Information security in high-impact areas


As the 2016 US presidential campaign was entering a fractious round of primaries, Hillary Clinton’s campaign chair, John Podesta, opened a disturbing email. The March 19 message warned that his Gmail password had been compromised and that he urgently needed to change it.

The email was a lie. It wasn’t trying to help him protect his account — it was a phishing attack trying to gain illicit access.

Podesta was suspicious, but the campaign’s IT team erroneously wrote the email was “legitimate” and told him to change his password. The IT team provided a safe link for Podesta to use, but it seems he or one of his staffers instead clicked the link in the forged email. That link was used by Russian intelligence hackers known as “Fancy Bear,” and they used their access to leak private campaign emails for public consumption in the final weeks of the 2016 race, embarrassing the Clinton team.

While there are plausibly many critical factors in any close election, it’s possible that the controversy around the leaked emails played a non-trivial role in Clinton’s subsequent loss to Donald Trump. This would mean the failure of the campaign’s security team to prevent the hack — which might have come down to a mere typo — was extraordinarily consequential.

These events vividly illustrate how careers in infosecurity at key organisations have the potential for outsized impact. Ideally, security professionals can develop robust practices that reduce the likelihood that a single slip-up will result in a significant breach.

Continue reading →

#141 – Richard Ngo on large language models, OpenAI, and striving to make the future go well

It does seem like we’re on track towards having many instances of the largest models rolled out. Maybe every person gets a personal assistant. And as those systems get more and more intelligent, the effects that they have on the world increase and increase. And the interactions that they have with the people who are nominally using them become much more complicated.

Maybe it starts to become less clear whether they’re being deceptive and so on… But we don’t really have concrete solutions right now.

Richard Ngo

Large language models like GPT-3, and now ChatGPT, are neural networks trained on a large fraction of all text available on the internet to do one thing: predict the next word in a passage. This simple technique has led to something extraordinary — black boxes able to write TV scripts, explain jokes, produce satirical poetry, answer common factual questions, argue sensibly for political positions, and more. Every month their capabilities grow.

But do they really ‘understand’ what they’re saying, or do they just give the illusion of understanding?

Today’s guest, Richard Ngo, thinks that in the most important sense they understand many things. Richard is a researcher at OpenAI — the company that created ChatGPT — who works to foresee where AI advances are going and develop strategies that will keep these models from ‘acting out’ as they become more powerful, are deployed and ultimately given power in society.

One way to think about ‘understanding’ is as a subjective experience. Whether it feels like something to be a large language model is an important question, but one we currently have no way to answer.

However, as Richard explains, another way to think about ‘understanding’ is as a functional matter. If you really understand an idea, you’re able to use it to reason and draw inferences in new situations. And that kind of understanding is observable and testable.

One experiment conducted by AI researchers suggests that language models have some of this kind of understanding.

If you ask any of these models what city the Eiffel Tower is in and what else you might do on a holiday to visit the Eiffel Tower, they will say Paris and suggest visiting the Palace of Versailles and eating a croissant.

One would be forgiven for wondering whether this might all be accomplished merely by memorising word associations in the text the model has been trained on. To investigate this, the researchers found the part of the model that stored the connection between ‘Eiffel Tower’ and ‘Paris,’ and flipped that connection from ‘Paris’ to ‘Rome.’

If the model just associated some words with one another, you might think that this would lead it to now be mistaken about the location of the Eiffel Tower, but answer other questions correctly. However, this one flip was enough to switch its answers to many other questions as well. Now if you asked it what else you might visit on a trip to the Eiffel Tower, it will suggest visiting the Colosseum and eating pizza, among other changes.

Another piece of evidence comes from the way models are prompted to give responses to questions. Researchers have found that telling models to talk through problems step by step often significantly improves their performance, which suggests that models are doing something useful with that extra “thinking time”.

Richard argues, based on this and other experiments, that language models are developing sophisticated representations of the world which can be manipulated to draw sensible conclusions — maybe not so different from what happens in the human mind. And experiments have found that, as models get more parameters and are trained on more data, these types of capabilities consistently improve.

We might feel reluctant to say a computer understands something the way that we do. But if it walks like a duck and it quacks like a duck, we should consider that maybe we have a duck — or at least something sufficiently close to a duck it doesn’t matter.

