Existential Risk Observatory: results and 2022 targets
Introduction
In our introductory post from last August, we introduced our new organization the Existential Risk Observatory. Our aim is to reduce existential risk by informing the public debate. We expect this to lead to a significant increase of:
Talent available to reduce existential risk.
Funding available to reduce existential risk (both private and government budgets).
Institutes trying to reduce existential risk.
Priority for reducing existential risk, even among actors who are in principle aware of this risk already.
Diversity in the existential risk space, which should lead to more robust responses.
We acknowledge information hazards, but we think we can achieve a net positive effect by communicating responsibly about these risks, as established research institutes have done as well. This point is explained in more detail in our introductory post.
We also formulated a target in this post: increase the number of articles in major news outlets in the Netherlands about existential risk by 25%. Now, we are happy to say that we have achieved our target. In this post, we will share our results and experiences last year, what our targets are for the coming year, and why we have chosen these targets. As before, we are open to feedback, and also to share additional information, either on this forum or e.g. by email.
Results and experiences
In 2021, there were seven articles published in national print media in the Netherlands about existential risk and AI without our intervention. We added two articles to this:
One in De Telegraaf (text in Dutch), the largest and oldest newspaper in the Netherlands.
One in Het Parool (text in Dutch), a leading Amsterdam-based newspaper.
We therefore achieved an increase in print media publications about existential risk of 29%, reaching our target.
These newspapers have a combined claimed readership of about 1.2 million, or about 12% of all Dutchmen between 20 and 65. About 540k of this is in print, the rest online. According to research, about 15-20% of readers change their opinion as a result of reading an op-ed. About half of these changes last over time. We therefore seem to have found an effective way to raise existential risk awareness in a general audience at scale, leading to the positive effects outlined above.
In our articles, we have had enough space to communicate the existential risk message in a relatively nuanced way, as opposed to what would be possible on social media. Also, the fact that the articles were published in print media, and one of them with an academic co-author from a locally known university, added credibility and weight to them.
We managed to get our existential risk op-eds published in roughly the following way:
Set up our organization (website, board, legal structure, etc.).
Form alliances with academics.
Hire a freelance PR person with journalistic experience.
Write op-eds with both of the above.
Write to journalists with an interest in the topic.
Submit, repeat if necessary.
We believe others with strong motivation and preferably some management experience, but not necessarily an academic background in existential risk or AI, can do the same in other countries. We are happy to give more advice about this, and you can always approach us by email.
Added to our work in traditional media, we have reached out to around 300k people through sponsored campaigns on social media as well. We have used these campaigns to increase the reach of our print articles, to inform people directly about existential risk, and to research which narrative worked best for this purpose. We tried a number of angles including AI, climate, pandemics, and personal impact. The results can be viewed here.
Additionally, we have organized the first Existential Risk Conference in Europe, with academics from e.g. CSER, FLI, Harvard University, and our local Groningen University, a climate NGO CEO, and a nuclear weapons activist.
Targets for 2022
With these results, we believe that we have proven that we can increase the societal debate significantly with our organization, and therefore we are now planning to scale up in the most impactful countries. According to us, for the Existential Risk Observatory in 2022, these are the Netherlands and the US. The Netherlands mainly because of local traction: we think the probability that we can achieve results, such as a significant increase in existential risk research, in the Netherlands, is for now significantly larger than anywhere else. Also, we can use the lessons we continue to learn in the Netherlands when scaling up internationally.
We chose the US as the first country to expand to, since it has large potential as a source of research funding, has one of the largest AGI research bases, we are relatively familiar with language and media landscape, and articles published here could have a global reach. The US may not be the most neglected, but the amount of articles about (AI) existential risk in print media is still small enough to be able to increase it by a significant percentage in the medium term.
We judged countries to expand to mainly by the criteria:
Necessity (for existential risk awareness to increase), and
Capacity (for us to operate, with a reasonable chance of success).
Our judgment on these criteria of the six countries we considered is summarized below.
| Netherlands | Germany | France | UK | US | China | |
| Necessity | Relatively low | Medium | Medium | Low | High | High |
| Capacity | Relatively high | Medium | Low | Medium | Medium | Low |
Necessity is mainly tied to GDP (since this could translate to research funding, either private or public) and regulatory relevance. Neglectedness and information hazard have also been included, but were not too dominant. Capacity is very organization-dependent. For us, it was determined mostly by in-house language skills, existing network, and familiarity with the local media and societal debate.
It would be great to raise awareness in many other countries, but because of our funding situation and focus, we do need to make hard choices. This underlines, in our opinion, how neglected existential risk communication still is.
At this moment, we are still heavily dependent on the founders for both work and funding. This is a key risk for us. To safeguard the long term stability of the organization, we therefore plan to acquire funding in the coming year. Also, we are at this moment funding-constrained for communication: more funding would enable us to increase awareness in more countries, and increase the chance of achieving positive outcomes.
