AI4WARD
Leading Institution: California State University, San Bernardino
Quarterly Research
DCWF aligned
public repository
Human + AI + Superior process
Rapid Research for the AI-Shaped Cyber Workforce
AI4WARD helps the CAE community keep pace with the way artificial intelligence is changing cybersecurity work. The initiative turns fast-moving changes in tools, tasks, threats, and workforce expectations into practical research, training, challenges, and community resources that institutions can use now — not years from now.
Built around four connected tracks — Probe, Prepare, Prove, and Promote — AI4WARD asks what students, faculty, and practitioners can do better when humans and AI work together. Probe supplies the evidence: quarterly research cycles grounded in the DoD Cyber Workforce Framework and shared across the national CAE community.
Explore Probe
Four entry points for researchers, ambassadors, faculty, and reviewers.
Rapid Research
TURN QUESTIONS INTO EVIDENCE
Probe runs quarterly research cycles on how AI is changing cybersecurity tasks, tools, and roles. Each project is scoped to be answerable in one cycle and ends in a useful public report.
Cycle diagram
Research with AI
USE AI, BUT VERIFY EVERYTHING
Topics focus on what AI is replacing, augmenting, or creating in cybersecurity work — and what cyber students need to demonstrate to be hirable in the current workforce.
Topics
CHOOSE WORK THAT MATTERS NOW
Topics focus on what AI is replacing, augmenting, or creating in cybersecurity work — and what cyber students need to demonstrate to be hirable in the current workforce.
Rubrics & Submissions
SUBMIT A FOCUSED PROJECT
Submissions are reviewed for feasibility, evidence access, predictive value, workforce relevance, and non-duplication. Strong proposals name evidence and define what would make the forecast wrong.
1. Rapid Research
Probe is built around quarterly rapid research cycles. The work starts with candidate topics, scopes them into questions answerable in twelve weeks, supports the researchers as a cohort, and publishes a quarterly report, forecast summary page, webinar, and public repository entry.
Research Paper Section Template
2. Research with AI
Probe studies AI-augmented work by using AI-augmented research methods. AI can help researchers search, compare, summarize, draft, and analyze, but it is never the sole source of factual claims. The researcher remains responsible for every source, citation, conclusion, and forecast.
Prediction Monitoring
Every submission includes at least one forecast with a claim, horizon, confidence, and falsifier. After publication, an agent searches public news and articles for evidence that supports or refutes the prediction. The next publication includes the prediction and the article list. A prediction continues to appear until clearly refuted, then moves to the archived section with the evidence that triggered the archive judgment.
3. Topics
Topics may come from AI Horizon findings, CAE community needs, government and industry priorities, published research, or prior-cycle nominations. The key test is whether a topic can become a focused question answerable in one quarter and useful to the CAE community.
4. Rubrics & Submissions
Strong submissions are focused, feasible, evidence-backed, workforce-relevant, and predictive. The proposal should name the question, identify the evidence, explain how AI will be used responsibly, map the work to DCWF roles or tasks, and describe what would make the forecast wrong.
Cohort and Awards Model
Selected researchers should not work alone. The current doctrine proposes a quarterly cohort with a shared Slack channel and an AI agent to support the process and alert the PI when issues arise. The doctrine also records unresolved budget questions around award levels and travel support.
Binding vs. Proposed
Binding: quarterly cycles, four reports per year, at least four webinars per year, a public repository, distributed CAE teams, DCWF mapping, and the Human + AI principle.
Proposed / still under review: the exact twelve-week schedule, six-part submission structure, four-part forecast form, AI methodology rules, scoring rubric, cohort details, and starter topic list.
Sources used: Probe Cycle Doctrine, draft dated 2026-08-24, and AI4WARD Project Narrative dated 2026-07-28. Vault mirror freshness checked: 2026-09-11 04:30 PDT. The doctrine is marked draft; proposed items should not be presented as approved requirements.
Contacts