Public Priorities for Open Philanthropy Project - Global Catastrophic Risks - September 2015
 Share
The version of the browser you are using is no longer supported. Please upgrade to a supported browser.Dismiss

 
View only
 
 
ABCDEFGHIJKLMNOPQRSTUVWXYZAAABAC
1
Summary (more detail to the right)Likelihood and magnitude of potential catastrophePossible philanthropic interventions (not already funded)
2
CauseStatusGoalLinkMore public content planned at the moment?Overall informal assessment of risk level (relative)Overall informal assessment of philanthropic opportunities to further reduce riskNeed for a full-time specialistHighest-damage scenarioHigh-damage scenarioRelative risk assessment for Level 1 event (as defined at http://blog.givewell.org/2015/08/13/the-long-term-significance-of-reducing-global-catastrophic-risks/#Levels)Relative risk assessment for Level 2 event (as defined at http://blog.givewell.org/2015/08/13/the-long-term-significance-of-reducing-global-catastrophic-risks/#Levels)
3
Biosecurity
Job search for full-time focus area
Make cause-specific hire
http://www.givewell.org/labs/causes/biosecurity
More in-depth writeup forthcoming
High
Relatively strong opportunities
Very high
Synthetic pandemic (possibly optimized to maximize death toll)
Natural pandemic
High: could come from a major natural pandemic (natural pandemics are common, could happen at any time, and are hard to predict), and would be fairly likely to occur if biotech capabilities continued to increase and someone accidentally or intentionally released a pathogen engineered/enhanced to maximize death tolls
High: "highest-damage" scenario is one of the easiest ways to imagine a Level 2 event
Many known (and expensive) preparations (e.g. disease surveillance, development and stockpile of countermeasures, improvement of preparatory plans and response capacity, and efforts (including regulation) to decrease the proliferation and increase the safety of risky emerging technology such as dual use research). Many gaps exist despite substantial government funding; philanthropy could fill gaps directly and/or aim to influence government funding.
4
Potential risks from artificial intelligence
Trial hire for full-time focus area
Medium-depth investigation, trial hire
http://www.givewell.org/labs/causes/ai-risk
Medium writeup forthcoming after investigation complete
High
Highly uncertain; credible opportunities exist but are difficult to evaluate
Medium
Extremely powerful artificial-intelligence agent with misaligned values
Use of extremely powerful artificial intelligence to disrupt geopolitics
High: would likely result from either of the scenarios described, though likelihood of these scenarios is hard to assess
High: "highest-damage" scenario is one of the easiest ways to imagine a Level 2 event
Additional technical research aimed at ensuring the robustness, predictability, and goal-alignment of advanced artificial intelligence systems; research in law, ethics, economics, and policy related to advanced artificial intelligence; and/or education related to such research
5
Geoengineering research & governance
Generalist priority 2 (unlikely to be a priority in the near term)
Infrastructure-building grant(s), preferably major
http://www.givewell.org/labs/causes/geoengineering
NoMedium-high
Possibly very strong opportunities (pending research on robustness of current system)
Medium
Geopolitical challenges from unilateral use of geoengineering or catastrophic effects of climate change
Major climatic impacts of either climate change or geoengineering
Medium-high: scenarios are relatively likely ("high-damage scenario" extremely likely) but Level 1 event would probably not result unless geopolitical instability were driven to extreme levels (e.g. global war) as a result
Medium: risk comes mostly from extreme war scenarios, and even then it's hard to see how these could lead directly to a Level 2 event
Policy-oriented research on how best to govern decisions about research and deployment. Building capacity and developing governance infrastructure. Scientific research on the effects of potential deployment. Very little funding in this space currently.
6
Geomagnetic storms
Generalist priority 2 (unlikely to be a priority in the near term)
Fund research on electrical grid robustness
http://www.givewell.org/labs/causes/geomagnetic-storms
NoMedium-low
Relatively strong opportunities
Very low
Regional (~continent-wide), sustained blackouts
Cascade of failing transformers, expensive and logistically challenging to respond to
Medium-low: would probably require highest-damage scenario (which appears quite unlikely based on our review), and even this would not necessarily be globally destabilizing
Low: even most extreme scenario probably could not cause Level 2 event
Hardening electrical grid; operational mitigation (such as unplugging components in advance of a predicted storm); building spare transformers; research.
7
Other cross-cutting work
Generalist priority 2 (unlikely to be a priority in the near term)
Evaluate specific grant opportunities
-
No (unless grants are made)
(Cross-cutting)TBDLow
N/A (different interventions will apply to different risks)
N/A (different interventions will apply to different risks)
N/A (different interventions will apply to different risks)
N/A (different interventions will apply to different risks)
N/A (different interventions will apply to different risks)
8
Food security
Generalist priority 3 (unlikely to be a priority in the near term)
More cause-level investigation
-No
Medium; highly uncertain and cross-cutting
Highly uncertain
Low
Global collapse of agriculture from extreme climate change, large volcanic eruption, nuclear winter, or other possibilities
Regional collapse of agriculture from extreme climate change, large volcanic eruption, nuclear winter, or other possibilities
Medium-low: scenarios listed are fairly extreme/unlikely
Medium: total collapse of agriculture could lead to Level 2 event. Idea that preparations would matter much in this case requires further conjunctivity.
We have seen a proposed research agenda but so far have not found a consensus that it is promising. More attention to food stockpiling is also a possibility.
9
Risks from nuclear weapons
Generalist priority 3 (unlikely to be a priority in the near term)
Monitoring cause for outstanding opportunities
