MIT recently released the final report of its Ad Hoc Committee on AI Use in Teaching, Learning, and Research Training. The committee was originally asked to assess how students and instructors were using AI, identify innovations in teaching and assessment, and propose an AI use policy. Instead, the committee concluded that generative AI raises much larger questions about what students should learn, how universities should assess that learning, and what higher education should accomplish in the first place.
For technical communicators, the report is particularly interesting because many of its recommendations extend well beyond the classroom. They concern how people decide when to use AI, how they evaluate its output, how they communicate expectations about AI use, and how human expertise should interact with increasingly capable technologies.
In other words, many of the problems MIT identifies are also technical communication problems.
Why Should We Care About the MIT Report on AI and Education?
AI is already deeply embedded in higher education.
MIT’s own surveys found widespread use among students, faculty, and staff. According to reporting on the committee’s survey data, 46% of MIT undergraduates reported using large language models daily, while another 30% used them several times per week. At the same time, only 25% of undergraduates believed MIT was adequately preparing them to use AI professionally.
This creates an interesting problem.
Students are already using AI extensively, but simply using AI does not necessarily mean that they understand how to use it effectively, appropriately, or strategically.
The committee therefore argues that higher education needs to become “AI-aware.” Rather than simply adding AI tools to existing courses or banning them from assignments, universities need to reconsider what students should know and be able to do in a world in which AI can already complete many traditional academic tasks. MIT’s leadership summarized this recommendation as revisiting learning goals, assessments, policies, and educational practices both inside and outside the classroom.
That distinction matters.
What Does AI-Aware Education Look Like?
One of the strongest aspects of the report is that it does not offer a simple answer to AI in education.
Instead, the committee identifies eight principles for approaching the problem:
- Be humble
- Be bold
- Put humanity front and center
- Lean into learning
- Teach with intentionality
- Recognize that no one size fits all
- Favor augmentation rather than automation
- Think beyond the classroom and campus
These principles reject two approaches that have dominated much of the discussion about AI in higher education.
The first is that universities should simply prohibit AI because students can misuse it.
The second is that universities should integrate AI everywhere because students will encounter it professionally.
Neither approach asks the more important question: What are students supposed to be learning?
MIT instead recommends starting with learning goals and determining where AI supports those goals and where it interferes with them. Different courses, assignments, and disciplines may therefore require very different approaches.
MIT has already begun translating this idea into practice. Rather than establishing one institution-wide rule for classroom AI use, it recommends that every course establish a clear policy explaining whether AI is permitted, restricted, required, or prohibited and, importantly, explain the reasoning behind that policy.
This is also a communication problem. Students need to understand not only what they are allowed to do, but why those boundaries exist.
Why Does the MIT Report Favor Augmentation Over Automation?
One of the report’s most useful principles is “augmentation not automation.”
AI can make many intellectual tasks faster. That does not necessarily mean that making those tasks faster improves learning.
The report is particularly concerned about students delegating intellectual work to AI before developing the knowledge needed to evaluate what the technology produces. It describes one version of this problem as “cognitive surrender,” in which producing a satisfactory answer begins to replace the process of developing understanding.
This problem is especially relevant to technical communication.
Technical communicators can already use AI to draft documentation, summarize research, generate content, analyze information, brainstorm designs, and perform many other tasks. These capabilities will almost certainly continue to improve.
The important professional skill is therefore becoming less about whether someone can generate an acceptable piece of content and more about whether they can determine what should be created, evaluate what AI produces, recognize problems, understand audiences, make strategic decisions, and revise outputs accordingly.
AI can automate production.
It cannot make professional judgment unnecessary.
What Does the MIT Report Mean for Technical Communication Education?
Technical communication programs have an opportunity here.
Our field has always existed at the intersection of people, information, and technology. Technical communicators regularly need to learn unfamiliar technologies, evaluate information, understand audiences, collaborate with specialists, and translate complex material into forms that people can actually use.
Generative AI makes those skills more important, not less important.
Teaching students how to write a document without AI is no longer enough. Teaching them how to prompt an AI system is not enough either.
Students need to understand the larger communication process surrounding these technologies.
When should AI be used? What information should be provided to it? How should its output be evaluated? When does automation save time without sacrificing quality? When does automation remove the very intellectual work someone needs to perform? How should organizations communicate AI policies? How do we design human-in-the-loop processes in which people retain meaningful oversight?
These are technical communication questions.
The report also reinforces the importance of experiential and project-based learning. MIT recommends reconsidering assessments and emphasizing educational experiences that make student learning more visible, including hands-on, social, and experiential activities.
That approach is particularly compatible with technical communication because our field already relies heavily on projects, research, collaboration, design, and the production of deliverables for real or realistic audiences.
What Should Technical Communicators Take Away from the MIT Report?
The most useful message from the MIT report is not that AI is good or bad for education.
It is that the question is too simple.
AI is now sufficiently capable and sufficiently widespread that educators need to reconsider what they are actually trying to teach. Professionals need to make the same assessment about their work.
The goal should not be to preserve every task people performed before generative AI. Nor should it be to automate everything that AI can perform.
Instead, we need to determine which human capabilities remain important, how AI can extend those capabilities, and where relying on AI undermines the expertise people actually need.
Technical communicators are well positioned to participate in that work because we already study the relationships among people, technologies, information, organizations, and communication practices.
The tools are changing.
The need to understand how people should use them is not.
Read the Full MIT Report
The complete Report of MIT’s Ad Hoc Committee on AI Use in Teaching, Learning, and Research Training is available through MIT:
Read the full MIT report on AI and education
Want to Hear about Mercer’s M.S. in Technical Communication Management?
Mercer’s M.S. in Technical Communication Management prepares working professionals to develop advanced skills in technical communication, content strategy, usability and user experience, instructional design, project management, and emerging technologies such as AI.