Defective Altruism#

What’s this?

This is summary of Wednesday 14th December’s Data Ethics Club discussion, where we spoke about Defective Altruism, a current affairs article written by Nathan J. Robinson. We welcomed members of Bristol’s Effective Altruism society to this discussion.
The summary was written by Huw Day with the aid of Chat GPT (see disclaimer), synthesising everyone’s contributions to this document and the discussion. “We” = “someone at Data Ethics Club”. Nina Di Cara and Natalie Thurlby helped with the final edit.

Disclaimer#

As an experiment, this writeup was done a little bit differently. Huw made use of the language model Chat GPT to make this writeup. You can read more about what he did in this separate blogpost.

This is still a summary of the discussion had at the original, as the chat GPT system was fed the questions and corresponding notes written during those discussions and instructed to collate these into a blog post summary, much in the way Huw usually does by hand. This was a one off experiment and feedback is encouarged and welcome. Everything you see below this point (not including the list of attendees at the end and the questions) is curated by chat GPT with Huw’s input.

Is it possible to measure how much good a person or action does? And is it a good idea?#

The discussion centered around the question of whether it is possible to measure the impact or good that a person or action does, and whether it is a good idea to do so. One perspective mentioned is that it may not be possible to be precise in measuring the impact or good that a person or action does, but that it may be possible to measure it in broad strokes. Some participants mentioned that even people who disagree with the concept of effective altruism believe that it is possible to measure the impact or good that a person or action does in some ways.

One issue raised in the discussion is that when attempting to measure the impact or good that a person or action does, it often ends up being measured in terms of money. For example, the Arts and Humanities Research Council in the UK has stated that “for every £ invested, we bring back £x to the economy,” which some participants pointed out can add a lot of bureaucracy and can also restrict the proposals that receive funding.

Another perspective mentioned in the discussion is that the existence of charities can be seen as a failure of society, and that measuring the impact or good that charities do may not make sense, especially when it comes to addressing basic needs. It was also noted that the process of measuring the impact or good that a person or action does can be reflexive, and that the question of who is doing the measuring and how many measurements are needed is an important consideration. Some participants argued that the process of measuring the impact or good that a person or action does can be a blinkered feedback loop that is only accessible to those with privilege.

The discussion also addressed the question of whether it is possible to measure the impact or good that a person or action does from a utilitarian perspective, as well as the subjectivity of the concept of benefit and the role of trust and certainty in the process of measuring impact or good. One participant argued that it may be possible to measure the impact or good that a person or action does from a utilitarian perspective, but that it is probably not possible to do so if one is not a utilitarian. Another participant mentioned that benefit is highly subjective and that it is therefore difficult to measure the impact or good that a person or action does.

One example mentioned in the discussion is a critique of utilitarianism by Bernard Williams, in which a person visiting a culture is offered the opportunity to kill one person in order to save nine others. This example was used to illustrate the idea that trust and certainty are important factors that are not often considered in utilitarianism, and that it can be difficult to know for certain that the actions we take will result in the benefits we expect.

The discussion also touched on the distinction between philanthropy and effective altruism. It was noted that while philanthropy is generally focused on giving money to causes that one cares about, effective altruism is more focused on trying to maximize the positive impact or good that a person or action does.

What else do you know about effective altruism? Have you heard of it before? Do you believe that the article is representative of the effective altruism movement or fair in its criticisms?#

The follow-up discussion on effective altruism focused on the idea that each person within the movement is an individual who prioritizes different things, and that global poverty and animal welfare are higher on the list of priorities than AI-related issues. It was also mentioned that there was a mix of previous knowledge about effective altruism among the participants, with some having heard about it before and others having learned about it through the article being discussed.

Overall, it was noted that the people involved in the effective altruism movement are trying to figure out how to do the most good, and that the article being discussed may not present the movement in the most appealing light. It was also mentioned that the article felt biased and was very long, and that it might have been more convincing if it had presented a more balanced and neutral argument.

Some participants in the discussion also pointed out that the article seemed to focus largely on Western philosophers and didn’t incorporate perspectives from other parts of the world, leading to the question of whether effective altruism is primarily a Western philosophy. It was also noted that the article missed the perspective within the movement that it is trying to do good on top of what is already being done in the world, and that there are concerns about diversity within the movement. Finally, the discussion touched on the idea that measuring and quantifying can cause problems and reinforce certain biases, as has been discussed in the context of development effectiveness and cooperation (DEC).

Imagine you were an effective altruist: (how) would you prioritise work on AI? Which dangers or benefits would be the most effective to work on?#

The discussion focused on the question of how effective altruists prioritize their work and how they decide what is “good.” It was mentioned that some people see the process of “recruiting” people to the effective altruism movement as a service in itself, and that there are organizations that make recommendations about which charities should be donated to.

It was also noted that the article being discussed mentions the concept of an “effective altruism pyramid scheme,” which may raise questions about how effective altruists prioritize their work and decide what is “good.”

Closing Thoughts#

The discussion raised a number of issues related to effective altruism. It was noted that the movement is built on the idea that each person is best placed to decide where they can be most effective, rather than having decisions made more centrally. While effective altruism is based on collective rather than individual decisions, this does not mean that effective altruists do not want to make decisions together.

Effective altruism does not include activism for higher taxes because it is voluntary, but this does not preclude people from doing both, giving lots of money to charity and advocating for higher taxes. Some participants pointed out that effective altruism can be undemocratic, as people with more money have more power within the movement. One participant described effective altruism as “a libertarian cult,” noting that the movement is interested in getting other people involved (in an evangelical or pyramid scheme-like way) and that this is not a standard philosophical practice, even though it is based on utilitarian philosophy.

The discussion also addressed the idea that Peter Singer’s original article, which formed the basis for the effective altruism movement, was based on the idea that middle class people in Western countries should give money to people in poorer countries. Some participants argued that this approach has the side effect of allowing people to benefit from high salaries and capitalism while also giving to charity, and that it can entrench the capitalist system and perpetuate inequalities that the movement claims to be trying to fix. There was a question raised about whether there could be a left-wing version of effective altruism that is still based on utilitarianism.

From a utilitarian perspective, one participant argued that the most effective and altruistic thing someone can do is “eat one billionaire.” The discussion also explored why activism, policy, democracy, and government are largely absent from the concept of “effectiveness” in effective altruism. It was noted that one reason for this could be that charity is more international in nature, and that effective altruists do sometimes back certain candidates for political office. However, it was also mentioned that effective altruists have sometimes backed both parties, leading to the question of whether this is a productive use of time and resources.

Finally, the discussion raised the idea that the existence of charities can be seen as a failure of society, and that certain practices within academic philosophy, such as engaging in blinkered feedback loops that ignore lived experience, may enable movements like effective altruism. It was also mentioned that around 2010, there was a trend in Australia towards focusing on narrative as an indicator of success, rather than quantitative key performance indicators, and that this trend has also been observed in the business world in the UK.

Attendees#

In total we had 9 attendees at this meeting. Those who provided their names are listed below. Name, Role, Affiliation, Where to find you, Emoji to describe your day

But is it FAIR?   Chat GPT Stole my Job