms race — one defined by the speed, intelligence, and adaptability of algorithms.

Balancing Prediction and Protection

Both Nishad et al. (2024) and Burton et al. (2025) underscore a central truth: AI is a double-edged sword. It can be the most powerful tool for preventing crime, but it can also become the most dangerous threat when misused.

To navigate this paradox, a multidisciplinary approach is essential, combining:

• Cross-sector collaboration between data scientists, criminologists, and policymakers.

• Strong AI ethics and governance frameworks to prevent bias and protect human rights.

• International regulations and standards to oversee AI applications in the field of digital security.

The principle of “Garbage In, Garbage Out” (Nishad et al., 2024) remains relevant — AI’s accuracy and fairness depend entirely on the quality of the data and governance behind it.

Artificial intelligence has fundamentally transformed the landscape of modern crime. From predicting potential offenses to combating deepfake scams and automated fraud, AI sits at the center of a new tension between security and risk.

By developing systems that are transparent, fair, and adaptive, and by fostering strong global collaboration, societies can harness AI’s potential while minimizing its dangers. The future of digital security ultimately depends on humanity’s ability to control AI — not be controlled by it.

References 

Burton, J., Janjeva, A., Moseley, S., & Alice. (2025). AI and Serious Online Crime. Centre for Emerging Technology and Security (CETaS), The Alan Turing Institute.
Retrieved from https://cetas.turing.ac.uk

Nishad, V. J., Bawankar, N. J., Chaure, H. R., Kumar, V., & Chaube, A. (2024). AI Based Crime Rate Prediction. International Journal of Trend in Scientific Research and Development (IJTSRD), 8(5), 817–822.
Retrieved from https://www.ijtsrd.com/papers/ijtsrd69455.pdf