Narrative Values Extractor
A simple tool demonstrator for seeing the moral story inside the news.
Today I’m posting about another simple demonstrator, the Narrative Values Extractor. This custom GPT is designed to take a news article and surface the values and conflicts that people and organisations (the actors) have between one another, for the reader. It produces an explainer which the reader can use to understand an issue in more depth.
The goal of this tool is to look beyond the positions that actors assert when making arguments, and to surface the underlying values that are used justify them. Most political or policy debates are really clashes between value systems that remain invisible. We argue about outcomes without first acknowledging the moral assumptions that shape what each side considers legitimate or fair.
I built it because I keep seeing people talk past one another. These are smart, well-intentioned individuals who weren’t disagreeing about the facts, they’re disagreeing about values. Our public debates have been flattened and have lost their moral literacy. The aim of this tool is to make those hidden assumptions visible again, so that conversations can start with understanding.
Our arguments aren’t just about facts or interests, they’re about what people care about most, often without realising it. By tracing those hidden values, we can start to see why certain conflicts feel unsolvable and where dialogue might actually begin. The Narrative Values Extractor doesn’t tell us who’s right or wrong; it helps us understand why people take the positions they do.
How it works
The custom GPT works by taking a narrative text such as a news article, editorial, or statement, and producing a short, structured values map. Instead of summarising events, it identifies the groups involved, the values they claim, how they frame the issue, and what solutions they prefer. It also surfaces the conflicts between groups and suggests possible ways forward. The result is a human-readable report that outlines the moral and normative information often hidden inside public narratives.
The tool follows a strict step-by-step process:
Is given a purpose: to read a single narrative and output a compact, structured values map.
Is given the output format and output mode.
Ingests URL, file, or copied block of text.
Discovers the actors named in the narrative:
Enumerates over name groups and actors.
Merges duplicates.
Requires that actors be relevant to values map before recording them.
Extracts values:
Extracts values nouns and noun phrases.
Separates stated values from inferred values.
Evaluates the evidence with discipline:
Using quotations where possible.
Includes citations when browsing is on.
Produces a conflict map:
Lists value-vs-value clashes as X ↔ Y pairs.
Notes narrative devices.
Surfaces asymmetries of power, voice, risk, or information.
Proposes bridging hypotheses:
2-4 practical ideas that honour both sides’ values.
Is concrete in recommendations.
Checks output for quality, bias, insufficient information, missing groups.
Limitations
The tool sometimes gives infers actors based on the text. This is referenced, but it’s something that I’m considering explicitly excluding because it can cause confusion. You can see that the example at the end of this piece references the NSW licensing authority who weren’t quoted in the article. I’ve kept it in for transparency, it shows the limitations of this approach.
The stated values don’t match any values framework. They are the best match of the LLM to what it considers human values to be. That’s ok for this project, because this is a first pass where you extract the values from a narrative before aggregating it across a group corpus and building the values map from that. In a larger project I’d be taking thousands (or hundreds of thousands) of these outputs, then collapsing them into the main threads to discover the fundamental values.
I also wouldn’t recommend using the outputs as a source of ultimate truth. These are designed like I build intelligence tools. They point a user in the right direction, reduce uncertainty, surface indicators that might inform more in-depth analysis, that kind of thing. We’re the moral agents here, not the LLMs. It means we need to use our own judgement, this is just to help.
Using the tool
Like the Terms of Service Evaluator, it’s pretty simple. All you need to do is open the custom GPT, paste the URL or text, and let it do its thing. If you can, turn on thinking mode, it gives a much better response.
The link to the tool is here.
This works best with articles that are rich in values statements and have at least two opposing sides. I have also tested it out with texts like the poem The Man from Snowy River and the French national anthem La Marseillaise. The results were very pretty cool. Still, I’d try it with investigative news articles first, particularly those which talk about a wrong being committed.
How it fits in the bigger picture
The aim of this proof of concept is to demonstrate how you can extract values from a text. The techniques it uses are the same as more complex processes I’ve built, such as the Political Values Analysis tool, simplified so that it can be used by the public. But what’s important is that it shows that LLMs can make explicit the values that are just under the surface of contested issues.
