344
Productivity & Workflow355
Automation & Workflow224
Software Development251
Marketing & Growth192
AI Infrastructure & MLOps174
Writing & Content Creation203
Data & Analytics141
Photography & Imaging156
Design & Creative170
Customer Support131
Sales & Outreach125
Voice & Speech135
Education & Learning131
Operations & Admin87
A Canadian legislator read a line that sounded like an AI tool’s editing instruction during a floor speech, and the clip later spread on social media.
In short: A New Brunswick legislator appears to have read an AI writing instruction out loud during an official floor speech.
Bill Oliver, a member of the Progressive Conservative Party in New Brunswick’s legislative assembly, gave a speech that included a strange line that did not fit the rest of his remarks. After making a point about the limits of “advocacy offices,” he then read: “here’s a more natural, flowing version of that section that reads like a legislative speech rather than a series of short points.”
That line sounds like text from an AI writing tool. Many large language models, which are AI systems that generate text (like an autocomplete that can write full paragraphs), will respond to a user by offering a rewritten version in a different style. The sentence Oliver read sounds like the tool explaining what it is about to do, not something meant for the public record.
The moment did not draw much attention right away. Video of the speech later spread on social networks including Reddit and Threads, and it then picked up wider coverage from Canadian news outlets.
Tools that help draft and rewrite text are now common at work, including in politics. This incident shows a simple risk: if someone uses AI to polish a speech and does not carefully edit the output, they can accidentally include behind-the-scenes instructions (like leaving the chef’s recipe notes on the plate). It also adds to a growing list of public mistakes tied to AI-written text, which can affect trust and credibility.
Source: Arstechnica