Back to the MIT repository
4. Malicious Actors & Misuse2 - Post-deployment

Impact on education: bypassing learning

Easy access to high-quality generative models might result in students that use AI models to bypass the learning process.

Source: MIT AI Risk Repositorymit1332

ENTITY

1 - Human

INTENT

1 - Intentional

TIMING

2 - Post-deployment

Risk ID

mit1332

Domain lineage

4. Malicious Actors & Misuse

223 mapped risks

4.3 > Fraud, scams, and targeted manipulation

Mitigation strategy

1. Redesign Curricula and Assessment Implement a systemic pedagogical shift towards authentic, process-oriented, and high-stakes in-class assessments that intrinsically require human critical thinking, application of unique course-specific knowledge, and individual judgment, thereby rendering unedited, AI-generated content insufficient for successful completion. 2. Enforce Comprehensive AI Literacy and Ethical Use Mandates Establish mandatory, institution-wide programs for both students and educators focusing on AI's capabilities, limitations, and ethical considerations, ensuring clear, published guidelines that delineate acceptable use (e.g., brainstorming, drafting with required disclosure) from academic misconduct. 3. Institutionalize Robust Policy and Governance Frameworks Co-develop and consistently enforce transparent academic integrity policies that mandate the disclosure and proper citation of any AI assistance, and establish a process for the fair and consistent application of penalties for non-compliant use, supporting these with continuous professional development for faculty.