For 2022, our targets are therefore to:
Publish at least three articles about existential risk in leading media in the Netherlands.
Publish at least three articles about existential risk in leading media in the US.
Receive funding for stability and future upscaling.
Conclusion
We think we have proven that a small organization such as the Existential Risk Observatory can influence a national debate about existential risks. Since we also believe existential risk communication to a general audience is strongly net positive, as outlined in more detail in our introductory post, we think we are creating a powerful tool in existential risk reduction. We are generating and showcasing this cool, but we don’t aim to monopolize it.
We would therefore like the existential risk community to actively seek possibilities to raise general audience risk awareness. Also, we hope the community is willing to help us to succeed, primarily by giving us advice on how to best achieve our goal of increasing awareness. We are hopeful that informing the public debate together can meaningfully help to reduce existential risk!
Enough happened to write a small update about the Existential Risk Observatory.
First, we made progress in our core business: informing the public debate. We have published two more op-eds (in Dutch, one with a co-author from FLI) in a reputable, large newspaper. Our pieces warn against existential risk, especially from AGI, and propose low-hanging fruit type of measures the Dutch government could take to reduce risk (e.g. extra AI safety research).
A change w.r.t. the previous update, is that we see serious, leading journalists become interested in the topic. One leading columnist has already written a column about AI existential risk in a leading newspaper. Another journalist is planning to write a major article about it. This same person proposed having a debate about AI xrisk at the leading debate center, which would be well-positioned to influence yet others, and he proposed to use his network for the purpose. This is definitely not yet a fully-fledged informed societal debate yet, but it does update our expectations in relevant ways:
The idea of op-eds translating into broader media attention is realistic.
That attention is generally constructive, and not derogatory.
Most of the informing takes place in a social, personal context.
From our experience, the process is really to inform leaders of the societal debate, who then inform others. We have for example organized an existential risk drink, where thought leaders, EAs, and journalists could talk to each other, which worked very well. Key figures should hear accurate existential risk information from different sides. Social proof is key. Being honest, sincere, coherent, and trying to receive as well as send, goes a long way, too.
Another update is that we will receive funding from the SFF and are in serious discussions with two other funds. We are really happy that this proves that our approach, reducing existential risk by informing the public debate, has backing in the existential risk community. We are still resource-constrained, but also massively manpower- and management-constrained. Our vision is a world where everyone is informed about existential risk. We cannot achieve this vision alone, but would like other institutes (new and existing) to join us in the communication effort. That we have received funding for informing the societal debate is evidence that others can, too. We are happy to share information about what we are doing and how others could do the same at talks, for example for local EA groups or at events.
Our targets for this year remain the same:
Publish at least three articles about existential risk in leading media in the Netherlands.
Publish at least three articles about existential risk in leading media in the US.
Receive funding for stability and future upscaling.
We will start working on next year’s targets in Q4.
I think that EAs generally haven’t pursued media outreach due to considerations such as covered in this post: What to know before talking with journalists about EA. Their worries seem to be mostly related to journalists misunderstanding or misrepresenting what was said, unfavorable quotes, or stories being fitted into a narrative.
I suppose Op-Eds manage to avoid most of these problems and add a lot of credibility to the field. I guess the main potential downside I can see is that we wouldn’t want existential risk to become a buzzword that people start adding to all kinds of proposals that have nothing to do with x-risk. However, it seems unlikely that just a couple of articles would have this kind of effect. So overall, I think having at least a small amount of this kind of work is important as it does improve the credibility of the field.
Thanks for that context and for your thoughts! We understand the worries that you mention, and as you say, op-eds are a good way to avoid those. Most (>90%) of the other mainstream media articles we’ve seen about existential risk (there’s a few dozen) did not suffer from these issues either, fortunately.
Thanks for the update ! This seems like very interesting work and important work. READI did a large survey with some questions on attitudes towards existential risks as part of our SCRUB survey work. We haven’t been able to write it up yet because of an embargo from our government partner, but I will reach out when we can because it might be an easy source of content.
Thank you for the heads up! We would love to have more information about general audience attitudes towards existential risk, especially related to AI and other novel tech. Particularly interesting for us would be research into which narratives work best. We’ve done some of this ourselves, but it would be interesting to see if our results match others’. So yes please let us know when you have this available!
Hey Otto,
We have 580 responses from international data collected during the SCRUB project (though these responses may not all be complete). We collected them in Waves 5&6. You can see the questionnaires used for each wave in the Word files in the linked folder. ‘GCR_’ are the relevant variables.
We also have 1400 domestic (Australian) responses. These are embargoed by our government partner. We can only get access by requesting the data for a paper (which we plan to do eventually). Let us know if you would be interested in collaborating on that. Let me know if you have any questions.