In today’s conversation, host Rob Wiblin and Richard discuss the above, as well as:

  • Could speeding up AI development be a bad thing?
  • The balance between excitement and fear when it comes to AI advances
  • Why OpenAI focuses its efforts where it does
  • Common misconceptions about machine learning
  • How many computer chips it might require to be able to do most of the things humans do
  • How Richard understands the ‘alignment problem’ differently than other people
  • Why ‘situational awareness’ may be a key concept for understanding the behaviour of AI models
  • What work to positively shape the development of AI Richard is and isn’t excited about
  • The AGI Safety Fundamentals course that Richard developed to help people learn more about this field

Get this episode by subscribing to our podcast on the world’s most pressing problems and how to solve them: type ‘80,000 Hours’ into your podcasting app. Or read the transcript below.

Producer: Keiran Harris
Audio mastering: Milo McGuire and Ben Cordell
Transcriptions: Katy Moore

Continue reading →

Marcus Davis on founding and leading Rethink Priorities

You might say there’s nothing you can do now [to help wild animals]… Something like that might be true, but one of the things we can do is figure out what we should be trying to attempt.

So if you try to understand the welfare of these animals, just gather basic facts about what their lives are like, this could help you understand how you should do this.

Marcus Davis

In this episode of 80k After Hours, Rob Wiblin interviews Marcus Davis about Rethink Priorities.

Marcus is co-CEO there, in charge of their animal welfare and global health and development research.

They cover:

  • Interventions to help wild animals
  • Aquatic noise
  • Rethink Priorities strategy
  • Mistakes that RP has made since it was founded
  • Careers in global priorities research
  • And the most surprising thing Marcus has learned at RP

Who this episode is for:

  • People who want to learn about Rethink Priorities
  • People interested in a career in global priorities research
  • People open to novel ways to help wild animals

Who this episode isn’t for:

  • People who think global priorities research sounds boring
  • People who want to host very loud concerts under the sea

Get this episode by subscribing to our more experimental podcast on the world’s most pressing problems and how to solve them: type ’80k After Hours’ into your podcasting app. Or read the transcript below.

Producer: Keiran Harris
Audio mastering: Milo McGuire and Ben Cordell
Transcriptions: Katy Moore

Gershwin – Rhapsody in Blue, original 1924 version” by Jason Weinberger is licensed under creative commons

Continue reading →

Four values at the heart of effective altruism

What actually is effective altruism?

Effective altruism isn’t about any particular way of doing good, like AI alignment or distributing malaria nets. Rather, it’s a way of thinking.

Last summer, I wrote a new introduction to effective altruism for In it, I tried to sum up the effective altruism way of thinking in terms of four values. (I wrote this newsletter before FTX collapsed, but maybe that makes it even more important to reiterate the core values of EA.)

  1. Prioritisation. Resources are limited, so we have to make hard choices between potential interventions. While helping 10 people might feel as satisfying as helping 100, those extra 90 people really matter. And it turns out that some ways of helping achieve dramatically more than others, so it’s vital to really try to roughly compare ways we might help in terms of scale and effectiveness.
  2. Impartial altruism. It’s reasonable and good to have special concern for one’s own family, friends, life, etc. But when trying to do good in general, we should give everyone’s interests equal weight — no matter where or even when they live. People matter equally. And we should also give due weight to the interests of nonhumans.
  3. Open truth-seeking. Rather than starting with a commitment to a certain cause, consider many different ways to help and try to find the best ones you can. Put serious time into deliberation and reflection,

Continue reading →

Why being open to changing our minds is especially important right now

If something surprises you, your view of the world should change in some way.

We’ve argued that you should approach your career like a scientist doing experiments: be willing to test out many different paths and gather evidence about where you can have the most impact.

More generally, this approach of open truth-seeking — being constantly, curiously on the lookout for new evidence and arguments, and always being ready to change our minds — is a virtue we think is absolutely crucial to doing good.

One of our first-ever podcast episodes was an interview with Julia Galef, author of The Scout Mindset (before she wrote the book!).

Julia argues — in our view, correctly — that it’s easy to end up viewing the world like a soldier, when really you should be more like a scout.

Soldiers have set views and beliefs, and defend those beliefs. When we are acting like soldiers, we display motivated reasoning: for example, confirmation bias, where we seek out information that supports our existing beliefs and misinterpret information that is evidence against our position so that it seems like it’s not.

Scouts, on the other hand, need to form correct beliefs. So they have to change their minds as they view more of the landscape.

