I’m the Secretary of Effective Altruism Australia and founder of Effective Altruism Australia: Environment.
After 15~ years in the Australian national security community, including in the Prime Minister’s Department and as a senior adviser to the Home Affairs Minister, I now work full-time advocating for longtermist policies in Australia.
Greg S
Karma: 226
Join the Founding Team: Australia’s AI Safety Institute
Thanks for the post @Luke Freeman 🔸 .
Maybe this is a mildly spicy take, but I think as AI risks become more concrete and nearer term, our community should transition a portion of our time from “scouting” to “soldiering”.
Scout mindset is great for finding new problems and new solutions. But once we have an idea of the problem and ideas for specific solutions, we need to execute the grind to make things happen in the messy real world of people and politics.
This is a great example of how a small bit of “soldiering” by our community helps achieve practical impact.
If you’re interested in this event, but not in Brisbane—we’ll be doing similar events around Australia. Sign up to the Good Ancestors’ newsletter to receive and email when the details are available, otherwise keep an eye on your local group.
Newsletter — The Good Ancestors Project
Thanks for this comment. I couldn’t agree more.
As you say, the overwhelming perspective in Australia up until a few months ago was in favour of acceleration for economic gain and somewhat interested in near-term ethical concerns. It’s probably only a marginal overstatement to say that the political environment was actively hostile to AI safety considerations.
There was a real risk that Australia was going to ‘drag the chain’ on any international AI safety work. Anyone who has been involved in international negotiations will know how easily one or two countries can damage consensus and slow progress.It’s obviously hard to calculate the counterfactual, but it seems certain that the work in the community you refer to has helped normalise AI safety considerations in the minds of senior decision-makers, and so there’s a real possibility that that work shaped how Australia acted at the Summit.
Introducing the Effective Altruism Australia Environment Fund (EAAE)
Thanks so much for the summary Zan. The letter has attracted a really good spread of AI expertise in Australia, has given us a vehicle to talk to other experts and government advisors less focused on safety issues. The letter is also attracting a reasonable amount of media attention this morning.
It’s hard to overstate how backwards the Australian government’s leadership is on AI safety concerns at this point in time. If things continue as they are, it’s essentially certain that the Australian government is going to be a skeptical voice in any multilateral negotiations relating to global agreements and standards setting etc. Given Australia’s geopolitical position, it would meaningfully harm global efforts if Australia is pulling in the wrong direction.
I’m really hopeful that this effort will have a meaningful impact in Australia correcting course. This is a great start, but it will require sustained effort.
Thanks Max—I’m glad this is a hot research topic.
At Good Ancestors Policy, we have begun advocating for the adoption in Australia of various pandemic prevention and mitigation approaches. The residual uncertainty (specifically that we don’t have enough evidence to confidently advise on how that risk-benefit calculus should be assessed in different contexts) makes it very difficult currently to advocate for anything specific relating to far-UVC.
My hunch is that government-directed advocacy for far-UVC is only likely to be successful if we can say “this technology has significant benefits during a pandemic, but provides meaningful ongoing benefit from reducing ‘colds and flus’ even when there isn’t a pandemic”. That is, if the pitch is instead “install this technology, turn it on if a certain risk threshold is crossed, and the cost-benefit works out because pandemics are super bad” governments might be unlikely to bite even if that cost-benefit assessment is robust.
Will keep following this closely!
I think the ‘good vibes’ that help policy advocacy come (in part) from benefiting from other people’s positive externalities. That’s to say, I’d like us to be in a position where we can say “we’re the movement that achieved X, Y and Z. So when we ask your nation to (put 5% of its aid budget to bed nets) you should take us seriously”.
To the extent that we’re more centralised and coordinated, it’s easier to say “we’re the movement that achieved X, Y and Z”. When we intentionally fracture and distance ourselves, we also fracture and distance ourselves from those positive externalities.
Again, I recognise that there is a give and take here—risks and opportunities. I just think that we need to put this potential path to policy impact as an ‘opportunity’ that we largely pass-up when we choose to distance our work and organisations from other another.
I think I mostly agree with what you’re saying. Perhaps the difference at the margins is that if 10 organisations get 10 grants for 10 researchers for 6 months, what they’ll each do is find a person for a short-term contract. If someone leaves or gets sick etc, a stream of work might not be completed and an organisation will be in a tight spot.
