The rapid ascent of generative AI has triggered an existential debate within UK higher education, placing traditional assessment methods under unprecedented scrutiny. Recent reports from the Higher Education Policy Institute (HEPI) and the Quality Assurance Agency for Higher Education (QAA) suggest that the sector is at a crossroads: either react with defensive, short-term fixes or seize this as a generational opportunity for meaningful assessment reform. For university leaders, particularly in disciplines like law, the path chosen will define the credibility and relevance of their programmes for years to come.
For too long, the conversation has been dominated by a narrative of risk and compliance. The arrival of tools like ChatGPT has led many institutions to double down on detection software and punitive policies. However, as a recent HEPI article argues, this approach is not only unsustainable but misses the point entirely. Generative AI is not the root problem; it is a powerful lens that exposes the long-standing limitations of assessment practices designed for a pre-digital age.
A Catalyst for Change, Not a Crisis to Manage
The core message from both HEPI and the QAA is a call to move beyond a compliance-driven mindset. In a February 2026 article, Dr Emma Ransome of Birmingham City University warns that a focus on policing AI use places an unsustainable burden on academic staff and fosters a culture of mistrust. It encourages students to focus on risk avoidance rather than deep learning, fundamentally undermining the educational mission. [1]
This sentiment is echoed in the extensive guidance published by the QAA. The agency has been clear that assessment practices developed under traditional paradigms are no longer fit for purpose. Their publications, including 'Reconsidering assessment for the ChatGPT era' and the February 2024 'Quality Compass' briefing, urge a move towards principled redesign. The QAA advocates for sustainable strategies that prioritise academic integrity and learning over the futile exercise of AI detection. [2] The consensus is clear: we cannot put the AI genie back in the bottle, so we must change the bottle.
The Shift to Authentic Assessment
The vulnerability of the traditional essay and the unseen exam is not just that AI can write them, but that they often fail to assess the most valuable graduate skills: critical analysis, complex problem-solving, and ethical judgment. The challenge for universities is to design assessments that are not only 'AI-proof' but are also more effective measures of these essential capabilities.
This is where authentic assessment comes in. This approach involves designing tasks that mirror the real-world challenges and contexts that graduates will face in their professional lives. Instead of asking a law student to simply write an essay on contract law, an authentic assessment might involve:
- Portfolio-Based Evaluation: Requiring students to compile a portfolio of work over a semester, including draft legal opinions, client letters, and reflective logs on their learning process.
- Oral Examinations (Vivas): Engaging students in a professional dialogue to defend a research project or justify their reasoning on a complex legal problem, a format that demands genuine understanding.
- Problem-Based Scenarios: Presenting a complex, multi-faceted legal scenario and asking students to produce a set of integrated deliverables, such as a client advice note, a draft pleading, and a risk analysis.
These methods do not just mitigate the misuse of AI; they actively cultivate the skills that are becoming more valuable in an AI-enabled workplace. They shift the focus from knowledge recall to knowledge application, from demonstrating what you know to showing what you can do with what you know.
A Framework for Trust-Based Reform
Transitioning to a model of authentic assessment is not a simple switch. It requires a deliberate, institution-wide commitment to a new philosophy of assessment built on trust and pedagogical purpose. This involves three core pillars:
1. Assessment as a Learning Tool
First, assessment must be reframed as an integral part of the learning process, not merely a final judgment. Formative assessment, which provides feedback during the learning journey, becomes critical. By creating opportunities for students to practice, receive feedback, and refine their work, institutions can build the skills and confidence needed for more complex, summative tasks. Platforms like LexIQ, for instance, can provide students with instant, AI-driven feedback on draft essays or problem questions, allowing them to iterate and improve while freeing academic staff to focus on designing the high-stakes, authentic assessments that truly matter.
2. Redefining Academic Integrity
A trust-based model requires a new social contract with students. Instead of assuming misconduct, institutions should proactively educate students on the ethical and effective use of AI tools. This means developing clear, co-created policies that distinguish between illegitimate use (e.g., submitting an AI-generated essay as one's own) and legitimate use (e.g., using AI for brainstorming, summarising complex texts, or checking grammar). The goal is to cultivate critical AI literacy, empowering students to use these powerful tools responsibly.
3. Empowering and Supporting Educators
Finally, and most critically, university leaders must recognise that assessment redesign is a significant and skilled undertaking. Academics cannot be expected to overhaul their modules overnight without proper support. Institutions must invest in professional development, provide dedicated time for curriculum redesign, and foster a culture where innovation in teaching and assessment is recognised and rewarded.
The Future of Legal Education
The emergence of generative AI is a watershed moment for higher education. For the legal sector, where the ability to think critically, argue persuasively, and act ethically is paramount, the stakes could not be higher. The reports from HEPI and the QAA provide a clear roadmap. The choice is between clinging to the familiar comforts of an outdated assessment model or embracing this moment as a catalyst to build a more resilient, relevant, and trustworthy approach to legal education. By focusing on authentic assessment and a culture of trust, universities can ensure they are preparing graduates not just for their final exams, but for a lifetime of professional success in a changing world.
References
[1] Ransome, E. (2026, February 6). What generative AI reveals about assessment reform in higher education. HEPI. https://www.hepi.ac.uk/2026/02/06/what-generative-ai-reveals-about-assessment-reform-in-higher-education/
[2] QAA. (n.d.). Generative artificial intelligence. Quality Assurance Agency for Higher Education. https://www.qaa.ac.uk/sector-resources/generative-artificial-intelligence