As more people turn to AI assistants like ChatGPT and Perplexity to answer their questions, an opportunity to go beyond traditional communication efforts like SEO arises. Answer Engine Optimization (AEO) refers to the practice of structuring content so it can be accurately retrieved and cited by large language models (LLMs). For EA-aligned organizations, AEO could be more than a communications tactic, as it could be an opportunity to strengthen program impact, improve fundraising efficiency, and ensure that evidence-based ideas surface in critical public discourse. This post outlines why AEO matters, how it works, and what practical steps organizations can take to begin integrating AEO, even with limited resources.
AI Chatbots Are Changing How People Search
Most effective organizations aim to influence how people think, give, and act based on rigorous research and thoughtful prioritization. But in today’s digital landscape, how people seek information has changed dramatically.
Search engines are no longer the primary entry point for knowledge. In a 2024 report by McKinsey, over 30% of Gen Z respondents reported using generative AI tools as a first stop when researching unfamiliar topics or making personal decisions.¹ Similarly, Pew Research found that young adults increasingly rely on LLMs for advice, summaries, and recommendations, skipping traditional websites altogether.²
If someone types into ChatGPT:
- What are the most effective charities fighting global poverty?
- What does cause prioritization mean?
- How can I reduce animal suffering?
Will your organization’s work show up in the response? Will it be accurately summarized?
AEO Is Like SEO for AI Conversations
Answer Engine Optimization (AEO) is a set of strategies designed to ensure that your organization’s ideas and materials can be retrieved, understood, and accurately represented by LLMs and generative search tools.
But unlike traditional SEO, which focuses on ranking high in search engine results, AEO is about being citable and useful to machines that generate direct answers. This includes tools like:
- ChatGPT
- Claude
- Perplexity
- Arc Search
- Google’s Search Generative Experience (SGE)
These systems often rely on structured content, trusted sources, and clarity of expression. There's an opportunity for organizations to proactively adapt their materials so they are more likely to be referenced, improving both their visibility and perceived credibility.
Note on Related Concepts:
AEO is closely related to a newer academic idea called Generative Engine Optimization (GEO)3. While AEO focuses on helping your content appear in AI-generated answer snippets and search results, GEO looks deeper into how large language models synthesize information from multiple sources to generate complete responses. GEO is mostly discussed in AI research, exploring the technical nuances of influencing model outputs. Both approaches share the goal of making valuable content more accessible and credible in the era of generative AI, but AEO is the more practical framework commonly used by marketers and communicators.
AEO Can Help Effective Organizations Amplify Their Impact
AEO is not just about traffic or brand recognition, but maximizing the reach of ideas that matter. Effective organizations often produce high-value content, but unless that content is machine-readable and query-friendly, it may be invisible to the very people whom we want to reach it most.
Here’s how AEO could support key goals:
1. Communications
- Help ensure accurate representations of complex ideas (e.g., moral circle expansion, long-term ethics)
- Broaden reach to audiences outside the EA bubble
- Position the org as a leader cited by trusted AI tools
2. Fundraising
- Improve discoverability in donor-driven queries like “Where to give for the highest impact?”
- Build implicit trust by appearing in authoritative AI-generated content
- Help cultivate newer, younger donor segments engaging through LLMs
3. Advocacy and Research Translation
- Ensure nuanced policy positions or recommendations (e.g., on AI governance or biosecurity) are surfaced in real-time.
- Prevent low-quality or outdated content from crowding out well-reasoned evidence.
Simple Steps to Start Optimizing for AI
Most of these actions can be implemented by a small team or individual contributor with no need for paid tools.
1. Anticipate and format for answering questions
- Make sure your content is publicly accessible. Avoid login walls or member-only sections for key materials, and whenever possible.
- Provide content in HTML format rather than PDF-only reports, which are harder for AI tools to parse and index.
- Use clear, consistent permalinks for foundational pages such as your theory of change, methodology, or donation recommendations with URLs like /how-we-work or /what-is-effective-giving that are easy for both humans and machines to understand.
- Write or revise evergreen content that answers big questions using clear headers that resemble real user queries:
- What is [concept]?
- Why is [intervention] effective?
