Learnings for Government Comms from Opening the Black Box Phase 2

LLMs are now shaping how the public perceives government performance, and the picture they present looks nothing like the news we all consume. That's the central finding of Medianet's Opening the Black Box Phase 2 report, a study of how large language models like ChatGPT, Gemini and Claude cite and represent brands and public institutions.

Phase 2 analysed 957 LLM responses and hand-coded 11,127 citations, with a focus on the Victorian Government's presence in AI responses tied to the My Melbourne Road campaign. Four findings stand out for anyone working in public sector communications.

Medianet Managing Director Amrita Sidhu frames the shift as a governance issue, not a technology trend. "LLMs are a stakeholder that will need their own type of management," she said. "They are an emerging reputational channel that needs to be measured, monitored and managed. Our research is the first step in building the framework to do that."

AI models rate government performance far more favourably than the press does

Traditional media coverage of the Victorian Government was 50% unfavourable. LLM responses on the same subject matter were 14% unfavourable and 43% favourable.

It is the same government in the same period with two opposite verdicts, depending on which channel someone asks.

Jacquie, our Head of Insights, traced the gap back to sourcing. "In Phase 2, we drew a connection between the sentiment towards Vic Government and the content information contained in LLM responses," she said. "These were similarly lifted from Vic Government websites."

When journalists write, they apply scrutiny and editorial judgement. AI models pull from what's published and structured. Government websites, fact sheets and policy pages are written to inform, not to critique, so the sentiment skews accordingly. A department's own site is doing more reputational work in AI environments than most communications teams have accounted for.

Owned content is the largest single source LLMs cite

Government sources made up 44% of all citations in the study, ahead of corporate content at 23% and traditional media at 16%.

Individual Victorian Government domains carried the weight. vic.gov.au was cited at 5.7%. tac.vic.gov.au 3.8% , transport.vic.gov.au 2.3%, bigbuild.vic.gov.au 2.2%, premier.vic.gov.au 2.2%, and parliament.vic.gov.au 2%.

Combined, these six domains account for close to a fifth of every citation recorded in Phase 2. Victorian Government sources generated more than 3,500 citations, much more than any single corporate brand in the study.

AI models treat such structured, publicly accessible government content as fact. Agencies that publish detailed, well-organised information get cited on the strength of it.

An issue that fades from the news can still get cited in AI

News coverage of a government subject typically runs its course in two to four weeks. AI models operate on a different clock. Once an issue is indexed, it can stay in a model's responses indefinitely.

Adam Palmer, Media Analyst at Medianet, in our recent webinar, pointed to the Royal Commission into Antisemitism and Social Cohesion as a live case study. University leaders were absorbing a rough week of headlines, but he argued the headlines weren't the real risk. "What won't pass is the RC's official report," he said. "That means being very careful with the statements and written submissions they make to the RC, because that is what will influence the official record."

The same principle holds for any government agency facing an inquiry or review. Formal submissions and the public record carry more weight in AI environments than daily headlines do. The data backs this up. The 2019 Banking Royal Commission was still surfacing in LLM responses in 2025, six years on. A December 2024 ASIC ruling that drew heavy traditional media coverage hadn't reached LLM responses at all by the time of analysis. Some issues take six months or longer to appear in AI responses. Once they do, they don't leave.

The accountability gap

The dominance of owned content raises a harder question. If government material drives most AI citations, and AI models soften controversy by default, does public accountability suffer?

A critical story about government spending or policy failure can reach readers within days through traditional media. The same story might not surface in AI responses for six months, if it surfaces at all. In that window, someone asking an AI assistant about the same issue gets a materially different answer to someone reading the news. Public sector reporting specifically, where scrutiny of spending and policy decisions is the point, the distance adds up.

It's compounded by a separate structural issue. Much of Australia's highest-quality journalism sits behind paywalls or crawl restrictions, making it largely invisible to AI models regardless of its accuracy or public interest value.

Cited, but not credited

The research also flagged a pattern worth watching. In 137 responses, AI models cited RACV content verbatim without naming RACV at all. For a campaign built around public education, that may not matter as the message still resonates, but for a government program designed to build recognition of a specific initiative, funding announcement or agency, it can be a measurement problem. The department can do the work and get none of the visibility.

What does this mean for government comms?

Four priorities that follow directly from the data.

• Treat owned content as core infrastructure. A well-structured, detailed government website is the strongest driver of favourable AI representation available to public sector teams. It is worth developing further.
• Extend monitoring well past the news cycle. An issue gone quiet in traditional media can still be building toward an AI citation months later, and once it's cited, it tends to stay for good.
• Take formal submissions seriously as long-term assets. Be it Royal Commissions, parliamentary inquiries, or reviews, they all shape how AI models remember government performance for years after the headlines fade.
• Audit AI visibility with the same discipline applied to media coverage. Automated tools may capture scale, but LLMs need to be monitored as a separate channel to understand their own sentiment and share of voice.

Medianet tracks this through its own LLM monitoring tool, part of the Medianet Labs Suite.

Medianet Labs is an exclusive access for clients where we launch some of our most AI-forward tools that empower media monitoring. It currently has our LLM Monitoring tool and PR Optimiser live, with new tools such as Hansard Monitoring and Media Database Chatbot along the way. Learn more about it here.

The findings are from phase 2 of our latest pioneering research into generative citation analysis - Opening the Black Box.

Read more about the Findings from the report.

Download the full report here.

FAQ's

Do AI models like ChatGPT and Gemini favour brands more than traditional media?

Yes. Medianet's Opening the Black Box Phase 2 research found LLMs are consistently more positive toward brands than traditional news coverage. Negative issues, like branch closures and job cuts, often don't show up in AI responses at all. That's because LLMs lean on structured, authoritative owned content rather than news reporting, so a brand's own website can end up doing a lot of the reputational heavy lifting.

How long do reputational issues stay visible in AI responses?

Much longer than in the news cycle. Traditional media issues usually fade within two to four weeks. Once an LLM indexes an issue, though, it can keep surfacing in responses indefinitely. The 2019 Banking Royal Commission is a good example. It still shows up in AI answers about banks years later, while more recent controversies with less formal documentation can disappear from AI outputs within a year.

What type of content do LLMs cite most when talking about Australian brands?

Owned content dominates. The research found corporate and government websites made up 67% of citations, compared to just 4% from major metro newspapers. This is partly a quirk of the Australian media landscape, where a lot of quality journalism sits behind paywalls, making it invisible to AI crawlers. That leaves the door wide open for brands with strong, freely available owned content.

What is Medianet doing to help brands manage their AI visibility?

Medianet has built Medianet Labs, a space inside the Medianet platform giving users early access to tools like LLM Monitoring, which tracks how brands are mentioned and cited across Gemini, Claude and ChatGPT. It's available now to Medianet account holders under the Medianet Labs tab. Medianet also runs its own ongoing research into this space through the Opening the Black Box report series, with Phase 2 available to download now and Phase 3 on the way.

How can PR and comms teams improve their visibility in AI search results?

You need to monitor and track your LLM citations and mentions just like you do media coverage.

• Invest in owned content depth: LLMs treat well-structured, authoritative owned material as fact.
• Extend your monitoring horizon: Reputational issues can take six months or longer to surface in LLM responses, and once they do, they tend to stick around.
• Go beyond automated tracking: Monitoring tools capture scale, but understanding sentiment, themes and source material still requires the same analytical depth applied to traditional media.