HTW BLOG: The State of AI in Hawaii

Treehouse Coworking
Sep 29
4 min read

Educators, healthcare leaders, bankers, technologists, and security professionals compared notes on what’s actually working with AI—where it breaks, and how to get your teams (and yourself) ready to ride the wave.

At 2025 Honolulu Tech Week, “The State of AI in Hawaiʻi” kicked off Hawaiʻi Tech Days at the Walter Dods, Jr. RISE Center in Honolulu, featuring leaders from Bank of Hawaii, Booz Allen, Hawaiʻi Pacific Health, ʻIolani School, and Servco, plus two Hyundai Card interns. The panel showcased real AI use cases and workforce upskilling, offered guidance for students, previewed Fall 2025 Tech Days events, and acknowledged sponsors: UH System, Honolulu Tech Week, CIO Council of Hawaiʻi, PACE, and the Chamber of Commerce Hawaiʻi.

1) The foundation of using the AI Literacy Triangle for adopting machine learning where we practice:

How it works by understanding the primitives: tokens, probabilistic prediction, model limits.
How to use it by applying it to your tasks, not your learning and true concept creation.
How you’ll engage ethically for your personal and professional brand choices: What’s acceptable use? Where do you draw lines in creative work, IP, or disclosure?

“Ten times zero is still zero. AI amplifies what’s there—skills, ethics, and intent.”

2) Healthcare: From Data Entry to Patient Connection

Healthcare leaders described a system long stuck in manual documentation. Ambient AI note-taking tools in clinics are already reducing after-hours charting from 15 hours a week to 1 hour for pilot cohorts—freeing clinicians to talk to patients (and to stay in practice).

Current wins and next steps:

Ambient scribing: Record visits with a medical app, generate structured notes, draft orders and follow-ups.
Operational forecasting: Staffing for flu surges, imaging triage, population-health signals.
Guardrails: “AI moves at the speed of trust.” Tools must be validated on local populations; FDA approval trained elsewhere isn’t enough. Costs matter for nonprofit systems—benefit must outweigh burden.

3) Banking & Enterprise: “GPT-Fluent” Talent Is a Force Multiplier

Real stories beat slides:

A new analyst learned Power BI via AI and rebuilt a legacy report pipeline in weeks, cutting a 3-day process to 15 minutes—then became the team’s expert.
Another hire automated trust-ops chores so effectively the org created a new role around AI enablement.

Companies are formalizing this momentum:

AI Ambassadors & Centers of Excellence to spread patterns and support.
Measure lift (time saved, throughput, quality) to scale what works.

Early-career pros often arrive “GPT-fluent” and can unlock quick wins. Pair their energy with structure (guardrails, coaching, metrics) and the compounding value shows up fast.

4) Security, Compliance & Government Work: Adoption Without Exposure

For defense and high-security environments, data handling determines tool choice. The path forward included:

Moving to segregated, compliant AI environments where sensitive and classified data never mingles with public corpora.
Updating MSAs and proposal processes to disclose when genAI is used (some RFPs now explicitly allow it—with transparency).
Org-wide access, not just devs: mission experts, analysts, and operators need safe, sanctioned AI, too.

5) The Real Bottlenecks: Infrastructure, Governance, and Culture Energy

Data + Cloud Foundations: Governance, pipelines, quality, GPUs/compute. No clean data, no durable AI.
Use-case discipline: Don’t “slap AI on everything.” Start with high-value, low-risk workflows. Prove concept, then scale.
Change fatigue: After years of disruption, employees are tired. Several orgs pair AI training with mindset work (“culture energy”) and hands-on learning labs tailored to teams’ real tasks to guide their mission alignment during the constantly changing demands.

6) Education: Adoption With Integrity

Students are already heavy users. Educators are balancing acceleration with academic growth:

No reliable AI detectors—so focus on teaching ethical choices, rather than monitoring.
Project-based learning where AI becomes a partner, not a shortcut.
Teach AI lifestyle applications (trip planning, side hustles, language learning) to get comfortable with the new technology—and then translate those patterns to school and work.

Build students’ introduction to AI and ethical views first, then amplify their tool application by teaching them how to use it.

7) What Employers Should Do This Quarter

Clarify boundaries: Approved tools, clear do/don’t, logging, and feedback.
Run an AI Sprint (4–6 weeks): Pick 3–5 repetitive workflows, benchmark time/quality, implement AI assist, re-measure, and implement AI in future workflows as appropriate.
Create multi-tier learning: Exec briefings, manager enablement (role-modeling), and team labs on real tasks.
Measure outcomes: Hours saved, cycle time, error rates, NPS/CSAT, employee sentiment.

8) What Students & Early-Career Pros Should Do

Build one personal project with AI end-to-end (site/app/report), then write a short case study with before/after.
Learn the stack you don’t know with AI as tutor (BI, automation, scripting).
Practice explainability: Be able to walk through how you used AI, where you verified, and why your approach is ethical.

9) Why This Matters (Especially in Hawaiʻi)

AI is a lever that favors smaller, faster players. We don’t need a billion-dollar data center to compete at the application and content layers. If we become the best adopters, Hawaiʻi can 10–100× outcomes in sectors that matter here—health, education, banking, defense, and entrepreneurship.

Quick CTAs to adopt AI

Internships & Early-Career Pathways: Multiple orgs are expanding AI internships and entry roles; keep an eye on local employers’ careers pages and university programs.
Community & Learning: Look for AI ambassador programs, hackathons, and workforce partnerships; many are open to students and career-switchers.
Educator Resources: Curricula and training materials are being shared publicly—tap them, adapt them, and share back.

AI won’t replace you but someone fluent, ethical, and explainable with AI will. Build the triangle. Go from experimenting to compounding impact.

Don’t forget! ChatGPT can make mistakes. Check important info.

–

We will be posting more about the adoption of AI, AI in Hawai‘i’s economy, and new technology tools. Follow us on social media or join or newsletter for live blog and event updates!