I think this might be the most practical of the custom GPTs I’ve built so far. It’s something anyone can use when they’re trying to make sense of a complex issue by understanding the moral terrain underneath.
There is an example of an output below that you’re welcome to try yourself. Next week we’re going to get to the good stuff, to Alignment, the central challenge of this series.
Appendix: Example output
The output below was produced by the tool after analysing the Australian Broadcasting Corporation article Would you shoot a wombat? In Taralga, the answer isn’t as simple as you’d think. by Tynan King. This article is a great one for this kind of analysis because it examines a contained local issue and details the values of the actors quoted.
Would you shoot a wombat? In Taralga, the answer isn’t as simple as you’d think — Values Map (ABC News, July 4, 2025)
1. Groups
A. Wildlife carers (Dianna & Warwick Bisset / Rocklily Wombats)
Who: Retiree couple caring for injured/orphaned bare-nosed wombats near Taralga; advocate humane coexistence.
Core values: Animal welfare; biodiversity protection; humane coexistence; legality; community care; (inferred:) stewardship.
Problem framing: Native wildlife is being illegally harmed; habitat is degraded; locals normalise violence against wombats; carers face hostility for speaking up.
Perceived harms: Illegal shootings and cruelty; habitat destruction from roadworks; social ostracism and threats (e.g., tyres vandalised).
Preferred solutions: Non-lethal measures (wombat gates in fences, mange treatment devices); release rehabilitated animals away from farms.
Legitimacy claims: Law (Biodiversity Conservation Act 2016 makes harming natives without licence an offence); evidence from caring work; moral claim of animal dignity.
Red lines: Killing as routine population control; habitat clearing that ignores wildlife. (Inferred from quotes and actions.)
Representative quotes/facts:
“I think they’re beautiful animals … and we’re so lucky to have them.”
“It is an offence under the Biodiversity Conservation Act 2016 to harm native animals without a licence.”
B. Local farmers (e.g., Chris Croker; some neighbours)
Who: Multi-generation producers managing stock and property; face damage from wombat burrows and fence breaches.
Core values: Property rights; farm safety; livelihood security; autonomy; practicality; (inferred:) fairness in regulation.
Problem framing: Wombats have increased locally, damage infrastructure and landscapes, create safety risks; rules tie farmers’ hands; city-style values imported by newcomers.
Perceived harms: Vehicle/quad bike hazards; erosion/gully formation; fence failure enabling pests; significant repair costs (“about a quarter of a million bucks”).
Preferred solutions: Easier farmer-led control (including lethal options); tighter control on where rehabilitated wombats are released.
Legitimacy claims: Practical knowledge; economic hardship; public safety; (inferred:) historical presence/identity.
Red lines: External “green” interference blocking farm operations; bureaucrats deciding case-by-case without local discretion.
Representative quotes/facts:
“They really need to allow the farmers to do what they have to do.”
Fence/creek damage and repair quote “about a quarter of a million bucks.”
C. Mediating local voice (Stirling Dixon, farmer & community bridge)
Who: Long-time farmer friendly with both sides; articulates the cultural split (“wombat cuddlers” vs “wombat shooters”).
Core values: Community cohesion; mutual recognition; historical identity; (inferred:) pragmatic coexistence.
Problem framing: Deep culture clash—moral concern for animals vs financial necessity; colonial/settler mentality persists.
Perceived harms: Community division; entrenched positions; rising incivility.
Preferred solutions: “Find neutral ground” (acknowledges both sides’ valid points).
Legitimacy claims: Long experience; relationships across the divide.
Red lines: (Inferred) Escalation that destroys social fabric.
Representative quote: “One’s a bloody wombat cuddler and the other one’s a wombat shooter.”
D. Upper Lachlan Shire Council / project contractors
Who: Local authority executing Wombeyan Caves Road upgrade; focal point in habitat dispute.
Core values: Regulatory compliance; tourism access; infrastructure delivery; environmental due process.
Problem framing: Works complied with legislation; ecologist approvals; paused to address issues after concerns raised.
Perceived harms: Allegations of habitat destruction; community dissatisfaction with finished road.
Preferred solutions: Proceed within legal frameworks; manage impacts per ecological advice.
