✕ Loaded Labels [5/10]: The headline uses a rhetorical question and a metaphor ('Agents of illusion?') that introduces skepticism and frames the technology as potentially deceptive. This leans into doubt without asserting it directly, bordering on editorializing in the headline.

"Peter Griffin: Agents of illusion? AI agents promise a smarter state but NZ is nowhere near ready to run one"

✕ Headline / Body Mismatch [8/10]: The lead paragraph neutrally introduces the technology and its capabilities by summarizing Google’s announcement without immediate judgment. It sets up the core subject clearly and factually.

"At its May 2026 Google I/O conference in California, the company unveiled Gemini Spark, a so-called AI agent designed not just to respond to prompts but to act."

✕ Loaded Language [6/10]: The article uses measured, analytical language overall, but includes phrases like 'we’ve seen this movie before' and 'a mess' that inject informal tone and subtle editorial judgment.

"The result wouldn’t be a sleek, automated state. It would be a mess, one where errors propagate faster, accountability becomes murkier..."

✕ Dog Whistle [5/10]: The phrase 'move fast and break things' is a well-known Silicon Valley ethos, and its use here carries critical connotation, subtly mocking the government’s approach by associating it with reckless innovation.

"Governments are not startups. They cannot simply 'move fast and break things' without breaking trust, services and, ultimately, themselves."

✕ Appeal to Emotion [9/10]: The article avoids overt fear or outrage appeals, instead building a case through logic and structural critique. Emotional language is minimal and mostly restrained.

✕ Single-Source Reporting [8/10]: The article relies primarily on the reporter’s analysis and general references to government policy, without quoting specific officials, experts, or stakeholders from the government, tech sector, or public service unions. There is no named source supporting the AI-driven restructuring plan.

✕ Vague Attribution [7/10]: While Google, OpenAI, and Anthropic are mentioned as developers, their claims are not directly quoted or attributed in detail. The article summarizes their work without reproducing promotional language uncritically.

"In Google’s framing, it’s less like search and more like staff, working away for you 24/7 even when you turn off your computer."

✕ Framing by Emphasis [9/10]: The article frames the issue as a critique of technological solutionism and premature automation, focusing on institutional readiness rather than political blame. This is a substantive, systemic angle rather than episodic or moralistic.

"Until we fix the foundations, promises of an AI-powered public sector are just that: promises."

✕ Narrative Framing [10/10]: The piece avoids reducing the issue to a simple conflict between 'pro-tech' and 'anti-tech' sides, instead arguing that AI has potential but is being misapplied. It resists moral framing and instead emphasizes structural limitations.

"But that requires investment – in infrastructure, skills, governance – and carefully staged implementation."