AI for Design & Visual Communication
Moderate · Communication & Dissemination track · ~35 min hands-on + readings and quiz
← Back to the Communication & Dissemination track
Make clear, accurate communication graphics with AI design tools — and pick the right tool for the data.
What you’ll be able to do
- Generate communication graphics (infographics, social posts, slides) with AI design tools
- Choose the right tool — Claude, Canva, Figma, or Adobe Express/Firefly — for the job and the data level
- Apply accuracy, accessibility, representation, and disclosure standards to AI-made visuals
Overview
Several AI tools now produce usable visuals fast. Claude can generate interactive infographics and web pages (“artifacts”); Canva and Figma have AI design features; Adobe Express with Firefly makes commercially-safe images and templates. You bring the message, the accuracy, and the design judgment. This is the self-paced companion to Workshop 4, which focuses on the Adobe tools.
The data rules differ by tool. Adobe Express, Firefly, and Acrobat are Drexel-approved for Moderate Risk data; Canva, Figma, and Claude are consumer tools — Low Risk Data only unless Drexel approves them. Use public, non-sensitive content, disclose AI-generated visuals, and check representation and accessibility before anything is published.
Practice activities
Activity 1 · Moderate — Build a one-page infographic
Time ~20 min · Tools Claude (or Canva)
Goal. Turn a finding into a clean, accessible visual you can share.
Setup. One public finding — e.g., “In Philadelphia, pedestrian-involved crashes peak between 4–7 PM and are more common on wide arterial roads.” Use public, non-sensitive content only.
Steps.
Describe the infographic (in Claude, ask for an artifact):
Create a clean, accessible one-page infographic (HTML) for this public-health finding: [finding]. Include a headline, one simple chart, and three takeaways. Use high-contrast colors and large text. Public data only.
(In Canva, use Magic Design / Magic Media with a similar prompt.)
Review for accuracy and accessibility — correct chart type, no fake precision, strong contrast, a caption or alt text.
Iterate one change at a time (simplify the chart, fix a label, adjust contrast).
Expected result. A shareable infographic draft.
Check your work. Is every number accurate, the text high-contrast, and a caption/alt text provided?
Common pitfalls. AI visuals can misrepresent data (wrong chart, invented precision) — verify against the source. Consumer design tools are Low Risk data only; use public content.
Stretch (optional). Export it and confirm it meets accessibility basics (color contrast, alt text) for publication.
Activity 2 · Moderate — Match the tool to the job and the data
Time ~15 min · Tools Claude + Canva or Figma + Adobe Express
Goal. Pick the right design tool for the task and the data sensitivity.
Setup. Use this concrete brief:
Design brief: Create a one-page flyer for a community pedestrian-safety event at Fairhill Square Park (4th & Lehigh, North Philadelphia) on Saturday, October 18, 2026, 10 AM–1 PM. The event is organized by DSPH and the Office of Transportation & Infrastructure Systems. It includes free bike helmet fittings for children, a crosswalk-safety demonstration, and information tables about Vision Zero. The flyer should be in English and Spanish, include a QR code placeholder for RSVP, and be suitable for posting at community centers, schools, and bus shelters. Audience: neighborhood residents, particularly families with children. Data sensitivity: all content is public.
Also write alt-text for the flyer’s main image. Here’s an example of good alt-text for a pedestrian-safety context:
Sample alt-text: “A parent holding a child’s hand crosses a marked crosswalk at an urban intersection. A pedestrian countdown signal shows 12 seconds remaining. Cars are stopped at the red light. The scene is set on a sunny day in a residential neighborhood with brick row houses.”
Steps.
Make a quick draft of the flyer in two tools (e.g., Adobe Express and Claude/Canva).
Compare: speed, control over layout, output quality, and the data classification each tool allows.
Get a recommendation:
This is public (Low Risk) content for community distribution. Which tool fits best for producing a bilingual pedestrian-safety flyer: Adobe Express, Canva, Figma, or Claude? List the Drexel data-rule consideration for each.
Write alt-text for the main image in your flyer, following the sample above — describe what’s shown, not what it means.
Expected result. A reasoned tool choice with the data-rule rationale, plus alt-text for your flyer image. Since the content is public/Low Risk, all four tools are acceptable here — the choice is about capability and workflow, not data restrictions.
Check your work. Did your alt-text describe the image content (who, what, where) rather than interpreting it (“a beautiful scene of safety”)? Could a screen-reader user understand the image from your description?
Common pitfalls. The flashiest tool isn’t always approved for your data — this case is Low Risk, but if the flyer included photos of identifiable community members, the classification would change. Always disclose AI-generated visuals.
Stretch (optional). Decide which tool you’d standardize on for routine team graphics, and why. Consider accessibility features (alt-text support, color-contrast checking) as a factor.
Check your readiness
Answer these, then check — your score suggests whether to dive in or skim the readings first.
Recommended readings
Available in the shared OneDrive folder Staff Faculty AI Workshop → Readings, and online where linked:
- Anthropic Has Caught Up to OpenAI in Image Understanding — what multimodal tools can do with images.
- These New Tools Could Make AI Vision Systems Less Biased (MIT Tech Review) — representation and bias in AI-generated visuals.
- Claude Code Now Supports Artifacts — generating interactive, shareable visual outputs.
Useful resources
- Adobe Firefly — commercially-safe image generation (Drexel Moderate-Risk data).
- Canva — AI design (Magic Studio); Low Risk data only unless approved.
- Figma — collaborative design with AI features; Low Risk data only.
- Claude — generate infographics and artifacts; Low Risk data only at Drexel.
- Drexel AI Tools — approved tools by data classification.