Skip to main content
AccessForce
Text size: 100 percent

Hackathon evidence

Type C: Accessibility Audit + AI Redesign

Everything a judge needs, on one page: what was wrong, what the AI found and recommended, and what replaced it.

Real public-sector experiences

Real experience → AI audit → AI redesign → After

  1. Step 1

    Real experience

    Existing Utah and SSA public-sector applications

  2. Step 2

    AI audit

    AccessForce identifies potential accessibility and usability barriers

  3. Step 3

    AI redesign

    AI determines how the experience could become more accessible

  4. Step 4

    After

    Agentforce delivers an adaptive conversational experience

Concrete evidence of the redesign

  • Document navigationConversational navigation
  • Program branchingAI-managed branching
  • Government terminologyPlain language
  • Structured fieldsNatural conversation
  • Help discoveryProactive assistance
  • Static interfaceAdaptive experience
  • Exact-answer pressureApproximate / follow-up workflow when appropriate
Utah Benefits case studySocial Security case study

Type C evidence — measured, not asserted

  1. Step 1

    Real application

    Published federal and state forms people use today.

  2. Step 2

    AI audit

    Seven accessibility dimensions examined and explained.

  3. Step 3

    AXS score

    One transparent 0–100 score for the current experience.

  4. Step 4

    AI redesign

    An adaptive Agentforce conversation, human-approved.

  5. Step 5

    AXS re-score

    The same dimensions scored again, point by point.

Prototype accessibility audits

  1. 1Housing Choice Voucher WaitlistLocal housing authority pre-applicationOpportunity: HighAXS 38 → 93Major Barriers Identified
  2. 2Help With a Utility BillLocal energy assistance applicationOpportunity: HighAXS 38 → 93Major Barriers Identified
  3. 3U.S. CitizenshipUSCIS Form N-400 naturalization applicationOpportunity: HighAXS 39 → 92Major Barriers Identified
  4. 4Utah Benefits31-page multi-program benefits applicationOpportunity: MediumAXS 42 → 93Significant Opportunities
  5. 5Filing a Federal Tax ReturnIRS Form 1040 and the Free File optionsOpportunity: MediumAXS 42 → 93Significant Opportunities
  6. 6California CalFresh Food AssistanceCalFresh application, administered by countiesOpportunity: MediumAXS 44 → 94Significant Opportunities
  7. 7New York Unemployment InsuranceFirst claim and weekly certificationOpportunity: MediumAXS 44 → 93Significant Opportunities
  8. 8VA Disability CompensationVA Form 21-526EZ federal disability claimOpportunity: MediumAXS 47 → 94Significant Opportunities

Illustrative prototype scores produced by AccessForce. Lower scores indicate larger improvement opportunities, not organisational failure.

Before and after isn't just visible.

It's measurable.

Before

The existing inaccessible experience

  • Form HS-1042-A — 17 pages, 214 fields
  • Grade 13.2 reading level, English only
  • 37 unlabelled fields, broken reading order
  • 15-minute session timeout, no save
  • Accommodation guidance on page 14

AI audit + redesign

Barriers identified, recommendations generated

  • Accessibility score: 42/100
  • 18 barriers across six categories
  • Each barrier scored, explained, and matched to affected people
  • Plain-language rewrite generated question by question
  • Workflow regenerated as an adaptive conversation

After

The redesigned Agentforce experience

  • 8 questions, one at a time, visible progress
  • Grade 6–8 plain language
  • Read aloud, keyboard-only, larger text, high contrast
  • Accommodations and language offered first
  • Save and resume, and a person on request

Concrete transformations

  • 17-page PDFGuided conversation214 fields reduced to 8 questions that matter
  • Grade 13 languageGrade 6–8 plain languageReading level rewritten and re-scored
  • All questions at onceOne question at a timeProgressive disclosure with 'Step 3 of 8'
  • Hidden accommodationsProactively offered assistanceAccommodation offered in the first message
  • English onlyMultilingualLanguage choice before the first question
  • Visual-first navigationMultimodal accessible interactionRead aloud, keyboard, voice, larger text
  • Generic workflowAdaptive experienceFlow changes with the support each person picks

Audit trail: the critical findings

  • Criticalseverity

    Reading level: Grade 13.2

    Rewrite every question in short, direct sentences and explain each term in context.

  • Criticalseverity

    37 form fields have no programmatic label

    Ask each question as labelled text in a conversation, with one clear input at a time.

  • Criticalseverity

    Accommodation request is buried on page 14

    Offer accommodation options in the first message, before any question is asked.

Judging criteria, and where to find the evidence

  • Accessibility

    Weight 20%

    WCAG-AA tokens, keyboard paths, visible focus, 44px targets, text scaling, high contrast, reduced motion, light and dark — on every page, adjustable from the header.

    Try the toolbar, then the experience

  • Responsible AI

    Weight 20%

    AI proposes and people decide, a human route at every step, no eligibility decisions by AI, data minimisation, bias review, and stated limits.

    Responsible AI

  • Clarity & communication

    Weight 15%

    Plain-language copy throughout, a numbered demo path, and original wording shown beside the rewrite.

    Before & after

  • AI fluency skills

    Weight 25%

    Understand, identify, reason, redesign, personalize, improve — each named at the step where it happens.

    AI redesign

  • Impact & insight

    Weight 25%

    Five people, the barrier that blocks each of them, and projections with the assumption behind every number.

    Who's excluded

  • Creativity & originality

    Weight 15%

    An audit that outputs a working experience rather than a report, and a site that adapts as it argues.

    This page

  • Scalability & adoption

    Weight 10%

    An accessibility layer over existing systems: pilot one form, expand across a portfolio, make audit standard. No legacy replacement.

    Architecture

What makes this original

  • Most accessibility tools produce a report. AccessForce produces a working replacement experience.
  • The audit is scored by human impact, not by rule count, so the worst barrier is the one that blocks a person completely.
  • Accessibility becomes a conversation setting a person chooses, not a separate 'accessible version' someone else maintains.
  • This site is the demo: the same adaptations it recommends are live in the toolbar you're using now.

What makes this adoptable

AccessForce is a layer, not a replacement. The legacy system keeps receiving the same fields, so a team can start with one form on Monday and expand across a portfolio without a migration project.

  • Pilot — one form, audited, redesigned, human-approved.
  • Expand — reuse the layer across forms on the same systems.
  • Standard — audit becomes part of publishing anything new.
Open the redesigned experienceResponsible AI commitmentsWhat's next

The AccessForce journey

Discover → Audit → Redesign → Translate → Complete → Improve

Type C asks for an accessibility audit and an AI redesign. AccessForce starts one step earlier, with finding the right process at all — and continues past submission, into the language someone speaks and the analytics that show what to fix next.

  1. Step 1

    Discover

    Tell AccessForce what you're trying to accomplish.

  2. Step 2

    Find

    AI identifies the likely government service and official form.

  3. Step 3

    Audit

    AXS evaluates accessibility and process barriers.

  4. Step 4

    Redesign

    Agentforce transforms the experience.

  5. Step 5

    Translate

    The experience is offered in the person's own language, not just translated word for word.

  6. Step 6

    Complete

    The person works through an adaptive conversation.

  7. Step 7

    Improve

    Aggregate friction signals show agencies what to fix next.

Accessibility starts before the first question.

Open the Form Finder