
The Office for Students, England’s higher education regulator, has concluded that artificial intelligence poses significant threats to the integrity of university degree assessments and has called for institutions to adopt traditional examination methods. The regulator noted that the widespread use of AI in essays, dissertations, and open-book exams has created challenges for academic staff attempting to evaluate student work fairly and accurately.
According to the investigation, which involved interviews with more than 300 students and staff members at 127 universities and colleges, AI use in higher education is prevalent and often undetected. Students reported that their peers frequently acknowledged using AI tools to complete assignments within minutes while receiving comparable grades to those who completed work independently. Some academics indicated receiving multiple submissions that appeared nearly identical in content, style, and references, raising questions about the validity of existing assessment methods.
The OfS warned that institutions failing to implement valid and reliable assessment measures could face sanctions or breach their registration conditions. Officials emphasized that many universities have already begun redesigning assessment practices, including increased reliance on in-person written exams, oral examinations known as vivas, and process-based assessments where students document their research and work methods. Some institutions have also introduced assessments requiring students to critique or edit AI-generated content in controlled settings.
Regulatory officials characterized the situation as presenting both generational risk and opportunity. Josh Fleming, director of strategy and delivery at the OfS, stated that the sector must make greater efforts to combat AI-related threats while still capturing potential benefits. The Department for Education indicated expectations that universities take the matter seriously and that regulators enforce standards effectively and develop robust frameworks governing AI use in educational settings.
Universities UK, representing 147 higher education institutions, acknowledged ongoing work to address assessment challenges through modified evaluation approaches, though acknowledged further progress remains necessary.
Article Attribution | Read More at Article Source
Article summary produced by Claude AI