If that was instead one organisation, maybe it brings on 5-6 researchers for a year (or ongoing), makes some assumptions about staff turnover and part-time arrangements, and when someone leaves or gets sick it’s not a big deal because work can be reassigned between the team and project timelines can be shuffled (surge people onto the more urgent work). Basically bigger numbers (dollars; staff; projects) gives managers more wiggle room to find ways to make things work.
But I agree with you, that if funders were hard over on smaller organisations, there are ways they could ease the employment model concerns in other ways. I just think the ideal situation from an employment and management perspective would be longer term and more centralised.
I’m not sure that’s true for two reasons:
First, there’s a lot of niche special interest groups that get their way with government. There’s lots of ways to pressuring government for a policy outcome that aren’t a simple popular appeal.
Second, I don’t think it’s impossible to build public support for many things we believe in. I think a message like “it’s unacceptable that in current year there are people who need something as simple as a net, but don’t have access to it—government could fix this today” could easily have popular appeal. Or at least sufficient popular that an alliance of governments would put a few per cent of their aid budgets on the problem and fix it. I agree that not everything we care about could work this way, but many things could.
A factor in favour of a more coherent EA that this post misses is the importance of policy advocacy.
I think in almost all of the spheres we care about governments hold most of the levers. For instance, it would be within the power of many countries to unilaterally solve the global insecticide-treated bed net problem if they were sufficiently motivated. I could make this a very long list.
As a public servant and ministerial adviser, I’ve been on the receiving end of well-coordinated campaigns by global not-for-profits. They’re extraordinal good at getting their way. The example I usually think of is Save the Children. It’s hard to know for sure, but StC seems to have fewer people involved than EA and less money. But they have significant access to the leaders of dozens of countries; the ability to drive multilateral agreements through international decision-making bodies (see the UN Declaration on the Rights of the Child); and genuine geostrategic influence.
The EA movement has all the ingredients (global reach; motivated talent; money) necessary to have influence of that kind, or more. But we chose not to (for many of the reasons outlined above), and I think our impact suffers hugely because of that choice. I think we’ve made a bad deal. I would much rather we paid the price of coordination, managed the risks that result, and used it to be serious players in global policy.
As a manager of a small “EA organisation” I strongly agree with this post.
To expand slightly on the points you made under the flexibility headings, I focus on two things more significant organisations can offer that smaller ones can’t:Better support to employees, particularly in fostering diversity. To oversimplify, a small organisation with the funding models and employment models that follow typically can’t offer the proper range of supports (e.g. parental leave, a degree of employment stability, internal mobility for career growth) that a diverse workforce needs. This directly causes many of the diversity challenges EA faces and harms the movement.
Linked to the previous point—talent identification and management. I know the people in my field and I know the people in my area. I know the other institutions where the talent is currently employed. I can find value-aligned talent and pull them into my organisation / our sphere. But the model needs to be legible to them. When I explain the bizarre series of runways and grants that my organisation is funded by, and the limits and dependencies that follow, that is a massive turn-off to people. And fair enough. Going back to point 1, risky employment structures are only tolerable for a subset of people.
There’s also a point about weight and influence. We greatly reduce our policy influence by fracturing ourselves so much. We are a significant movement that ought to have a weight in the policy conversation equal to our significance. But we present ourselves poorly for that purpose and are weaker as a result. I understand there are risks we manage through the current approach, but I think we are very much on the wrong point of the spectrum.
In my specific case, I would much prefer a world where funders who wanted activities to occur in Australia were able to pool money into one or a small number of top-level organisations and communicate to the board their top-down view of prioritisation and the degree of autonomy + trust they are or aren’t happy with through funding agreements. And then the organisation can do fundraising and community building and policy development and advocacy and organising conferences and coordinating volunteers and all the other things. Presumably, it would also be easier for donors to interface with one or a few organisations, rather than dozens.
I agree with the bulk of this post, but disagree (or have misunderstood) in two specific ways:
First, I think it is essential that EA is a “practical project”, not merely (or primarily) an intellectual project. Peter Singer shared an intellectual insight linked to practical action, and the EA community used that insight to shape our thinking on impactful interventions and to make a practical difference in the world. It’s the link to practical action that, for me, is the most important part of what we do. I think EA is, and should be, fundamentally a practical project. Of course, we should do sufficient intellectual work to inform that practical project. But, if we lose sight of the practical project, I don’t think it’s “effective” or “altruism”.