Legitimacy claims: Law/policy compliance; expert certification (qualified ecologist).
Red lines: (Inferred) Halting works absent process-based cause.
Representative facts: Council says all clearing approved by a qualified ecologist; identified wombat burrows inactive.
E. Wider local community & business/tourism voices (inferred/partly quoted)
Who: Residents and businesses concerned about jobs/tourism; some hostile on social media to carers.
Core values: Local economy; place identity; social solidarity; (inferred:) respect for established norms.
Problem framing: Conservation activism blamed for sub-par road outcome and economic loss.
Perceived harms: Threat to livelihoods/tourism; social friction and boycotts.
Preferred solutions: Less obstruction to development; (inferred) prioritise access/amenities.
Legitimacy claims: Community interest; economic outcomes.
Red lines: Activism seen as harming town prosperity (inferred).
Representative facts: Calls on Facebook “to not serve [Dianna] in town.”
F. NSW licensing authority / environmental law (implicit actor)
Who: State agencies administering licences to harm protected fauna under Biodiversity Conservation Act 2016.
Core values: Rule of law; conservation; risk-based control; proportionality.
Problem framing: Harming wombats illegal unless licensed; licences rare and require exhausting non-lethal options.
Perceived harms: (Inferred) Illegal harm; unmanaged conflict.
Preferred solutions: Case-by-case licensing; evidence thresholds.
Legitimacy claims: Statute and administrative process.
Red lines: Lethal control without need and process.
2. Conflict map
Value clashes
Animal welfare / biodiversity protection ↔ Property rights / livelihood security.
Legality / procedural compliance ↔ Local autonomy / practical discretion.
Tourism/economic growth ↔ Habitat protection / precaution.
Community cohesion ↔ Adversarial activism / social sanctions.
Narrative devices
Labeling: “wombat cuddlers” vs “wombat shooters.”
Identity cues: “greenies”; “colonial mentality”; newcomer vs old-timer.
Moral vs economic framing: “living creatures” vs “sacred entitlement to make a living.”
Asymmetries
Power: Licensing authority holds coercive/legal power; farmers hold land-access power; carers have limited formal power.
Voice: Farmers rooted in local identity; carers face ridicule/boycotts; council speaks via official statements.
Risk/Exposure: Farmers bear direct repair/safety costs; carers face harassment; wombats face lethal harm.
Information: Ecologist assessments vs lay observations; disagreement about release rules (note: article states national parks release is barred, while a farmer claims “no rules on where they let them go”).
3. Bridging hypotheses
Co-funded “wombat-safe infrastructure” program
Establish a local fund (council + state environmental grants + landholder co-pay) for: wombat gates at fence lines, culvert/passages near known burrow corridors, and erosion control at creek banks. Eligibility tied to sharing before/after data (photos, maintenance logs). (Aligns animal welfare with property protection; reduces repair burdens.)Transparent, tiered control protocol with independent verification
Publish clear local criteria for when non-lethal steps are considered “exhausted” (e.g., documented use of gates/barrier designs, deterrents over a set period) before any lethal licence is supported. Include an on-farm visit by an independent adviser (ecologist/ag extension) and time-bounded trial. (Respects law while giving farmers a predictable pathway.)Release registry + buffer-zone rules
Create a simple registry for rehabilitated wombat releases: record GPS vicinity, land-tenure consent, and minimum distance from active grazing areas and national parks (parks already excluded). Share aggregated heatmaps with farmers to reduce conflict. (Addresses farmers’ concern about “where they end up” and carers’ need to release safely.)Roadworks wildlife protocol with community oversight
For future upgrades, seat a small oversight group (council, ecologist, carer, farmer) to sign off on pre-clearance burrow surveys, stop-work triggers, and remediation steps; publish a short post-works impact note. (Builds trust that habitat is considered while enabling infrastructure delivery.)
Notes on evidence sufficiency
Direct quotes and factual claims are taken from the ABC article; where the piece offered conflicting perceptions (e.g., release rules), both are recorded with citations. If fuller regulatory detail is needed (precise licensing thresholds, rehabilitation release protocols), additional sources beyond this article would be required.