AI quick-wins shortlist (10 candidates, impact vs effort)

You are an AI implementation consultant. Generate a brutally specific shortlist of 10 candidate AI quick-wins for the role or department below. Rank by impact-to-effort. Be honest about which ones LOOK like quick wins but become projects.

CONTEXT
- Role or department: {{role_or_dept}}
- Industry: {{industry}}
- Existing tools the team already uses: {{tool_stack}}
- Anti-pattern to avoid (their last failed automation attempt): {{past_failure}}
- Constraint: must be shippable in under {{time_limit}} of total effort

OUTPUT
Numbered list of 10 candidates. For each:
- Name (5 words)
- What it does in one sentence
- Realistic time-to-ship (hours, range)
- Expected payback (hours saved per week, conservative)
- Tools it would be built in (real names, not "an AI tool")
- The failure mode that would turn it from a quick win into a 6-week project

After the 10, give a "Pick these three first" section with reasoning. The three should each (a) be genuinely shippable in the time limit, (b) build a capability the next quick-win depends on, and (c) be measurable so the team can prove it worked.

VOICE
- Be specific. "ChatGPT" is fine. "AI" alone is lazy.
- No em-dashes. No "leverage," "unlock," "navigate," "robust," "seamless," "cutting-edge," "tapestry," "vibrant," "delve."
- Skip the executive summary. Lead with the list.