Bill Oliver, a Progressive Conservative member of New Brunswick's legislative assembly, accidentally read an AI-generated prompt during a recent floor speech. The gaffe went viral on social media, highlighting the risks of relying on large language models for official government communication.
The "natural, flowing version" that exposed Bill Oliver
During a speech last month, Bill Oliver was discussing the potential dangers of establishing advocacy offices when he inadvertently read a phrase typical of an AI chatbot: "here's a more natural , flowing version of that section that reads like a legislative speech rather than a series of short points." According to the report, this slip-up went largely unnoticed during the live session but later gained significant traction on social media platforms including Reddit and Threads.
The error occurred while Bill Oliver was arguing that citizens often develop expectations for advocacy offices that exceed the actual powers granted to those entities. By reading the AI's instructional text aloud, the legislator effectively revealed the mechanical process behind his speechwriting.
The "Borg" divide and the National Post's critique
The incident has sparked a wider conversation about the delegation of intellectual labor in the public sector. The National Post characterized the gaffe as evidence of a growing societal divide between "the elites," who are happy to delegate their duties to "the Borg," and the general public who find such automation objectionable.
This reflection suggests that the public may view the use of AI in legislative speeches not merely as a tool for efficiency,but as a surrender of authentic representation. As generative AI becomes more common in political offices, the gap between a politician's actual voice and their AI-curated persona becomes a point of political vulnerability.
Sullivan & Cromwell's "AI hallucinations" in the Prince Global Holdings case
The risks of AI automation extend beyond political speeches into the high-stakes legal sector. as the report notes, Andrew Dietderich, co-head of Global Finance & Restructuring at the Wall Street firm Sullivan & Cromwell, admitted in a letter to Chief Judge Martin Glenn that a previous filing for Prince Global Holdings contained "AI hallucinations."
In the legal context, these hallucinations invovle the fabrication of case citations, the misquoting of authorities, or the generation of entirely non-existent legal sources. The admission from Sullivan & Cromwell underscores that even elite institutions are struggling to maintain quality control when integrating artificial intelligence into professional workflows.
Whether Bill Oliver or the New Brunswick assembly will issue a formal apology
Despite the viral nature of the video footage, the source provides no record of a formal apology or explanation from Bill Oliver or the Progressive Conservative Party of New Brunswick. It remains an open question whether this was an isolated incident of poor editing or if the legislator has a broader, systemic practice of using AI for his official duties.
Furthermore, the report does not specify if the New Brunswick legislative assembly has established formal guidelines regarding the use of AI in official records. The lack of a public response from Bill Oliver leaves a vacuum regarding the transparency standards expected of Canadian legislators using generative tools.
Wynton Hall's "AI Nation" and the MAGA strategy
The broader geopolitical struggle for AI dominance is highlighted by Wynton Hall,the social media director for Breitbart News. Hall's book, AI Nation: A Plan to Create a Trillion Jobs, Solve Inequality, and Save Democracy, argues for a MAGA-led approach to the technology.
Hall claims that a specific strategic plan is necessary to create AI positions that benefit humanity without handing control of the United States to "leftists of Silicon Valley" or allowing the Chinese government to dominate the global AI landscape.
Comments 0