Acting like a scout isn’t always easy:

  • There’s lots of psychological evidence suggesting that we all have cognitive biases that cloud our thinking.

Continue reading →

Regarding the collapse of FTX

The collapse of FTX is likely to cause a tremendous amount of harm – to customers, employees, and many others who have relied on FTX. We are deeply concerned about those affected and, along with our community, are grappling with how to respond.

Though we do not know for sure whether anything illegal happened, we unequivocally condemn any immoral or illegal actions that may have taken place.

Prior to this, we had celebrated Sam Bankman-Fried’s apparent success, had held him up as a positive example of someone pursuing a high-impact career, and had written about how we encouraged him to use a strategy of earning to give (for example here). We feel shaken by recent events, and are not sure exactly what to say or think.

In the meantime, we will start by removing instances on our site where Sam was highlighted as a positive example of someone pursuing a high-impact career, since, to say the least, we no longer endorse that. We are leaving up discussions of Sam in places that seem important for transparency, for example this blog post on the growth of effective altruism in 2021.

In the coming weeks and months we will be thinking hard about what we should do going forward and ways in which we should have acted differently.

If you are out there trying the best you can to use your career to help solve the world’s most pressing problems with honesty and integrity,

Continue reading →

    #140 – Bear Braumoeller on the case that war isn’t in decline

    Imagine I have a deck of 96 cards. The most common card has 1,000 battle deaths, but one of the cards is World War I, and one of the cards is World War II. How worried should you be about drawing a card from that deck?

    You could say, “Well, most of them are 1,000 battle deaths, so I shouldn’t be too worried.” But at the same time, World War I and World War II are in there, and if the deck hasn’t changed, we really need to be thoughtful about when it is we’re going to draw another card.

    Bear Braumoeller

    Is war in long-term decline? Steven Pinker’s The Better Angels of Our Nature brought this previously obscure academic question to the centre of public debate, and pointed to rates of death in war to argue energetically that war is on the way out.

    But that idea divides war scholars and statisticians, and so Better Angels has prompted a spirited debate, with datasets and statistical analyses exchanged back and forth year after year. The lack of consensus has left a somewhat bewildered public (including host Rob Wiblin) unsure quite what to believe.

    Today’s guest, professor in political science Bear Braumoeller, is one of the scholars who believes we lack convincing evidence that warlikeness is in long-term decline. He collected the analysis that led him to that conclusion in his 2019 book, Only the Dead: The Persistence of War in the Modern Age.

    The question is of great practical importance. The US and PRC are entering a period of renewed great power competition, with Taiwan as a potential trigger for war, and Russia is once more invading and attempting to annex the territory of its neighbours.

    If war has been going out of fashion since the start of the Enlightenment, we might console ourselves that however nerve-wracking these present circumstances may feel, modern culture will throw up powerful barriers to another world war. But if we’re as war-prone as we ever have been, one need only inspect the record of the 20th century to recoil in horror at what might await us in the 21st.

    Bear argues that the second reaction is the appropriate one. The world has gone up in flames many times through history, with roughly 0.5% of the population dying in the Napoleonic Wars, 1% in World War I, 3% in World War II, and perhaps 10% during the Mongol conquests. And with no reason to think similar catastrophes are any less likely today, complacency could lead us to sleepwalk into disaster.

    He gets to this conclusion primarily by analysing the datasets of the decades-old Correlates of War project, which aspires to track all interstate conflicts and battlefield deaths since 1815. In Only the Dead, he chops up and inspects this data dozens of different ways, to test if there are any shifts over time which seem larger than what could be explained by chance variation alone.

    Among other metrics, Bear looks at:

    • Battlefield deaths alone, as a percentage of combatants’ populations, and as a percentage of world population.
    • The total number of wars starting in a given year.
    • Rates of war initiation as a fraction of all country pairs capable of fighting wars.
    • How likely it was during different periods that a given war would double in size.
    Image source.

    In a nutshell, and taking in the full picture painted by these different measures, Bear simply finds no general trend in either direction from 1815 through today. It seems like, as philosopher George Santayana lamented in 1922, “only the dead have seen the end of war”.

    That’s not to say things are the same in all periods. Depending on which indication of warlikeness you give the greatest weight, you can point to some periods that seem violent or pacific beyond what might be explained by random variation.

    For instance, Bear points out that war initiation really did go down a lot at the end of the Cold War, with peace probably fostered by a period of unipolar US dominance, and the end of great power funding for proxy wars.

