A good summary. Note that marginal per paper cost does not include our overhead, communications, building our network and tools, etc.

We're considering pushing this further and investing in a more bespoke, shareable, and automated platform:

Improving Technology and User Experience: We want to build better tools for scholars, policymakers, journalists, and philanthropists to engage with impactful research and The Unjournal’s work. This includes developing interactive LLM tools that allow users to ask questions about the research and evaluation packages, creating a more interactive and accessible experience. We also want to extend this to a larger database of impactful research, providing tools to aggregate and share users’ insights and discussion questions with researchers and beyond.

This may take the form of research/evaluation conversational notebooks, inspired by tools like NotebookLM and Perplexity. These notebooks would be automatically generated from our content (e.g., at unjournal.pubpub.org) and continuously updated. We envision:

• Publicly shareable notebooks, also enabling users to share their notebook outputs
• One notebook for each evaluation package, as well as notebooks covering everything in a particular area or related to an identified pivotal question.
• A semantic search tool for users to query "what has The Unjournal evaluated?" across our entire database
• Embedded explanations of the evaluation context, including The Unjournal’s goals and approach
• Clear sourcing and transparent display of sources within both our content and basic web data. Academic citation and linking support
• Fine tuning and query engineering to align the explanations with our communication style (and, in particular, to clarifying which points were raised by Unjournal managers versus independent evaluators, versus authors)

We aim beyond this, to 

• Incorporate a wider set of research (e.g., all the work in our prioritized database)
• Leverage users’ queries and conversation (with their permission) to provide feedback to researchers and practitioners on
  • Frequently-asked-questions and requests for clarification, especially those that were not fully resolved
  • Queries and comments suggesting doubts and scope for improvement
  • Ways users are approaching and incorporating the research in their own practice
• … and similarly, to provide feedback to evaluators, and feedback that informs our own (Unjournal) approaches

Ultimately, this tool could become a "killer app" for conveying questions and feedback to researchers to help them improve and extend their work. In the long term, we believe these efforts could contribute to building future automated literature review and evaluation tools, related to the work of Elicit.org, Scite, and Research Rabbit.

We will support open-source, adaptable software. We expect these tools to be useful to other aligned orgs (e.g., to support ‘living literature reviews’). 

Interesting. But I wonder if they (or anyone) has considered this 

... as a source of funding (student fees, government aid if accredited, training subsidies from government and companies) 

rather than merely as an outflow?

Meant to post this in funding diversification week. A potential source of new and consistent funds: EA researchers/orgs could run research training programs.

Drawing some of the rents away from universities and keeping it in the system. These could be non-accredited but focus on publicly demonstrable skills and offer tailored letters of recommendation for a limited number of participants. Could train skills and mentor research particularly relevant to EA orgs and funders.

Students (EA and non-EA) would pay for this. Universities and government training funds could also be unlocked.

(More on this later, I think, I have a whole set of plans/notes).

Also, fwiw, a Perplexity search mainly agreed with you, with some caveats:

likely that a future Trump administration would attempt to undermine or preempt state legislation on animal welfare standards, particularly those related to farm animals.

Perhaps a 'side point', but as bad as the RFK anti-science skepticism is, I think it may actually cut in favor of animal welfare. Unwarranted health fears may combine with warranted considerations to bring concerns over the human risks of large-scale animal agriculture? Bad epistemics but may put this on the map? 

(Note, as I discuss here  I think RFK is a huge risk/disaster on the health and pandemic risk front).

I just commented here on the original post. I mention that we can separate discussion of the issues without discussing the elections themselves, but I think we should not be overly delicate about mentioning links to politicians and parties.

I give some further examples to emphasize the close connection between high-value issues and likely US policies in the coming Trump administration. 

I agree, it's too hard to disentangle these from any political discussion. I guess we could separate discussion of the issues (which we can reasonably link to politicians and parties) but not discuss the elections themselves. I (weak to moderate confidence) think EA and this forum could play a valuable role in bringing good epistemics, reasoning, and expertise to the discussions of the issues. And I think we should not trip over ourselves to refrain from tying these to the politics, where this is obvious. 

In line with your list, I think it's plausible that

1. HEALTH/PANDEMIC RISK: Putting RFK in charge of health, or giving him influence over it will strongly escalate the risk of pandemics and large-scale disease outbreaks in the US and internationally. As I understand it, RFK rejects vaccines and the integrity of the current scientific institutions. He will spread this as a matter of policy as well as through his social influence; both will boost vaccine hesitancy. He will likely drastically reduce US support for vaccine development and promotion, with global implications. 

(Representative evidence: He proposed allocating  half of the NIH's research budget to "preventive, alternative, and holistic" medicine.) 

2. NUCLEAR RISK: The US will fully or substantially abandon NATO and other institutions. This will be destabilizing and raise the risk of adventurism from Russia, China, and others, in turn raising the risk of nuclear war.

Fwiw these claims were broadly backed up by a query on Perplexity.  Full disclosure, Perplexity was less convinced than I am that that Trump will undermine foreign development aid budgets. They had a mixed response on animal welfare issues, but they agreed it is "likely that a future Trump administration would attempt to undermine or preempt state legislation on animal welfare standards, particularly those related to farm animals." 

I need to consider the visibility of the personal blog posts. If they are really ~invisible one possibility could be combining politics with the community section.