The debate surrounding the ethical dilemmas associated with education agents has persisted for decades. At the same time, the use of agents has grown rapidly, suggesting that many higher education institutions and prospective international students continue to perceive value in their services. This article examines the key drivers behind agent engagement, explores the ethical concerns it raises, and considers how generative artificial intelligence may reshape the future of international student recruitment.

The education agent industry has a history spanning over 30 years. In the past decade, the contracting of agents by higher education institutions has become a normalized international student recruitment practice in many countries across the world. However, agent-based international student recruitment remains controversial in parts of the sector.

Education agents are for-profit actors assisting education providers to recruit fee-paying international students across the globe. Many of these relationships are formalized in a written contract. A typical compensation model involves education providers paying a commission for each student recruited (for instance, 10-20 percent of the first year’s tuition fee or a flat fee), but other financial arrangements also exist. In some countries, such as China, it is common for agents to receive payments from higher education institutions, while also charging students a service fee. The international student recruitment sector also includes other actors, such as independent consultants, who are fully compensated by prospective students.

Ethical dilemmas associated with agent use, such as limited transparency, financial and document fraud, and misinformation, are frequently discussed by academics, policy makers, students, and industry professionals. The key problem with this recruitment model is the principal-agent relationship between higher education institutions and agents, which incentivizes agents to direct students to their partner institutions in exchange for financial compensation. This raises concerns about the extent to which the guidance provided by agents serves prospective students’ interests. Indeed, the terms underlying this arrangement often are not clearly disclosed to prospective students.

The awareness of this ethical dilemma suggests that the continued popularity of agent use is driven by the belief that the benefits outweigh the perceived risks. This article explores the drivers for agent use and how artificial intelligence may disrupt them.

Why Do Education Agents Endure?

The book Student Recruitment Agents in International Higher Education outlines international trends and challenges in agent use and discusses a number of drivers that incentivize agent engagement. Drivers can be divided into the rationales that higher education institutions have for contracting agents and the rationales prospective international students have for using intermediaries, whether university-contracted agents or independent, fully student-contracted counselors. 

On the institutional side, the drivers for agent engagement often originate from the financial objectives that higher education institutions aim to meet by recruiting international fee-paying students. For higher education institutions, agents are often considered an easy market entry strategy to gain access to prospective students across borders without upfront investment. Agents have local knowledge and networks and help institutions maintain a physical presence in countries across the globe. In some key markets, prospective students often engage intermediaries, so deciding not to work with agents may make international student recruitment challenging. 

On the other hand, students often report that they are satisfied with the services that agents offer. Students can benefit from agents’ local and international knowledge when navigating the eye-watering range of study opportunities across the globe and the complexities of application and admission processes. Students and their families may prefer to communicate with someone from the same culture and in their own language. Some institutions may also either require or strongly recommend that prospective students from specific countries work with one of their contracted agents. 

If these factors drive agent engagement, it is important to explore how artificial intelligence may impact these drivers and the ethical dilemmas associated with agents.

Generative Artificial Intelligence as a Disruptor?

The emergence of generative artificial intelligence (GenAI) has the potential to disrupt the education agent sector. Large language models, such as ChatGPT, Copilot, Gemini, and others, are examples of GenAI technology that enables prospective students to have easy access to information about study destinations and institutions.

There is evidence that international students are well aware of these tools and are employing them. For instance, in a 2023 survey of more than 10,000 international students, 45 percent of the respondents either intended to use or had used GenAI tools when selecting their institution. For China, the percentage was significantly higher at 75 percent. With the rapid development and popularity of large language models in the past two years, it can be assumed that this percentage is now much higher.

Prospective students’ ability to use large language models to explore suitable study options erodes the value of the guidance that has been traditionally provided by agents. As we move toward a future in which large language models can provide information about study options across the globe, as well as assist prospective students in their application and visa processes, the drivers for agent use will likely erode, unless agents can demonstrate the unique value of humans and their specialized skills in this process.

It is important to note that prospective students’ reliance on GenAI platforms will not eradicate ethical issues. Content generated by large language models is known to include hallucinations, biases, and misinformation. In the future, we are also likely to see more stakeholder attempts to influence the content generated by public large language models, which will create new types of conflicts of interest. This means that some of the current ethical issues associated with the education agent industry may be reproduced by GenAI tools.

The Future of AI-Agent Engagement

Education agents have sustained their core business model for over three decades by providing added value to prospective students and higher education institutions. The education agent recruitment model has endured, despite its susceptibility to ethical dilemmas. However, the adoption of generative AI might finally erode the drivers for agent engagement.

Regardless of the long-term prospects for the education agent industry, it is clear that generative AI will further complicate what is already a challenging situation. Government agencies and higher education institutions must start paying increasing attention to guidance provided by large language models as it relates to study destinations and specific educational institutions. They will also need to consider strategies that can help mitigate issues associated with AI-agent biases, hallucinations, misinformation, and conflicts of interest.

Pii-Tuulia Nikula is associate professor at the Eastern Institute of Technology, New Zealand. E-mail [email protected].