http://www.givewell.org/labs/causes/nuclear-security
NoMedium-high
Not many suitable remaining opportunities
High
Mass casualties and/or nuclear winter following war between major nuclear powers (U.S. and Russia are the plausible candidates today but proliferation or increased stockpiles could create additional candidates)
Mass casualties and/or "nuclear autumn" following war between moderate nuclear powers (such as contemporary Pakistan and India)
High: would probably require all-out nuclear war, would be quite likely in that case
Medium: risk comes mostly from extreme war scenarios, and even then it's hard to see how these could lead directly to a Level 2 event
Policy development in countries other than the U.S. has little funding but would be very difficult. Some experts say policy advocacy and public engagement in the U.S. are underfunded.
10
Potential environmental risks from engineered microorganisms
Generalist priority 3 (unlikely to be a priority in the near term, though we expect to finish shallow investigation)
Complete shallow investigation
-
Shallow writeup forthcoming
Still investigating specifics, but likely to conclude that risk level is low
Still investigating specifics, but likely to conclude there are few opportunities
Low
Still investigating specifics
Still investigating specifics
Still investigating specifics, but probably will conclude that relative risk is moderate-low
Still investigating specifics
Convenings to analyze potential risks and develop safety guidelines for relevant technologies (genomically recoded organisms, mirror organisms, artificial photosynthesis)
11
Anthropogenic climate change (other than geoengineering)
Generalist priority 4 (unlikely to be a priority in the near term)
Not prioritized
http://www.givewell.org/shallow/climate-change
NoMedium-high
Not many suitable remaining opportunities
High
Geopolitical challenges following disastrous climate change
Major damage to global agriculture and/or refugees due to rising water levels
Medium-high: scenarios are relatively likely ("high-damage scenario" extremely likely) but Level 1 event would probably not result unless geopolitical instability were driven to extreme levels (e.g. global war) as a result
Medium: risk comes mostly from extreme war scenarios, and even then it's hard to see how these could lead directly to a Level 2 event
R&D on clean tech, adaptation preparations, and working toward carbon pricing are all possibilities but all generally highly funded already.
12
Large volcanic eruptions
Generalist priority 4 (unlikely to be a priority in the near term)
Not prioritized
http://www.givewell.org/shallow/volcanoes
NoLow
Not many suitable remaining opportunities
Low but uncertain
Global collapse of agriculture following major eruption
Major eruption cools atmosphere, interferes with travel and agriculture, may increase risk of conflict
Low: scenarios are quantifiably low-probability, especially the "highest-damage scenario" (probably necessary for Level 1 event)
Low: scenarios are quantifiably low-probability, especially the "highest-damage scenario"
Have heard few ideas for interventions other than research or improved prediction. Very little funding currently.
13
Atomically precise manufacturing
Generalist priority 4 (unlikely to be a priority in the near term)
Not prioritized
http://www.givewell.org/labs/causes/atomically-precise-manufacturing
NoMedium-low
We are not aware of many suitable remaining opportunities, but it's possible some would exist if the area received more attention
N/A (not prioritized)
Nano-scale machines self-replicate to an extreme degree, leading to "grey goo" scenario
Weaponized atomically precise manufacturing used in global war
Medium-low: would probably require global war, and in that case the specific contribution of this technology would be questionable
Medium-low: "highest-damage scenario" seems highly unlikely
We have not heard strong ideas for interventions focused specifically on mitigating risks. Highly uncertain how much better proposals could be if the area received more attention.
14
Near-earth asteroids
Generalist priority 5 (unlikely to be a priority in the near term)
Not prioritized
http://www.givewell.org/shallow/asteroid-detection
NoLow
Moderate opportunities
N/A (not prioritized)
Impact from object >10 km in diameter
Impact from object >1km in diameter
Low: scenarios are quantifiably low-probability, especially the "highest-damage scenario" (probably necessary for Level 1 event)
Low: scenarios are quantifiably low-probability, especially the "highest-damage scenario"
Tracking smaller asteroids, along the lines of http://sentinelmission.org/ . Largest asteroids already tracked.
15
Antibiotic resistance
Generalist priority 5 (unlikely to be a priority in the near term)
Not prioritized
http://www.givewell.org/shallow/antibiotic-resistance
NoLow
Relatively strong opportunities
N/A (not prioritized)
Complete failure of antibiotics
Failure of some antibiotics
Low: hard to picture how scenarios would lead to Level 1 event
Low: hard to picture how scenarios would lead to Level 2 event
More R&D on future antibiotics, as well as attempts to prevent antibiotic overuse. There is already a reasonable amount of funding in this area, but could be room for more.
16
Disaster shelters
Generalist priority 5 (unlikely to be a priority in the near term)
Not prioritized
http://www.givewell.org/labs/causes/disaster-shelters
No
Medium-low; highly uncertain and cross-cutting
Not many suitable remaining opportunities
N/A (not prioritized)
Major (synthetic?) pandemic or other situation in which shelters are key
Scenario (such as particular global-war or pandemic scenarios) where shelters could help key decisionmakers survive or put individuals in a much stronger position to recover.
Medium-low: covers many scenarios, but shelters being crucial (especially on the margin) requires a lot of conjunctivity
Medium-low: covers many scenarios, but shelters being crucial (especially on the margin) requires a lot of conjunctivity, and is particularly unlikely regarding potential risks from advanced artificial intelligence
Could build more or improved disaster shelters, but believe we are already at the point of diminishing returns.
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Loading...
 
 
 
Global catastrophic risks