- How does [organization] make decisions?
- Include short, summary-style answers below each question, as these are often the snippets AI tools extract.
2. Add structured data to your website
Use schema.org markup (particularly FAQ and HowTo schemas) to help LLMs understand your page structure. Free tools like Yoast or Google’s Structured Data Markup Helper can help non-developers implement this, too.
3. Simplify and clarify language
- Many effective organizations produce excellent but complex material. Large language models are more likely to cite short, clear, and well-structured summaries, especially those written in plain language.
- While English remains the most reliably parsed language for LLMs, the same principle applies across languages: direct phrasing, concise answers, and skimmable sections make your content easier for AI tools to understand and reuse.
- Including summaries, ideally in both English and your local language, can significantly improve visibility and citation.
4. Strengthen presence in public knowledge graphs
Ensure your organization has complete and accurate entries on:
- Wikidata
- Wikipedia
- Google Knowledge Panel
These databases inform many of the LLMs' internal knowledge graphs.
5. Link to credible sources
- The more your content is cited or linked to by high-authority media and organizations, the more likely it is to appear in LLM outputs because AI tools are more likely to consider your site trustworthy.
- Offer guest posts, collaborate on joint content, or contribute to Wikipedia/Wikidata.
6. Monitor your presence and check if your AEO is working
Create a log of responses from different AI tools to key questions (e.g., “best charities for climate change”). Track whether your org appears, and whether the summary is accurate. You could:
Risks to Keep in Mind and Tradeoffs
While AEO could be powerful, it’s not without limitations:
- Unpredictable output: LLMs may summarize your content in ways you didn’t intend.
- Opaque impact: You may not be able to directly trace how users arrive at your donation or content via AI.
- Maintenance required: AEO is not a one-off fix. It will benefit from regular updates and content iteration.
However, neglecting this space may result in missed opportunities and the risk of misinformation prevailing where evidence could have led the way.
Looking Ahead
Generative search and AI assistants are not a passing trend. As public reliance on these tools increases, so too does the need for aligned, accurate, and legible ideas to populate their outputs.
Effective organizations are especially well-positioned to lead in this space, with high-quality content, clear theories of change, and an emphasis on truth-seeking. AEO could make that work easier to find, easier to trust, and ultimately, more impactful!
Call for Collaboration
If you're experimenting with AI visibility, I’d love to connect. Some ideas to explore:
- A shared visibility tracker or dashboard to monitor how effective organizations appear in AI tools.
- A community resource with FAQ-style content for EA cause areas.
- A working group or tutorial series on low-cost AEO implementation for nonprofits.
Let’s help each other bring more thoughtful, evidence-based ideas into the spaces where people might be asking the world’s most important questions.
Sources and Further Reading
¹ McKinsey & Company. The State of AI in 2024: Generative AI’s Breakout Year. 2024. https://www.mckinsey.com/mgi/overview/the-state-of-ai-in-2024
² Pew Research Center. How Americans Use ChatGPT. 2023. https://www.pewresearch.org/internet/2023/12/07/how-americans-use-chatgpt
³ Gao, Y., Liu, Z., Si, C., Meng, Y., Xiong, C., & Lin, J. Generative Engine Optimization: Rethinking Content Creation for LLMs. arXiv preprint (2023). https://arxiv.org/pdf/2311.09735
Adapted in part from:
Additional examples and recommendations by the author.
Other resources:
Luckily those suggestions are all useful for SEO too!
Some other things to consider (from figures like Tyler Cowen[1] and Patrick McKenzie[2] (edit: also Gwern) who talk about how their primary audience is now LLMs):
https://marginalrevolution.com/marginalrevolution/2025/01/should-you-be-writing-for-the-ais.html
Search for "Generative Engine Optimization (GEO)" in this podcast
Thank you so much for this thoughtful comment, Rasool. I appreciate the references to the discussion and podcast, and I’m excited to check them out!
I love how these tips deepen the connection between AEO and SEO, and highlight how writing with LLMs in mind can also improve clarity and usability for human readers. The points on a modular structure, prompt-mirroring headlines, and licensing are especially insightful.
Thanks again for sharing!