    But that drop came after a period of somewhat above-average warlikeness during the Cold War. And surprisingly, the most peaceful period in Europe turns out not to be 1990–2015, but rather 1815–1855, during which the monarchical ‘Concert of Europe,’ scarred by the Napoleonic Wars, worked together to prevent revolution and interstate aggression.

    Why haven’t modern ideas about the immorality of violence led to the decline of war, when it’s such a natural thing to expect? Bear is no Enlightenment scholar, but his book notes (among other reasons) that while modernity threw up new reasons to embrace pacifism, it also gave us new reasons to embrace violence: as a means to overthrow monarchy, distribute the means of production more equally, or protect people a continent away from ethnic cleansing — all motives that would have been foreign in the 15th century.

    In today’s conversation, Bear and Rob discuss all of the above in more detail than even a usual 80,000 Hours podcast episode, as well as:

    • What would Bear’s critics say in response to all this?
    • What do the optimists get right?
    • What are the biggest problems with the Correlates of War dataset?
    • How does one do proper statistical tests for events that are clumped together, like war deaths?
    • Why are deaths in war so concentrated in a handful of the most extreme events?
    • Did the ideas of the Enlightenment promote nonviolence, on balance?
    • Were early states more or less violent than groups of hunter-gatherers?
    • If Bear is right, what can be done?
    • How did the ‘Concert of Europe’ or ‘Bismarckian system’ maintain peace in the 19th century?
    • Which wars are remarkable but largely unknown?
    • What’s the connection between individual attitudes and group behaviour?
    • Is it a problem that this dataset looks at just the ‘state system’ and ‘battlefield deaths’?

    Get this episode by subscribing to our podcast on the world’s most pressing problems and how to solve them: type ‘80,000 Hours’ into your podcasting app. Or read the transcript below.

    Producer: Keiran Harris
    Audio mastering: Ryan Kessler
    Transcriptions: Katy Moore

    Continue reading →

    The importance of considering speculative ideas

    Let’s admit it: some of the things we think about at 80,000 Hours are considered weird by a lot of other people.

    Our list of the most pressing problems has some pretty widely accepted concerns, to be sure: we care about mitigating climate change, preventing nuclear war, and ensuring good governance.

    But one of our highest priorities is preventing an AI-related catastrophe, which sounds like science fiction to a lot of people. And, though we know less about them, we’re also interested in speculative issues — such as atomically precise manufacturing, artificial sentience, and wild animal suffering. These aren’t typically the kind of issues activists distribute flyers about.

    Should it make us nervous that some of our ideas are out of the mainstream? It’s probably a good idea in these cases to take a step back, reexamine our premises, and consult others we trust about our conclusions. But we shouldn’t be too shocked if some of our beliefs end up at odds with common sense — indeed, I think everyone has good reason to be open to believing weird ideas.

    One of the best reasons for this view relates to another of 80,000 Hours’ top priorities: preventing catastrophic pandemics. I’d guess few people think it’s strange to be concerned about pandemics now, as COVID-19 has killed more than 6 million people worldwide and thrown the global economy into chaos.

    Continue reading →

    #139 – Alan Hájek on puzzles and paradoxes in probability and expected value

    Length is not bounded, volume is not bounded, time, spacetime curvature — various things are not bounded. And why should utility be?

    Normally when you do have a bounded quantity, you can say why it is and you can say what the bound is. Think of, say, angle: if you think of it one way, angle is bounded by zero to 360 degrees, and it’s easy to explain that. Probability is bounded with a top value of 1, bottom value of 0. Not so easy to say in the case of utility.

    Alan Hájek

    A casino offers you a game. A coin will be tossed. If it comes up heads on the first flip you win $2. If it comes up on the second flip you win $4. If it comes up on the third you win $8, the fourth you win $16, and so on. How much should you be willing to pay to play?

    The standard way of analysing gambling problems, ‘expected value‘ — in which you multiply probabilities by the value of each outcome and then sum them up — says your expected earnings are infinite. You have a 50% chance of winning $2, for ‘0.5 * $2 = $1’ in expected earnings. A 25% chance of winning $4, for ‘0.25 * $4 = $1’ in expected earnings, and on and on. A never-ending series of $1s added together comes to infinity. And that’s despite the fact that you know with certainty you can only ever win a finite amount!

