Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


How to help the world

How different empirical and value judgments can lead you to different focus areas for making the world a better place

Daniel Kokotajlo

on 24 August 2015

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of How to help the world

More details?
How to help the world
This flowchart is designed to start conversations, not end them. There are many simplifying assumptions, but we hope it can help you clarify your own reasoning and understand that of others.
We think identifying a cause to support is hard and complex. There are many valuable ones, and anyone deciding may find there are several front-runners. Moreover the factors which affect the decision can't be perfectly compressed to a flow-chart, so we have made some simplifications. Nonetheless, we hope this chart may help readers to clarify their thinking and perhaps make better judgements when giving time or money.
Where numbers appear in this chart they are based on the most robust estimates we found, but we don't think you should trust them blindly. If your decisions depend sensitively on them, we encourage you to research them further.
More detail on the content of this flowchart, with links to relevant articles and a discussion of the structure and weaknesses, will be available in a forthcoming online version.
OK, got it.
I want to do the most good I can!
OK, I have a cause. How best to achieve it?
almost any
cause, there are four main ways for our time or money to work
towards it:
Direct Work
Movement Building
Researching New Solutions
Political or Social Activism
OK with indirect methods?
We should be very skeptical about indirect methods.
Maybe indirect methods are best for this cause.
Young movement?
Is the movement supporting your cause well-established, or is it still growing quickly?
Need to establish legitimacy?
Does your movement need to prove that it can do good work?
Movement Building
Promising leads for new solutions?
Sometimes new technology or new methods can make problems much easier. For example, perhaps a disease can be cured instead of treated.
No chance of a magic bullet.
Research might well pay off.
Legal or social solution feasible?
Achieving large-scale legal or social change is difficult, but often it has the potential to completely solve a problem.

Moreover, sometimes partial solutions can be achieved relatively easily this way by finding common ground.
Researching New Solutions
Political or Social Activism
Direct Work
Building the movement, raising awareness, etc. can be an incredibly effective way to do good. Attracting more volunteers and donors means more resources for the cause, some of which can be reinvested to grow the movement further.
In many cases, researching new solutions is the best option, because it usually meets lighter opposition, and because the potential payoff for success is huge: whereas it usually takes a massive amount of direct work to completely solve a problem, often it takes only a relatively small amount of research and development.
For a variety of reasons, sometimes the best thing to do is to try to solve the problem directly. For example, maybe the problem is especially urgent.

Usually you won't be able to solve the whole problem this way, but you can be very confident that you are solving part of it. Not only does direct work avoid some of the risks of the other methods, in some cases it is important for the legitimacy of the movement.
Often problems can be solved or mostly solved by appropriately designed legislation. For issues that are not yet politicized, this can be remarkably effective: For the cost of some lobbying and/or political ads, you can convince politicians and/or people to vote for a complete or mostly complete solution! Even for issues that are politicized, sometimes this is the only viable way.
When do you want to help?
Long Run
Should we restrict our attention to the next few decades, perhaps due to an obligation to our generation, or because we can't reliably affect the future? Or is it worth planning for the very long-term?

We should focus on the next few decades.
Future people are just as valuable, and we can help them.
Specific responsibility?
Do you have an overriding obligation to help a particular group or cause?

More generations better?
People probably won't be around forever. All other things being equal, is it significantly worse for extinction to happen sooner rather than later?
Dying is bad, but never being born isn't.
It would be a disaster to miss out on a long-lasting future.
OK with uncertain projects?
Is it OK to prioritize risky projects if the potential benefits of success are high enough?

For example, would you rather do something that you think has a 1% chance of saving 1,000 lives than save one life for sure?

We should make sure we are making a difference.
We should do whatever has the most
Can we do something about existential risk?

Can we reduce our chance of extinction in the future (e.g. by building institutions for the purpose), even a little bit?

The risks are so remote that we cannot have a meaningful effect on them.
We can do something now to significantly lower the risk in the future.
Focus on measurable goals?