Second, I think the trend of “disassociating” is typically unhelpful. You note elsewhere in your post that brand damage risks negative feedback loops (and hence a reduction in tractability) and that there is a chance that the community tears itself apart in the face of setbacks. I think disassociating fuels those fires. Personally, I plan to keep associating the EA brand (appropriately) with the good work that I do, because I want everyone in this community to benefit from the positive externality of my work (as humble as that might be). It would be nice if others did the same. That’s how we rebuild and come out of this stronger. If everyone doing good work disassociates, that’s the definition of the community tearing itself apart.
On the balance of things we agree on (and my own priors separate to recent events), I think EA becoming more centralised is right. I agree that we need to focus on good governance and integrity. Doing that effectively requires a sufficient centre-of-mass. The current situation with very many dozens (hundreds?) of small organisations and grant-funded individuals largely doing their own things is not well suited for good governance and high integrity. I think funders should strongly update to provide funding to established organisations (geographically based or subject matter based) to achieve particular outcomes, rather than providing grants to individuals. Empowering peak bodies on issues or in locations lets them have the weight and expertise necessary to drive integrity and enforce norms. But that’s verging on a different conversation.
I strongly agree that as we mature we should push towards a more centralised model. Two specific reasons which expand on your post:
First, the ability to influence policy is essential to many cause areas. Because our efforts are so decentralised, we aren’t building “brand capital” with policymakers as effectively as possible (or even commensurately with our value). In my experience, organisations like Save The Children can easily get meetings with senior public servants and Ministers because they are smart about managing their global brand. Because we don’t harness the positive externalities of one another’s good work, we don’t have anything like the heft that we could. (I appreciate that there’s risk here as well, but I think it’s easier to manage risk in a more mature organisation. I think a lot of recent errors could have been addressed by more mature organisational structures, rather than being caused by them.)
Second, you mention the importance of professional HR operations and health insurance. That shortcoming is directly connected to the DEI issues our community is facing. The decentralised model is fine if you’ve got a relatively privileged background and the right connections. But a more centralised model is essential to be more inclusive and offer people things like parental leave and a safety net. More centralised models could also have better recruitment practices including a greater ability to find diverse talent in their field and their community – something that solo grant applications are not well suited for.
There’s a point for funders here. If you’re minded to fund certain work in a certain cause area – you might do more good by approaching established organisations that already have a “back end” and offering to fund them to find more talent and expand in certain directions (rather than giving a new grant and spinning off yet another person or small organisation).
Other comments joke about mergers and acquisitions. But in seriousness, my vote on boards of the EA-related organisations I’m on has been, and will continue to be, in favour of sensible centralisation.
TL;DR of the below post is that I agree with the brief remarks about 1.5 Track Dialogues made in the Founders Pledge document you cite at f/n 68.
I’d like to cheerlead briefly for a diversity of channels. That is, sometimes we’ll see countries put a public freeze on one another where the most senior and high-profile figures (presidents, prime ministers, foreign affairs ministers etc) don’t talk to each other for reasons of posturing over an issue.
In countries where there isn’t a diversity of channels (by which I mean, most interactions occur between those most senior officials and a formal diplomatic channel), this can create risky situations because there’s no longer a way to clarify (that is, communications become indirect via the oblique public statements, and prone to cross-cultural and other confusion). “Hotlines” are less relevant in this situation because the point of the posturing is that the countries aren’t talking to one another. Picking up the hotline would be off-message.
What reduces risk in that situation is a diversity of channels. The post discussed diplomatic channels, and I won’t repeat that. We also often think about a ‘back channel’ in the sense of a confidant of one leader talking to a confidant of another leader—a proxy conversation that allows the posturing to continue but some more direct communication to occur. And that can help (appreciating the clarity and timeliness points made in your post).
The key thing I’d add to that point (in support of the points raised about track 2) is at-level connections within a bureaucracy (and to a lesser extent people-to-people connections). That is, where lots of officials in a country know their counterpart in the other country, a formal diplomatic freeze is of much less practical concern because the bulk of all those at-level communications means each country remains pretty much in tune with what the other is doing (there isn’t likely to be a spiral of miscommunication because, when the president says X in a public forum, mid-level officials will be saying to other mid-level officials “oh, when the President said X what s/he really meant was 1, 2, 3 not a, b,c ”). Obviously that’s helpful outside of the ‘freeze’ scenario as well.
