What is Understories and who is it for?
Understories scores climate messages — a press release, a social post, a campaign email — against eight evidence-based criteria and flags ethical risks. You get structured, research-backed feedback on what's working, what isn't, and why.
It is built for the people who need it most: grassroots organisers, underfunded NGOs, advocates writing their “final final” copy at midnight before a campaign launch, and those who often lack access to the communications expertise and research. Understories closes that gap.
How does Understories use AI?
Understories’ AI architecture is built on governance layers: research, narrative, and ethics. It combines RAG (Retrieval-Augmented Generation) with a large language model, currently Claude by Anthropic, to apply a structured framework to evaluate messages. All is grounded in the research library: no black boxes. Understories suggests. You decide.
AI also assists in native language editing for report page summaries, proofreads all native English content, and—via Framer’s assisting agent—corrects and refines UI/UX elements, SEO, and regulatory compliance, from desktop to mobile.
A general AI tool synthesizes an answer from whatever it's trained on or retrieves in the moment: you can't see what was excluded, how rigorous the underlying research was, or whether it's been peer-reviewed. Understories takes the opposite approach: every message is analysed through evidence from the research library, whose entries are human-selected, scored against eight evidence-based criteria, and reviewed through an ethics lens before publication. So instead of a synthesized answer, you're seeing the underlying evidence and its provenance — which matters if you're building a campaign message you'll need to defend.
How is the research exactly used?
The Understories Research Library draws on over 50 peer-reviewed studies, meta-analyses, practitioner guides, and toolkits — organised across five categories: Communications, Psychology, Narrative, Public Opinion, and Mobilisation.
Each entry has been assessed against the eight narrative criteria and the six ethical principles. Concretely: when Understories tells you your message scores low on Hope/Doom Balance, it's applying findings from specific studies — Moser & Dilling (2011) on how doom without solutions triggers psychological withdrawal, or Peters & Salas (2022): it is not just that doom feels bad, but that it structurally removes the condition necessary for action. The research isn't used to generate generic advice. It produces a specific, evidence-grounded assessment of your specific message.
What is the methodology behind the Research Library?
The Research Library is a personally curated collection, shaped by the expertise and perspective of Caroline Prak, its creator. It is built across four criteria to ensure rigor and diversity: Focus: It centers on strategic communications, serving advocates without ideological bias. Content Mix: It includes foundational, empirical, and practical materials for practitioners across communications and content production. Temporal Scope: It spans multiple decades, featuring research that reinforces, contradicts, or extends prior work.
Authorship & Representation: It prioritizes peer-reviewed, publicly authored works by a gender-balanced mix of authors and institutions, with explicit attention to geographical diversity (North/South) when possible. Note: While peer review is the gold standard, exceptions may be made for seminal works that meet other criteria. The library excludes private, corporate, or profit-motivated research to maintain independence and focus on public-interest advocacy. All works are included with respect to copyright and licensing, and authors/institutions were contacted for permission where necessary.
Does Understories work for audiences outside the Global North?
Honestly: with caveats. The Library reflects where most climate communication science has been conducted: the United States, the United Kingdom, Western Europe. Many studies were run on WEIRD samples — Western, Educated, Industrialised, Rich, Democratic. That's a limitation of the field, and we flag it.
The Library includes research from Latin America, Africa, Asia, and the Global South — and where findings diverge from the Northern baseline, the scoring reflects that. Expanding the Library southward is a core objective.
What does the ethical framework mean in practice?
Climate misinformation has real consequences: undermining trust, delaying action, harming vulnerable communities. The six ethical principles are constraints the tool operates within and flags when a message risks crossing them. They reflect a commitment to AI that is transparent, accountable, and centered on human agency. Human voices and stories cannot be scripted from outside. Understories applies this by offering narrative frameworks as scaffolding, not templates to copy.
What is the environmental impact of such a tool?
Running AI inference has an energy cost. Every message generates a computational load: servers, cooling, electricity. We're working to quantify it accurately, which, now in May 2026, is harder than it sounds: the per-query carbon footprint varies significantly depending on the model, the provider, and the data centre's energy mix.
The research pilot is deliberately very light: the initial evaluation agent runs under 2,000 tokens. Understories is intentionally built with a cap to prevent runaway computation, and as a RAG system, its efficiency improves with use. A rough estimate at scale — 10,000 users, 30 messages a month — puts the total cost at the equivalent of two round-trip flights from Paris to Tokyo. A full pilot will eventually be heavier, so to ensure efficiency and prevent the tool from being a content generator without any cap, users’ parameters would need to be reduced. An AI tool for climate communication is justified only if it produces better climate communication at scale and has a positive net impact. We take that seriously and will publish an environmental account as part of the annual transparency report if the tool develops.
Is Understories free? Who funds it?
Understories is at research stage, being built between classes in February and March 2026, following an MIT capstone project. Messages are currently evaluated manually, not automatically. The tool is free, following a design principle: the science of climate communication should not be available only to those who can afford a research team.
It runs on free tools and is self-funded at a very low financial cost but with higher time and energy expense. Commitment and autodidactism made it possible. If, and as Understories grows, given what looks like a real interest from practitioners, questions around technology, community, and funding may follow. Users are already raising them: embedding, tailoring, and licensing the tool are possible paths. For now, all is work in progress and built in public. If you want to support open-source climate communication infrastructure and women in AI and tech, get in touch: caroline@understories.io
Is it possible to have the evaluation straight on and make the tool automatic?
Yes, that's one of the directions. The pilot requires a manual process. The tool’s natural next stage is automation, in something closer to real time, directly on the website, or embedded in the workflows advocates already use through an API: content management systems, email platforms, social scheduling tools, and ambient, automatic evaluation is where this goes. The pilot is about listening to users, practitioners, and researchers before building the infrastructure. Users and non-users can raise questions and offer suggestions by email to caroline@understories.io
Can Understories integrate with our organization’s existing tools and strategies?
This is one of the most interesting things Understories could evolve into. The tool currently applies a universal framework: the same eight criteria, research, and ethics for everyone. But climate communication doesn’t happen in a universal context. An organisation has existing positions, a tone of voice, a specific audience, and data it’s spent years understanding, as well as strategic commitments it can’t compromise on.
We’re exploring how organisations can layer their own context on top of the framework: uploading your theory of change, audience segmentation, messaging guide, or red lines. Understories would then aim to evaluate your message against both the research baseline and your organisational reality. This is still in the research phase; users and partners help shape what’s next. If your organisation wants to help build this, get in touch: caroline@understories.io.
Last update: 10 July 2026