Should we focus on projects where it is feasible to measure our impact and compare it to that of alternative projects?

This division is not sharp, so we suggest you also take a look at the path you don't choose, just in case.
Permanent environmental damage possible?
Is there a chance that something (like runaway climate change or nuclear winter) could make life much harder, not just in the next century, but for thousands of years?
No significant chance of damage that lasting.
The welfare of hundreds of generations is endangered.
Your specific responsibility
Which is more important?
How can we have a greater effect in the long run: by improving institutions; by improving societal morals or values; or by increasing human knowledge?

This is a hard question that depends on many factors. For example: Will values or institutions inevitably drift towards certain outcomes? Will they get "locked in" at some point, perpetuating whatever state they were in at the time? Will certain technologies be crucial to navigating the near future?
Reducing Existential Risk
Preventing Permanent Environmental Damage
Better Values and Morals
Better Institutions
Human Rights
Preventing Catastrophe
Institutions have been crucial in guiding the development of society so far; they will probably continue to do so.

We might improve international cooperation to make it easier to regulate dangerous technologies and avoid arms races.
We might improve government transparency and media standards to prevent the creation of a permanent totalitarian regime enabled by new technologies.
We might try to found long-lasting institutions that we think are good, that for whatever reason might never be created in the future.

The values people live by could have a significant impact on our long-run trajectory. For example, perhaps "Value lock-in" will happen in the next century or two: new technology or institutions will prevent values from drifting or changing further. If so, we should make societal values as good as possible before then.

Racism and other forms of discrimination might persist indefinitely due to people not having enough empathy, and there may be something we can do to prevent that.
If people are more altruistic, they may be more willing to collaborate and solve global problems.
Nonhuman beings might end up being given too few rights or too much power unless a moral consensus is reached early.
Since more generations would be better, and since civilization could last for thousands or millions of years, the value of the future is very large. Even a tiny decrease in the probability of extinction has tremendous value.

Improving international cooperation and raising awareness about risk in general to prepare for unforeseen new technologies.
Guarding against threats that might undo this cooperation.
Working to understand and prevent extreme runaway climate change.
Raising awareness about AI risk and researching safety methods.

If you have an overriding obligation towards a particular group or cause, then you should follow that. If you have multiple obligations, or are considering exploring additional cause areas, you might find the remainder of this chart helpful.

Also, the "How to achieve it" section can be a useful resource for almost any cause.

Though civilization can probably recover from anything in ten thousand years, that's still thousands of years to multiply the badness by. Even if there is only a small chance that we can prevent damage like this, it is worth it.

Feedback effects like melting permafrost releasing methane might turn most of the planet into desert.
Depleted natural resources could make rebuilding society after a catastrophe more difficult.
Global catastrophes are likely to be neglected, because many are unprecedented and because their effects are spread thinly across many countries.

Climate change will probably cause massive casualties due to famine, flooding, and refugees.
Epidemologists predict that a natural pandemic will happen eventually and kill a hundred million people. An artificial one might be worse. (http://tinyurl.com/pvdayb6)
The chance of a nuclear war is high enough to cause concern.
Some consider aging a catastrophe: It kills millions every year, and far more money goes towards fighting the symptoms than researching a cure.
Some consider wild animal suffering a catastrophe; many millions of animals are eaten alive or die of starvation daily.

For many people, the world is an unfriendly and unsafe place due to the violent and oppressive behavior of others. For some people, such as slaves, things are even worse. Moreover, solving these problems will likely have beneficial flow-through effects for years to come.

We can fund and advise citizens in totalitarian regimes, giving them the tools to resist and/or escape.
We can work with non-totalitarian governments to legislate against criminal abuses they have thus far ignored.
We can work independently of governments to provide services the legal system is too slow or entrenched to handle.
We can advocate for many kinds of structural change in political and economic systems.