One of the reasons I think this perspective is relevant to EAs is because it is tractable. A philanthropist might struggle to get a “hotline” built between countries X and Y. However, it is explicitly open to an EA NGO to establish a 1.5 track dialogue (or similar concept) that seeks to get together sets of officials in a way that’s focused on building relationships and interpersonal connections (I don’t even think it needs to be issue-specific in the way this post suggests).
For the benefit of readers, the PF document in the f/n says:
[W]e recommend Track II and Track 1.5 diplomacy programs. This intervention stands out because it has (i) a strong theoretical case for effectiveness, (ii) some supportive empirical evidence and support from experts, (iii) seemingly high upside and minimal downside risk.
Thanks for this post. I’ll make a short comment here (noting that there’s more comments in the other forum) from the perspective of someone who has spent 10+ years in middle management positions in large (1000+ person) organisations.
I’ve seen things like many of the above anecdotes play out, so I mostly agree that these things are all possible, a problem, and worth mitigating. But I’d make two defences of middle management:
There’s also a failure mode where senior managers think they can do both senior management and middle management. In a large organisation, the CEO investing time in understanding the widget manufacturing processes is a poor investment of their time. Lots of senior managers are drawn into that trap. (Particularly if they started out as a widget maker and therefore have specific views. Often views that are now outdated for reasons they’ve missed since they were last widget-making.) So I wouldn’t want any current or prospective senior manager to read this post and imagine they can somehow lead a large organisation without middle managers.
Middle managers do necessary work. The example company that makes widget-1 in small quantities in a single location likely doesn’t need middle management. But once the company is making widgets 1 to 100, there are all kinds of failure modes that develop at that working level. The people making widget-1 will come up with what they see as an efficiency, but actually it’s a problem for the integration of widget-1 with widget-68 in a way that the widget-1 makers don’t understand.
The people that make widget-5 want to make it blue because it’s easier to make. But the marketing people want widget-5 to be red because it will increase sales. And the logistics people think the key change to widget-5 would be making it smaller so it stacks better. Settling that dispute can’t be done by any of those working-level teams (the person running the widget machine shouldn’t be spending time synthesising logistics reports), and it’s below the pay grade of the CEO (their job is to say ‘balance those three factors in the best way’). So you need a process where teams of middle managers can sit down and compare those things. And the outcome of that process is very likely two working-level teams thinking ‘these middle managers have no clue what they’re talking about’. That’s just how the world goes.
It’s no accident that there have been middle managers since the dawn of civilisation. It’s worth thinking about how the CEO can keep a good understanding of their organisation with sensible time investment. And it’s worth thinking about how to ensure widget makers can see the overall strategic vision with sensible time investment. It’s worth thinking about how to ensure middle managers resolve disputes with evidence and reason rather than internal game-playing. I don’t think it’s worth thinking about how to get rid of middle management. I think it would be harmful for a CEO to hold off on hiring managers when their organisation starts to need them.
Which part of 11 USC 548 do you think applies?
It doesn’t seem to me the payments were made to hinder / defraud (I guess that’s a stage-of-mind point I don’t have information on) ; the payments weren’t undervalued ; FTX was solvent when (at least some of) the payments were made.
Thanks for the reply Sean.
I guess if I was in your position I’d consider running a pilot of the service (using as-similar-a-method-as-possible) in a demographic that is seen as desirable to the health industrial complex (New York or California or something). Perhaps I’d also talk to an intellectual property lawyer and see if there are any elements of your method that can be subject to protections (patents, copyright material or whatever).
If it’s so impactful and cost-effective it might be immediately profitable if offered to other demographics (including because market-provided mental health services seem hugely expensive in many western countries—it’s possible that this model could undercut the market dramatically).
Thanks for this post.
Regarding your point that F/P doesn’t have any potential defences, I would have thought that there’s some chance F/P could argue:
1) Only a certain amount of money was received from FTX-proper within the 90 clawback window (impossible for us outsiders to know how much that would impact anything, if at all, but it could reduce the quantum from ‘anything F/P has ever done’ to ‘the recent activity of F/P’), and
2) Notwithstanding the above, that the new value defense applies. That is, F/P was given money to do a particular thing (make grants) and that it discharged its obligation to do that thing (it made the grants).
Any views on that?
(Usual caveat of not being a US bankruptcy lawyer)
Thanks for this link. This made me laugh:
“Everyone Should be a Mapmaker and Fear that Using the Map to Actually Do Something Could Make Them a Worse Mapmaker” would be a much less rousing title, but this is how many EAs and rationalists have chosen to interpret the book.