grantmaking.ai Launch Round
$8,000
$45,000
ARC aims to strengthen coordination and build the communication layer of the AI safety field through frameworks, trainings, practical content, communication tools, and field activation.
+4
ARC aims to strengthen coordination and build the communication layer of the AI safety field through frameworks, trainings, practical content, communication tools, and field activation.
+4
Project Details
Updated 07/13/26 · Edited by orgHow can communication work help? ARC consists of 5 workstreams: Research, Framework Development, Field Visibility and Coordination, Field Outreach, and Format Innovation.
- Research
Since April, I have conducted initial research to identify coordination and field-building gaps in the field, including speaking to field-building, training, advocacy, policy organizations and private companies, including Arcadia Impact, MATS, SafeAI, CeSIA, Control AI, Pause AI, Cooperative AI Foundation. Apart from interviews done during EAG London, I have also spoken online with organizers, researchers and communicators in the field. Content-wise I have created a on-going survey and conducting a comms landscape mapping.
The research so far has revealed significant coordination, terminology, field-building and visibility gaps in the field. It also reveals the need to educate about risk communication and innovate current content format to better engage the public. These research efforts are far from finished. They will continue to inform better designs and direction for the other workstreams.
- Framework Development
AI safety risk communication is a new field. I have drawn from existing literature from health care, nuclear, finance, environmental and scientific uncertainty and put together a simple playbook that was shared on LinkedIn.
I see the need to develop a framework that is tailored to our field. I want to rally more academic support to research and develop this field.
Communicators need more tools and guidance to better communicate uncertainty and abstract AI safety debates. With this work, I can further develop it into guidelines, trainings and fellowships.
- Field Visibility and Coordination
ARC aims to improve the coordination between technical research, policy, and advocacy by 1) increasing visibility 2) providing training 3) building shared-terminology.
I. Reducing redundant efforts and designing more targeted programs can be achieved when orgs knows what each other is doing. Communication work such as but not limited to an industry newsletter can amplify best practices/organizational models/effective programs.
In the same way, advocacy can better design campaigns and mobilize public support if they know the progress and development of the field.
II. Training communicators to translate research into policy language to create stronger exchange between researchers and policymakers. I am going to help other orgs that are already doing this by drafting course materials and providing support.
I have identified that researchers are often reluctant to speak about their work to non-technical people. By recruiting qualified researchers, and creating communication tools and trainings for researchers, it will help them better communicate their work.
III. A research organization and a private company both revealed to me that teams from policy, red-teaming and research often use different terms for the same thing. There is also a lack of communication between them. By building a shared-language tool it can help reduce friction and increase workflow efficiency.
- Field Outreach
I. AI Safety Journalism Network
To expand the field, more professionals needs to help amplify our progress and break the silo. By developing a PR toolkit, I can recruit tech journalists and PR to pitch regular AI safety coverage in existing technology publications and industry channels.
II. Arts and Cultural sectors
Building on my strong network across the United States, Europe, the Middle East, and Asia, I plan to publish a call-to-action paper that makes the case for greater public engagement with AI safety, and invites artists and curators to develop work around the issue.
- Format Innovation
The field has grown and matured to the point where explainers alone are no longer enough. I plan to use formats such as quizzes and games, both common in science communication, as well as profiles of key figures in the field to broaden public engagement.
Recent communications hackathons have also produced promising project ideas. I plan to help these projects get off the ground not just as a communicator but also as a product developer.
Click here to view the ARC Initiative Framework workstream architecture diagram
Theory of Impact
Updated 07/13/26 · By grantmaking.aiWhen we mention 'communication', most people naturally think of public engagement.
Communication work goes far beyond that.
As a veteran communicator who has worked across journalism, corporate communications and advocacy, the best way for me to contribute is to identify the bottlenecks beneath the surface and help break down silos. My counterfactual case is that, without this work, many of these coordination and translation gaps will remain nobody’s clear responsibility. I plan to reduce x-risk at the infrastructure level by building the communication capacity the field currently lacks.
The 5 workstreams aims to tackle different bottlenecks that I have identified so far. By strengthening each building block in the ecosystem through communication work, it helps the people already working on technical safety, policy and governance communicate more effectively, find one another’s work, and move useful research into practice.
Right now, the field is asking for more people who can translate across research, policy, advocacy and public communication, but that talent pool is still very thin. By developing frameworks, training communicators and improving visibility across the field,
People
Updated 07/13/26 · By grantmaking.aiTeam Member
Funding Asks
Applied
Discussion
Sign in to comment
I met Wun at EAG London and we have been in contact since. She is sharp at analysing what is actually happening in this space and turning that into workable answers, and she brings real communications experience to a field that needs much more of it. Recommended.
I interviewed Wun as an AI safety communications expert for my research and am exchanging thoughts with her ever since. I am deeply impressed not only by her knowledge of what the field is currently lacking, but also by her sharp understanding of effective communication. Her work will be of huge benefit to the evolution of the field. Highly recommended!
AI risk comms is needed more than over. We need folks trained in nuanced thinking about risks to counter clickbaity discourse.
I had a great experience meeting Wun at the 2026 EAG in London, and have been in contact since. I really believe that her long experience in communications could bring great benefit to the AI Safety Comms landscape. The space definitely needs more coordination, and I think this project would be a step forward in that direction. Recommended!