How different empirical and value judgments can lead you to different focus areas for making the world a better place
This chart can help you think about the best strategy for a movement working on a given cause, or for an individual deciding where to donate money.
This only partially addresses the question of how you should donate your time, since that question is more complicated and depends on many factors about your particular situation and skills.

Measurable projects allow us to continually improve our impact.
Hard-to-measure projects are more likely to be neglected by others.
Cause Prioritization
Maybe which cause to support depends a lot on facts about how likely certain catastrophes are, or what the future will be like, or whether or not certain ethical positions are correct. And maybe the cost-effectiveness of different causes varies widely, so that it is very important for us (both as individuals and as a society) to figure out which ones are best.

In that case, we should make cause prioritization itself our cause, at least for now. In a sense, it is a method for working on every cause simultaneously.
More research needed?
Would having more research about the different causes than is currently available lead to significantly better choices?
This is hard. Which is more important?

Expanding Human Knowledge
Research is usually the sort of thing that would be done eventually by someone else, so in that sense it isn't a long-lasting improvement. However, sometimes having improved knowledge just a few years or decades sooner makes all the difference.

Inventing alternative energy sources before oil scarcity escalates international conflict.
Inventing better agricultural methods to counteract the effects of climate change on food production.


Expanding Human Knowledge
Human Rights
Preventing Global Catastrophe
The ethical and empirical assumptions we have discussed so far do not pick out a single focus, because there are now too many relevant factors to put in flowchart form. The magnitude of various problems must be weighed against their tractability and neglectedness, with no single question on which the decision clearly pivots. Here are some suggestions for causes to look into:
Do you think indirect methods--like movement building, political activism, or research--might work for the cause that you are prioritizing?

You might instead think that indirect methods are very risky, or that they take too long, or that there is something unethical about prioritizing them over direct work.
Which cause to prioritize?
How to achieve it?
Even after choosing a cause and method, it is an important and difficult problem to identify the best interventions. This flowchart does not address this question. For some causes, there is excellent work done on it by organizations such as J-PAL, GiveWell, or the Open Philanthropy Project.

The best intervention in an average cause may be better than a typical intervention in the best cause, so research into intervention effectiveness can bear on cause selection. The relative importance of this depends on how much you think cause effectiveness varies.

Even after choosing an intervention, there remains the separate issues of what you personally should do to help. Should you donate your time or your money? How much? Now, or after you have advanced your career? These questions are also beyond the scope of this chart. For information and discussion about them, we recommend career consultancy 80,000 Hours.

As these qualifications make clear, the purpose of this flowchart is not to tell you what you should do with your life. Rather, this flowchart is about cause prioritization, which is a rather high-level choice between different categories of charities and interventions. The main goal is to inspire people to think more carefully about cause prioritization, and to give them resources for doing so. The secondary goal is to aid communication and mutual understanding: Readers can see the reasons why others might disagree with them by looking at the different pathways on the chart.

We welcome feedback on this chart, which is still a work in progress. Contact us at flowchart@prioritisation.org, or visit our website at
Version 1.1
To make this flowchart, we first came up with a list of causes that effective altruists tend to support, and then we tried to come up with questions that would decide between the various causes. Initially we put many questions in, and included a lot of information in each question. Over many days, with feedback and focus groups, we honed down the number of questions by combining or eliminating the less essential ones. We looked to keep only the most robust divisions and challenging decisions. We tried to make it intuitive and comprehensible for non-experts, even though this sometimes came at the cost of precision.
This flowchart is not by any means the final word on the matter. We learned a lot through working on this project, and we expect to learn a lot more in the conversations it will generate. We look forward to incorporating new insights gleaned from feedback into a continuously better product.
Most of the work on this chart was done in summer 2015 by Daniel Kokotajlo and Owen Cotton-Barratt. Many thanks go to the dozens of people who gave feedback on early drafts and participated in focus groups.
Intervention Prioritization
For a given cause, the value of interventions aimed at that cause might vary dramatically. If this is the case, and people aren't already very good at identifying the best ones, it could be more valuable to investigate this further and improve the quality of the choices made than to work directly for the cause.

