AB 2392 was signed on September 30, 2026. The California Community Colleges Chancellor’s Office and California State University must establish a joint working group on generative artificial intelligence, with the University of California requested to participate. The law pairs procurement safeguards with employee training and requires recommendations on responsible use by January 1, 2028.
Source checked October 1, 2026 · CHAPTER 855, STATUTES OF 2026 · GENAI + HIGHER EDUCATION + PROCUREMENT · Official source ↗ · AI-assisted
WDB decision strip
| Status | Action | Primary owners | Key dates | Impact |
|---|---|---|---|---|
| SIGNED LAW | USE THE HIGHER-ED FRAMEWORK AS A BENCHMARK FOR LOCAL AI GOVERNANCE | College Partnerships + Data/AI Governance + Training / Procurement Leads | Recommendations due 1/1/28 · annual training review | College partnerships · AI literacy · Procurement · Data governance |
The bottom line
AB 2392 does not regulate local WDBs. Its workforce value is indirect but useful. Community colleges are core workforce-system partners and training providers, and the law creates a concrete California model for evaluating generative AI procurement, privacy, data use, accuracy, and employee training. WDBs can use that direction as a benchmark for their own AI governance without claiming the higher-education requirements automatically apply to them.
What changed
| Change | What the law does | WDB implication |
|---|---|---|
| Joint working group | The California Community Colleges Chancellor’s Office and California State University must establish a joint working group, with the University of California requested to participate. | WDBs should expect statewide higher-education AI practices to become more standardized over time. |
| Procurement protections | The framework must address risk, bias, potential harms, data use and retention, data minimization, auditing, and transparency about data sources and vendors. | These are useful screening questions for workforce technology purchases even where the statute does not apply. |
| Employee training | Covered systems must provide training related to generative AI policies, responsible use, privacy, and limitations. | AI literacy is moving from optional experimentation toward a normal part of organizational capacity. |
| Recommendations due | The working group must submit recommendations by January 1, 2028. | Local partners should watch the state framework before creating conflicting or unnecessarily complex local standards. |
| Annual review and records | The law requires annual review of training content and retention of training-completion records for four years. | Responsible AI practice includes ongoing learning and documentation, not one-time training. |
| Interim procurement reporting | Until standards are adopted, specified systemwide generative AI contracts trigger a report within 60 days after procurement. | Transparency about major AI purchases becomes part of the governance model. |
Operational considerations for WDBs
- Designate a higher-education AI contact. Know who at the local community college or university is following the statewide working group so workforce partnerships are not surprised by changes in procurement or training practice.
- Compare local AI practice with the state framework. Review privacy, data minimization, vendor transparency, output limitations, human review, and staff training in any WDB AI tools.
- Align AI literacy where useful. If college partners develop staff or student training that fits workforce-system needs, explore shared learning rather than building a separate curriculum from scratch.
- Use procurement questions before buying tools. Ask what data enters the system, where it goes, how long it is retained, what the vendor can reuse, how outputs are checked, and what risks have been assessed.
- Watch the January 2028 recommendations. Treat the final state guidance as an evidence source for future local governance decisions rather than guessing now what the framework will require.
Required / local choice / good practice / watch out
| Label | How to apply it |
|---|---|
| Required | The statutory duties apply to the specified public higher-education systems. Local WDBs are not made subject to AB 2392 by the law. |
| Local choice | A WDB may voluntarily borrow the law’s procurement, training, privacy, and documentation principles for its own AI governance or joint work with education partners. |
| Good practice | Treat responsible AI as an operating system: clear procurement questions, data protections, human accountability, staff training, regular review, and documentation. |
| Watch out | Do not describe AB 2392 as a statewide AI credential, a WDB mandate, or a finished set of statewide standards. The working group recommendations are still to be developed. |
Source + Trust Record
| Primary authority | California Legislative Information – AB 2392 |
|---|---|
| Signing confirmation | Governor of California – September 30 AI and worker-protection package |
| Current status | CHAPTER 855, STATUTES OF 2026 |
| Source checked | October 1, 2026 |
| What we verified | AB 2392 is Chapter 855, Statutes of 2026. It requires a joint public higher-education working group, recommendations by January 1, 2028, employee training and annual review, four-year training-completion records, procurement safeguards, and interim reporting for specified systemwide generative AI contracts. |
| Important limitation | AB 2392 applies to the public higher-education segments identified in the statute. Workforce Wonkery’s suggestions for WDB use are voluntary operational applications, not statements that the law governs local workforce boards or all training providers. |
Official sources control. Workforce Wonkery is AI-assisted and does not receive human legal or compliance review.
Leave a Reply