Radio’s AI Foes Exposed in New Rules

AI Voices on Australian Radio: New Rules Mandate Disclosure

Australia is stepping into the future of broadcast media with new regulations that will require commercial radio stations to clearly disclose when they are utilising computer-generated voices to host regular programming. These changes, part of the updated Commercial Radio Code of Practice 2026, aim to enhance transparency and address growing community expectations surrounding the use of artificial intelligence in broadcasting.

The Australian Communications and Media Authority (ACMA) revealed the revised rules on Tuesday. While the new code introduces mandatory AI disclosure for certain programs, it also includes significant exemptions. These exclusions cover AI used in radio streaming services, advertisements, and crucial updates such as weather and traffic reports.

These reforms come in the wake of a notable incident where a Sydney radio station, CADA, employed an AI-generated host named Thy for a regular music show without informing its audience. This move sparked debate and prompted calls for greater accountability from industry bodies, including a group representing Australian voice actors.

Key Changes to the Commercial Radio Code

Effective from July 1, the revised code mandates that commercial radio stations must explicitly reveal the use of an AI-generated, synthetic voice when it fronts a regularly scheduled music or news program. This disclosure is a crucial step towards ensuring listeners are aware of the nature of the content they are consuming.

The required disclosure can take several forms:

• During the Program: The announcement can be aired live within the program itself.
• Online Platforms: Stations can also make the disclosure available on their official websites or social media channels.

Beyond AI disclosures, the updated code also places a greater emphasis on content during sensitive times for young students. Commercial stations will be required to “exercise special care” on school days between 8am and 9am, and again from 3pm to 4pm. This provision aims to ensure that programming during these periods is mindful of children who may be listening, particularly during school drop-off and pick-up times.

ACMA Chair Nerida O’Loughlin highlighted that these reforms are a direct response to community concerns. “Broadcasting rules must keep pace with technology and with community expectations,” she stated. “Listeners want greater transparency about when AI is being used.” This sentiment underscores the authority’s commitment to aligning regulatory frameworks with evolving societal demands and technological advancements.

Exemptions and Future Considerations

While the new AI disclosure rules are a significant development, it’s important to note their limitations. The mandatory AI disclaimers will not extend to all forms of radio content. Specifically, the following categories are exempt:

• Weather and Traffic Updates: These essential, often time-sensitive reports will not require AI disclosure.
• Advertisements: Commercials utilising AI voices will not fall under the mandatory disclosure requirements.
• Music and Advertisements within Programs: Pre-recorded music or advertisements played as part of a larger program are also excluded.
• Streaming Platforms: Broadcasts delivered exclusively via radio streaming services are exempt from the new rules.

Despite these exemptions, Ms O’Loughlin encouraged broadcasters to adopt a more comprehensive approach. She suggested that stations should “voluntarily include AI disclaimers across all platforms, including on-demand services.” The ACMA has also indicated a willingness to consider further regulatory action if voluntary disclosures do not become widespread. “If they do not, we expect that new regulations may need to be put in place so that listeners know what to expect no matter how they tune in,” she warned.

Industry Reactions and Ongoing Debate

Commercial Radio Australia has welcomed the registration of the code. Chief Executive Lizzie Young commented that the AI transparency provisions represent a proactive approach to managing technological changes within the industry. “The code’s registration is a significant milestone, and we are pleased to see the ACMA working collaboratively with the industry to update the Commercial Radio Code of Practice,” Young stated.

However, the changes have been met with some criticism from those advocating for broader AI regulation. Simon Kennedy, president of the Australian Association of Voice Actors, expressed that while the AI transparency rules are a step forward, they do not go far enough. He argued that AI-generated voices are increasingly prevalent in commercial settings and that consumers have a right to know when they are not interacting with a human voice.

“It’s not close to good enough,” Kennedy told AAP. “We’d like to see it legislated that people have the right to scrutinise the media that they’re consuming based on the facts.” This perspective highlights a broader concern about the pervasive nature of AI and the need for robust consumer protection measures across all media platforms.

The federal government’s National AI Plan, released in December, did not propose specific legislative measures for AI governance, leaving the door open for further industry-led initiatives and ongoing public discourse on the ethical and practical implications of artificial intelligence.