Such work will additionally tend to increase the ability of people to correctly choose whether to support this cause.
Would more analysis of the various interventions lead to significantly better choices?
How many animals per human?
How many years of chicken suffering would we have to prevent, to do better than preventing one year of human suffering?

We used chickens as a comparison because ~95% of farmed animals are chickens.
Animal pain is just as real as human pain, and just as morally relevant.
Helping humans is better, perhaps because their minds are bigger or because it has additional indirect effects.
Free-range farm worth living?

Is an animal's life on a free-range farm better than never having been born?

Focus on opportunity or health?
Generally speaking, is it better for us to focus on reducing poverty and providing opportunities, or on preventing death and suffering? Or on something else?

We probably missed a few. If you are thinking of any cause or question that should be on this chart, please let us know by contacting us at

Global Health
Millions of people suffer from diseases or vitamin deficiencies that can be easily prevented. Millions more lack access to clean water and proper nutrition.

Essential vitamins and nutrients can be distributed in poor areas, costing $0.30 per person per year, and helping recipients to grow up smarter and healthier. (http://tinyurl.com/o566hbs)
Governments can be lobbied to raise awareness about the economic payoff of healthcare investments.
Research can be conducted to prevent, treat, and cure neglected diseases.

In many countries education is scarce, particularly for girls. Schools are often poorly equipped and understaffed. As a result, hundreds of millions are illiterate and find it extremely difficult to escape poverty.

Low-cost private schools in Lagos cost an average of $35 per pupil per term. http://tinyurl.com/q3p74s4
The annual salary of a teacher in Niger is about $1,500. (http://tinyurl.com/qfsmw6x)
Deworming children allows them to attend school more and leads to better life outcomes later on. (For a discussion of deworming and recent controversy, see http://tinyurl.com/p939e5x)
Reducing Animal Suffering
Although raising animals for food can be done in a humane way, factory farming causes a great deal of unnecessary suffering.

We can boycott or regulate factory farms.
We can boycott or regulate fish farms.
Saving Lives
In developing nations, thousands of people die every day from easily preventable diseases. For an infographic of causes of death worldwide, see http://tinyurl.com/mxy7k7l

Anti-mosquito bednets can be bought and distributed; every $3000 donated is estimated to prevent one child from dying of malaria. (http://tinyurl.com/qyuyqpf)
Research can be conducted on neglected diseases, either for treatment, prevention, or cure.

Preventing Animal Death
The sheer number of animals that die at the hands of humans is staggering.

We can promote vegetarianism or veganism to reduce meat consumption.
We can develop better artificial meat to reduce demand for farmed animals.
Living Standards
About one billion people live on less than $1.25 a day. Given systematic inequalities and current government policies, legislative changes or direct interventions could dramatically improve their quality of life for relatively little cost.

Governments can be lobbied to provide basic services and infrastructure to the people who need it most.
Services and infrastructure can in some cases be provided directly by NGOs supported by private donations.
Even giving money directly to disadvantaged communities, when properly implemented, can result in significant returns on investment. (http://tinyurl.com/plfx9ug)
Fighting Poverty
Providing Education
Standard of living, or education?
Preventing death or suffering?
Double the income of three families for a year
Pay for 10 years of school for 3 children
Prevent one child from dying of malaria
Provide 10 years of vitamins to 1,000 people
According to currently available statistics, there are some charities that can do the following for each $3,500 they receive:
According to currently available statistics, there are charities that can do the following for each $3,500 they receive:
Can we permanently improve society?
Can we make any kind of improvement to society that will make an extremely long-lasting difference?

(Even if our improvement itself doesn't last, it might help society successfully navigate through a difficult period.)
Eventually either the change will be erased with time, or would happen naturally.
We can make improvements that will last a very long time.
Our changes could last long enough to improve our long-run trajectory.
Full transcript