Technology system oversight and accountability

---

Recognizing the expanding influence of digital systems in the workplace, unions have pursued contractual provisions that strengthen oversight, clarify responsibilities, and protect worker rights in the governance of technology. Provisions in this section reflect efforts to maintain human judgment in system deployment, require clear policy communication, and preserve legal protections in evolving technology contexts.

The examples below are organized by focus area:

• Human oversight and system testing and validation: Ensures that new technologies are reviewed, tested, and subject to editorial or employee oversight.
• Technology policy development and disclosure: Requires employers to develop, share, and update clear policies around technology use.
• Legal and contractual rights, and avenues for recourse: Clarifies that technology-related disputes remain subject to existing agreements, protections, and legal claims.

Human oversight and system testing and validation

These provisions aim to prevent unchecked automation and ensure that technologies—especially those involving AI—are reviewed, validated, and subject to direct human oversight before being used in the workplace.

Examples:

• “[The Employer] agrees that all application software utilized by UNION-represented employees in the normal course of their duties shall be tested, prior to installation and roll out, for proper functionality.” (CTU-AFT-PCBA11)
• “[The Employer] will commit to follow Associated Press (AP) and Society of Professional Journalists (SPJ) guidelines on the use of Artificial Intelligence (AI) and will treat AI-generated content as source material that requires editing and fact-checking by a human prior to being published.” (MNCG-CWA-PCBA32)
• “Generative AI may be used to perform the work of news production only with the direct involvement and oversight of employees in compliance with AP standards.” (TNG-CWA-PCBA33)
• “An internal standard will be developed to control remote access (externally) of equipment [new technologies and artificial intelligence]. The goal will be to maintain the safety, security and integrity of the equipment while being accessed.” (IATSE-PCBA46)
• “The Company shall develop procedures and processes for oversight and evaluation of AI systems in the workplace.” (IAMAW-PCBA54)
• “Written stories, articles and images created by GAI, in part or in whole, shall first be reviewed by an editorial employee (bargaining unit or non-bargaining unit), and will have distinct signifiers in the byline indicating that it is GAI-created, or if no byline, then a disclaimer in the body copy.” (WGAE-PCBA78)
• “Any content that is created by Generative AI shall only be done at the direction of and with the editorial review of human beings with editing responsibilities, and no such content shall be published without such involvement.” (ZDCG-TNG-CWA-PCBA111)

Technology policy development and disclosure

Unions have negotiated for transparency and inclusion in the development of technology use policies, including obligations for timely disclosure and opportunities to review and comment on proposed changes.

Examples:

• “The Company shall establish a policy on the use of GAI within 12 months of ratification of the CBA, and shall first provide the union with a draft of the policy with a twenty (20) calendar day opportunity to provide comments and input, which the Company will consider in good faith. If the Company intends to change the policy, it shall provide the union with a twenty (20) calendar day notice with an opportunity to provide comments and input, which the Company will consider in good faith.” (WGAE-PCBA78)
• “Should a Company furnish a writer with written material produced by [Generative AI] GAI which has not been previously published or exploited, and instruct the writer to use the GAI-produced material as the basis for writing literary material: The Company shall disclose to that writer that the written material was produced by GAI.” (WGA-PCBA1)
• “The [Employer] is responsible for documenting training, and providing applicable rules and guidance on personal use of material or [information systems and telecommunication] equipment to employees (i.e. to include updates, and disciplinary and adverse action guidance) as required, or not less than annually.” (AFGE-OPM123)

Legal and contractual rights, and avenues for recourse

These agreements affirm that workers retain enforceable rights when new technologies are introduced, including arbitration pathways and protections over the use of their work and identity.

Examples:

• Both parties recognize the evolving legal landscape surrounding Generative AI (GAI) and reserve rights regarding GAI-related claims and the use of literary material in AI training. For instance, nothing in this article prevents a writer who has reserved rights, or the Union on the writer’s behalf, from claiming that using their literary material to train, inform, or otherwise develop GAI software or systems falls within those rights and is not allowed under applicable law. (WGA-PCBA1)
• Claims for violations of this agreement arising from the Producer’s creation, utilization, or unauthorized alteration of digital replicas or creation and use of AI-generated “synthetic performers” shall be subject to arbitration. The available remedies shall be limited to monetary damages. (SAG-AFTRA-PCBA2)
• “Employees who are assigned to utilize an AI System to perform services, including by inputting prompts or otherwise overseeing the use of the AI System, shall continue to be covered under the terms of the applicable Agreement while performing such work.… Except as explicitly set forth herein, it is understood that this Article does not expand or contract any existing rights and obligations under applicable Agreements. Nothing herein alters the scope of coverage under those Agreements.” (IATSE-PCBA46)

---