7 karmaJoined Jan 2020


Vladimir Arutyunov

Robotics engineer at NASA JPL

New here




Hi James, this is great, we should be in touch! Bryce Rogers (at highimpactengineers.org) and I are working on cross-cause-area technology-focussed mapping project and biosecurity is one of the areas we're starting with for our prototype. Some comments below mention how it would be useful to link/attach information to different nodes in the map and that's precisely what we're doing. I believe Bryce emailed you and I'm CC'd. I'd love to get in touch and get your thoughts on the matter.

Really happy to see this starting and eager to learn how I can help. Looking forward to talking soon at EAG.

I resonate with your pitfall statement. Although I haven't been actively exploring the non-SW engineering space in EA for very long, it does seem that there is a scarcity of technical information publicly available about where and what the specific needs are from a ME, EE, CE, MatSci, robotics, systems, manufacturing, or other non-SW perspective on a variety of problems.

For example, take the biosecurity cause area needing more materials science expertise for better PPE. Intuitively there is a core MatSci aspect to this most people can easily recognize, but there are 1) probably a lot more types of challenges you would discover once you dig into the details of the problem or begin basic conceptual synthesis, and 2) even the MatSci-specific aspects of this problem are hard to find without actually just digging into the decades-long literature on the general topic of PPE. And I think this might be generally true across domains, though I haven't tried looking exhaustively yet.

Having dashboard-like visibility (whether it's a literal SW-enable dashboard or not) into the technical problem landscape across disciplines and across cause areas sounds like the holy grail to me, which is obviously a lot to ask for at the moment, but that's the general direction I personally hope for.

Kudos to you all for starting this.

An EA Space Agency

Effective Altruism, Space Governance

Let’s build an organization which formulates and implements space programs, missions, and systems, which are aimed at the highest-priority things that humanity can be doing in space. There is currently no space organization, public or private, which formulates and implements programs and missions aligned solely with doing the most good, in an impartial and longtermist sense. There are many organizations which do some or much good, such as NASA, ESA, SpaceX, and others, but there is no example today which spends its budget on a portfolio of projects all aimed at doing the most good for humanity, most effectively, in an impartial and broad sense, such as how many EA grantmaking or evaluating organizations operate today. That should change. Costs for spacecraft development and launches to space are about to drop by so much that it might drop enough for an independent organization to formulate and implement its own missions aimed solely at the highest-priority actions for which space access is required. Traditionally, costs have been prohibitive such that only nations or very large organizations could afford access. There is reason to think this is about to change within this decade.

Costs for launching payloads to space are about to (~5 year timeframe) drop dramatically relative to the status quo (1/100x or less) because of SpaceX Starship, and its eventual competitors. Underdiscussed and underappreciated in this topic area is that payload development costs, not just launch costs, will also drop dramatically as a direct result of the combination of lowered launch costs and the increase in absolute payload capacity. The scale of that second cost drop could be comparably large, in my view, meaning a total mission cost reduction on the order of 1/10,000x could be on the table. Payload capacity to Earth orbit will 5x (100T to LEO on Starship vs 20T typical on many platforms). Payload capacity (to LEO, in this example, but to anywhere, generally speaking) is a variable which many key factors are highly sensitive to in aerospace systems. Technical factors are directly sensitive to this, of course, but programmatic factors, the main cost drivers, are also highly sensitive to this. A 5x increase (20T to LEO vs 100T to LEO) is not merely a multiplier on top of cost, simply lowering the cost per kg further by a factor of 5. It’s not just more kg’s in the denominator with a fixed cost numerator. Rather, when you can actually fly a payload 2x or 5x the mass you normally would, for a fraction of the launch cost, the nature of the engineering problem undergoes a step change. The kinds of engineering, the types of resources, the timeframe, and the scope and scale of the organization required to pull it off all change. The bar is lowered. For tens of millions per mission, which could be tens of millions per year or just a few million per year over a few years, an organization could develop and launch missions to close the gaps in space opportunities that can be used to reduce existential and catastrophic risk.

While NASA continues to develop and fund many climate science and near-Earth object detection projects (even one deflection demonstration), the overall budget leaves much to be desired from an effective altruist viewpoint, in my estimation. As a U.S. agency subject to U.S. congressional budgets… do I need to say more? I fully support aligning NASA more with EA over time, but that is a massive ship to steer, and I don’t necessarily support it more than creating a new space organization which is aligned from the ground up, as hard as that could be. While SpaceX is laser-focussed, dedicated, and actually serious about colonizing Mars, which I fully support, becoming multi-planetary, as necessary as it is, is not the only space-based EA near-term idea we should be able to come up with - far from it. Some things like climate science targeted towards extreme climate change scenarios, better/more supervolcano monitoring from orbit, and direct observation of nearby exoplanet systems for signs of alien megaprojects using dedicated megatelescopes, come to mind but I think part of the reason the ideas in this area are limited might be because we haven’t been considering having our own space organization as an option. Further work should go into understanding the cost-effectiveness of this line of thinking, as well as the actual funding model for such an organization. Do you go fully-philanthropic (is that feasible?), or do you start with philanthropic seed funding to create a revenue-generating for-profit or non-profit entity? These questions bear greatly on the range of possibilities and degree of prioritization of this idea. What might we think of to be able to do or learn by doing it in space, if we believed we were able to?

[opinions are my own, not those of my employer]