Comments
53
In recent months, I’ve been navigating a career change. I’ve shifted from working on global catastrophic risk management with a broad focus to concentrating on artificial intelligence. Although I had some prior involvement in AI regulation, I didn’t think I could make my greatest contribution in that specific area based on my skills and knowledge. So, while thinking about how I could best contribute to the field, I decided to experiment with writing—a skill where, at least in my native language (Spanish), I have a comparative advantage.
Before committing to a specific path, I considered several related options, such as leading communications for an AI-aligned organization, exploring journalism to cover global debates for a regional audience, or building an independent project to address topics I found neglected and important, growing an audience over time. Below are some reflections from this career transition process.
With the idea of testing a writing-oriented path, I attended EAG London and spoke with over twenty people from different fields. This led me to refine my expectations and update several beliefs.
The first change was realizing that writing quality alone is not enough to maximize impact, especially in highly competitive niches or in a second language. For instance, journalism in English demands not only advanced language skills but also strong networks, fast turnaround times, and a high tolerance for public exposure—none of which are my strong suits.
The second insight was that writing rarely reaches its full potential in isolation. It becomes much more effective when combined with other elements, such as solid research, domain expertise, and a clear understanding of which stakeholders need to receive the message and when. More than one person pointed out that even without a wide reach, a piece of writing can have a significant influence if it reaches the right people at the right time.
Finally, I came to see the value of engaging the target audience not just during dissemination but from the research stage onward. This means treating writing as part of an interactive process—a feedback loop—rather than a finished product.
Following these conversations, I ran a small experiment on the Substack of a new organization I'm working with (https://aixo.substack.com/). Over the past month, I published weekly, alternating between two types of content: shorter posts commenting on a recent article or news item, and longer, more research-intensive essays. The short posts take two to three hours, allowing me to maintain a steady flow and react quickly to current events. The longer ones can take twice as long but offer a different kind of value: they help me connect ideas, identify gaps in the discussion, and propose future research avenues.
I’ve found that the process of writing these longer pieces imposes a kind of intellectual discipline. It improves not only the final text but also the quality of my thinking on the topic, even before anyone reads it. This is a key reason for others to consider writing, even in a seemingly crowded field. The first and most important audience for your early drafts is often yourself; the clarity you gain is a benefit in itself.
A preliminary conclusion from this exercise is that writing is not my only skill, nor necessarily my greatest comparative advantage in my current publishing context, but it acts as a multiplier for the skills I do have. My main strength lies in research and analysis, and writing amplifies that work by making it legible, reviewable, and reusable by others.
Even with a small audience, a well-targeted post can serve as a prototype for ideas that can later be developed into a more substantial report, a policy recommendation, or an academic paper. In contexts where attention is scarce, the ability to distill a conceptual framework or key evidence into an accessible format can be more valuable than producing large volumes of scattered content.
I have also found value in replicating and discussing ideas that I believe don't get enough visibility but are relevant, and even in raising new perspectives on fields where I feel I have some authority to speak, such as those at the intersection of advanced AI risk management and social resilience and adaptation.
After this month of experimentation, I’ve decided that I don’t want to pursue writing as an isolated activity. I want to integrate it into a workflow where writing and research reinforce each other. This means:
This approach also helps me prioritize posts with the greatest potential impact relative to the time invested. I've noticed that short news commentary adds little value in an already saturated environment.
I will continue writing on Substack—not as a standalone goal, but as a strategic tool to amplify my research and professional impact. I am interested in the cumulative effect of weekly writing on my career trajectory, beyond the impact of any single post.
If there's interest, I might share concrete metrics and results from this approach in the future, along with examples of how an initial post evolved into a larger project. For now, this first month has shifted my perspective: writing is no longer just about its communicative power, but about serving as a laboratory for ideas that, when used strategically, can act as a catalyst for impact. I encourage others who may be on the fence to consider writing not as a performance, but as a process. It might be one of the most effective tools we have for improving our thinking and creating targeted value.
