Canadian Politician Reads AI Chatbot Prompt in Legislature

A Canadian legislator used a Large Language Model (LLM) to generate a speech and read the resulting AI-generated prompt aloud during a session in the legislature, according to reports from Ars Technica and 1News. The incident highlights the increasing integration of generative artificial intelligence within government legislative processes and the potential for automated content to enter official public records.

The politician read the chatbot’s response directly from a device or script during a floor speech, according to Ars Technica. While the specific identity of the legislator and the exact contents of the prompt were not detailed in the initial reports, the act of reading an AI-generated response into the record marks a distinct shift in how legislative rhetoric is produced and delivered in Canada.

Generative AI tools, such as those developed by OpenAI or Google, are designed to predict and produce text based on patterns in training data. When used for speechwriting, these tools can create structured arguments or summaries, but they are known to occasionally produce “hallucinations” or factual inaccuracies, a risk that becomes central when such text is read as official testimony or political positioning in a government body.

The use of AI in the Canadian legislature occurs as governments worldwide grapple with the ethics of automated content. According to Ars Technica, the event serves as a practical example of how LLMs are moving from private productivity tools to public-facing communication instruments.

The integration of AI in legislative settings raises questions regarding transparency and the authenticity of political discourse. Because LLMs synthesize existing information rather than engaging in original critical thinking, the use of such prompts in a floor speech suggests a delegation of rhetorical drafting to an algorithm.

This development follows a broader trend of AI adoption in North American governance, where tools are being tested for drafting legislation, analyzing complex policy documents, and managing constituent communications. However, the direct reading of a prompt into a legislative record is a more visible application of the technology than the behind-the-scenes administrative uses.