15 Equity, access, and GenAI
Equity in GenAI can mean several things. It includes fair access to GenAI tools, fairness in the information GenAI learns from, and fair chances for everyone to benefit from GenAI.
True GenAI equity requires that every student has equal access not just to the software, but to the hardware, instruction, and ethical protections necessary to use it safely and effectively.
Technology considerations
Cost per use and subscriptions
GenAI tools can be expensive, not everyone can easily access or use them. Some tools also offer “premium” services through a fee-based subscription. Premium GenAI models can offer higher accuracy, and advanced data analysis, but they cost recurring monthly subscription fees. When professors design coursework around specific tools, or permit the use of AI, students who can afford monthly subscriptions can gain an advantage over those relying on University-licensed or free commercial models.
The Hardware Gap
Advanced GenAI tools and interfaces require updated laptops with modern processors and sufficient RAM. Students using older, low-spec laptops or relying solely on smartphones are left at a disadvantage.
Bandwidth & Network Limits
Running AI-driven web applications or generating outputs requires high-speed, stable internet. Under-resourced students living off-campus or in rural areas with poor connectivity face friction that well-connected peers do not experience.
Example: A student on a limited mobile data plan may be unable to use AI tutoring platforms or image-generation tools that require significant bandwidth, while a peer on campus gigabit Wi-Fi uses them seamlessly.
Gaps in AI literacy training and course policies
Inconsistent Policies & Faculty Preparedness
Policies around student GenAI use vary drastically between departments and professors. Some instructors ban GenAI completely, while others require it, leaving students to navigate a confusing landscape of rules and expectations.
Informal Learning vs. Formal Guidance:
Students with established social or professional networks often learn prompt engineering and GenAI strategy through peers. First-generation or underrepresented students may lack access to these informal networks and suffer without explicit in-class instruction or support.
Disparities in Prompt Literacy:
Knowing how to prompt effectively requires an understanding of academic expectations. Students unfamiliar with academic terminology are less equipped to guide the GenAI toward high-quality, relevant responses. If learners rely too much on GenAI in their academic work, it could make achievement gaps worse, and change the way their brain operates. This is especially true if students from schools with fewer resources don’t get good guidance on how to use these tools properly.
Penalties for GenAI use
Disproportionate False Positives for Non-Native English Speakers:
AI detection software routinely flags writing by ESL students as “AI-generated” because their structured grammar and vocabulary closely match the statistical patterns AI detectors target.
Unfair Penalty Risks:
Neurodivergent students or those using assistive writing software (like basic grammar checkers) are also at higher risk of false accusations of academic dishonesty.
- Example: A multilingual student wrote an essay entirely on their own but received a failed grade because an unreliable AI detector falsely flagged their academic style as machine-generated.
Accessibility concerns
Screen Reader & User Interface Barriers:
Many GenAI chat interfaces and visual outputs lack proper alt-text formatting, keyboard navigation, or compatibility with screen readers, creating barriers for blind or low-vision students.
Over-Reliance on Visual outputs:
As GenAI shifts toward visual and multimodal outputs, these platforms often neglect accessible, text-based alternatives for users.
- Example: An AI tool generates an complex infographic summary of a lecture, but fails to provide a structured text equivalent, locking out students using screen-reader technology.
