Balancing Personalisation and Protection: Ethical and Legal Safeguards for AI-Driven Recommender Systems

The rapid growth of online content and global internet use has intensified the problem of information overload, prompting widespread adoption of AI driven recommender systems to deliver personalised content. While these systems enhance user experience by reducing search time and improving content relevance, they also raise significant ethical and legal concerns. This paper critically examines the risks posed by general purpose AI recommender systems, particularly those deployed on social media platforms, including privacy violations, algorithmic bias and behavioural manipulation. It further explores associated legal challenges such as data protection compliance, transparency obligations and intellectual property rights. Drawing on both technological analysis and policy perspectives, the study proposes practical measures for mitigating these risks, including regulatory reforms, enhanced transparency and systematic auditing. By addressing the dual imperative of fostering innovation and safeguarding users, this work offers a framework for policymakers, industry stakeholders and researchers to ensure that AI powered recommendations serve the public interest without undermining fundamental rights.