How ibl.ai Helps Build AI Literacy

AI is now part of everyday academic work. The real risk isn’t “AI in the classroom”; it’s unstructured AI use that muddies assessment, frustrates students, and erodes trust. At ibl.ai, we help campuses turn AI into a transparent, scoped teaching partner—anchored in pedagogy, backed by clear guardrails, and supported with hands-on coaching until it works in real courses. We’ve refined this approach alongside institutions like Syracuse University, Morehouse College, and long-time collaborators at George Washington University. Prof. Lorena Barba at GWU recently published her reflections on a genAI-enabled engineering course echoing two core principles we teach: (1) set expectations early and precisely, and (2) design for process and attribution, not just final products. Those insights inform the playbook below.

---

What Faculty Need From AI Literacy (& Why)

Prof. Lorena A. Barba’s (GWU) recent paper documents a pattern many instructors meet in their first AI-heavy term:

• Illusions of competence. Students can appear fluent because AI scaffolds too much, too soon. Without safeguards, perceived mastery outpaces real understanding.

• Assessment validity. Traditional take-home tasks can become poor measures of learning when AI is widely available; validity, not just “cheating,” becomes the core problem.

• Boundary-setting. Bans are impractical; clarity about acceptable AI use is essential (what’s allowed for exploring vs. drafting vs. final submission).

• Student sentiment. Tensions spike if expectations shift late. Prof. Barba reports this led to unusually negative evaluations and a need to redesign activities and assessments mid-course.

Her takeaways are actionable: guide AI use (e.g., exploration and code explanation), prompt reflection after AI use, and rebuild assessments for validity under realistic time or supervision constraints. These lessons align tightly with the goals of AI literacy and shape how we train and support faculty. Reference: Lorena A. Barba, “Experience Embracing GenAI in an Engineering Computations Course: What Went Wrong and What’s Next” 2024.

What Our AI Literacy Program Looks Like On Campus

Group Track (Cohort Learning + Open Support)

Kickoff: Principles → Practice

We align on responsible use, then demo concrete tasks instructors can try the same day: drafting rubrics, creating feedback banks, and configuring course-aware assistants that cite the assigned materials.

Pedagogy Lab

Bring an assignment; leave with an AI-robust version. Together we:

• map learning outcomes to allow / constrain / disallow AI use,

• add “show-your-work” and reflection steps to surface reasoning,

• shift trivialized take-home tasks into quick in-class checks.

Assessment Studio

We co-create structured comment banks and feedback workflows where AI proposes suggestions and instructors review and select—speeding turnaround without outsourcing judgment.

Policy & Student-Communication Clinic

Faculty leave with course-level AI guidelines (what’s allowed, where, how to cite; consequences and appeals), plus syllabus/LMS language ready to paste.

Weekly Drop-In Office Hours.

Open sessions for questions, quick builds, and troubleshooting as instructors iterate from early usage analytics.

One-On-One Faculty Coaching

1) Course & Outcomes. Pick 1–2 pilot activities; script how you’ll introduce AI expectations to students. 2) Assistant Build & Retrieval. Configure a course-aware assistant: upload/link readings and slides; scope sources; tune answer style to cite course materials and encourage deeper study. 3) Safety, Scope, Integrity. Set custom boundaries so assistants refuse out-of-scope questions; add productive friction (reasoning checkpoints, versioning). 4) Launch, Observe, Iterate. Define success signals; run a phased rollout; adjust prompts, retrieval sets, and activities from real usage—not guesswork.

What ibl.ai Sets Up For Instructors

• Cited answers by design. Assistants reference the course’s slides/readings so students can verify claims and explore further.

• AI-robust assessments. Redesigned items, rubrics, and in-class checks that remain valid with AI in the mix.

• Clear, consistent policies. Discipline-specific templates that remove ambiguity for students and TAs.

• Prompt & reflection packs. Templates that nudge metacognition and responsible tool use.

• Ongoing human support. Cohort workshops, weekly office hours, and rapid 1:1s to keep momentum.

Why AI Literacy Matters—Right Now

• Assessment validity. General-purpose tools trivialize some take-home tasks; literate faculty redesign for reasoning, evidence, and proper attribution.

• Student trust. Clear, consistent rules reduce workarounds and improve course climate.

• Timely, equitable feedback. Review-and-select workflows deliver faster, targeted comments without losing the human voice.

• Workforce alignment. Graduates learn to scope tasks, check sources, and document AI use—habits they’ll need on day one.

• Institutional governance. Literate departments craft policies that scale good practice without blanket bans.

Where We’ve Learned What Works

Across deployments at Syracuse University, Morehouse College, and GWU, we’ve paired platform rollouts with practical faculty enablement: group workshops, office hours, and one-on-ones that meet instructors where they are. And while every campus context is different, the same pattern holds: clarity up front, process-centric assessment, and assistants that cite the assigned materials lead to better learning and fewer policy headaches.

---

In Conclusion

Our work with universities like Syracuse, Morehouse, and GWU shows that AI literacy grows fastest when pedagogy and tooling move together—clear rules, AI-robust assessments, and assistants that cite what you actually teach. If you’re planning faculty AI training—or want to pilot the kickoff + assessment studio with a department—reach out at ibl.ai/contact. We’ll help your instructors teach more of what only they can teach, with AI as a transparent, well-scoped partner.