    Today’s guest — philosopher Alan Hájek of the Australian National University — thinks of much of philosophy as “the demolition of common sense followed by damage control” and is an expert on paradoxes related to probability and decision-making rules like “maximise expected value.”

    The problem described above, known as the St. Petersburg paradox, has been a staple of the field since the 18th century, with many proposed solutions. In the interview, Alan explains how very natural attempts to resolve the paradox — such as factoring in the low likelihood that the casino can pay out very large sums, or the fact that money becomes less and less valuable the more of it you already have — fail to work as hoped.

    We might reject the setup as a hypothetical that could never exist in the real world, and therefore of mere intellectual curiosity. But Alan doesn’t find that objection persuasive. If expected value fails in extreme cases, that should make us worry that something could be rotten at the heart of the standard procedure we use to make decisions in government, business, and nonprofits.

    These issues regularly show up in 80,000 Hours’ efforts to try to find the best ways to improve the world, as the best approach will arguably involve long-shot attempts to do very large amounts of good.

    Consider which is better: saving one life for sure, or three lives with 50% probability? Expected value says the second, which will probably strike you as reasonable enough. But what if we repeat this process and evaluate the chance to save nine lives with 25% probability, or 27 lives with 12.5% probability, or after 17 more iterations, 3,486,784,401 lives with a 0.00000009% chance. Expected value says this final offer is better than the others — 1,000 times better, in fact.

    Insisting that people give up a sure thing in favour of a vanishingly low chance of a very large impact strikes some people as peculiar or even fanatical. But one of Alan’s PhD students, Hayden Wilkinson, discovered that rejecting expected value on this basis requires you to swallow even more bitter pills, like giving up on the idea that if A is better than B, and B is better than C, then A is also better than C.

    Ultimately Alan leans towards the view that our best choice is to “bite the bullet” and stick with expected value, even with its sometimes counterintuitive implications. Where we want to do damage control, we’re better off looking for ways our probability estimates might be wrong.

    In today’s conversation, Alan and Rob explore these issues and many others:

    • Simple rules of thumb for having philosophical insights
    • A key flaw that hid in Pascal’s wager from the very beginning
    • Whether we have to simply ignore infinities because they mess everything up
    • What fundamentally is ‘probability’?
    • Some of the many reasons ‘frequentism’ doesn’t work as an account of probability
    • Why the standard account of counterfactuals in philosophy is deeply flawed
    • And why counterfactuals present a fatal problem for one sort of consequentialism

    Get this episode by subscribing to our podcast on the world’s most pressing problems and how to solve them: type ‘80,000 Hours’ into your podcasting app. Or read the transcript below.

    Producer: Keiran Harris
    Audio mastering: Ben Cordell and Ryan Kessler
    Transcriptions: Katy Moore

    Continue reading →

    Open position: Recruiter

    The role

    You’ll be managed by Sashika Coxhead, our Head of Recruiting, and will have the opportunity to work closely with hiring managers from other teams.

    Initial responsibilities will include:

    • Project management of active recruiting rounds. For example, overseeing the candidate pipeline and logistics of hiring rounds, making decisions on initial applications, and managing candidate communications.
    • Sourcing potential candidates. This might include generating leads for specific roles, publicising new positions, reaching out to potential candidates, and answering any questions they have about working at 80,000 Hours.
    • Taking on special projects to improve our recruiting systems. For example, you might help to build an excellent applicant tracking system, test ways to improve our ability to generate leads, or introduce strategies to make our hiring rounds more efficient.

    Depending on your skills and interests, you might also:

    • Take ownership of a particular area of our recruiting process, e.g. proactive outreach to potential candidates, our applicant tracking system, or metrics for the recruiting team’s success.
    • Conduct screening interviews where needed, to assess applicants’ fit for particular roles at 80,000 Hours.

    After some time in the role, we’d hope for you to sit on internal hiring committees. This involves forming an inside view on candidates’ performance; discussing uncertainties with the hiring manager and committee; and, with the other committee members, giving final approval on who to make offers to.

    Continue reading →

      Anonymous advice: If you want to reduce AI risk, should you take roles that advance AI capabilities?

      We’ve argued that preventing an AI-related catastrophe may be the world’s most pressing problem, and that while progress in AI over the next few decades could have enormous benefits, it could also pose severe, possibly existential risks. As a result, we think that working on some technical AI research — research related to AI safety — may be a particularly high-impact career path.

      But there are many ways of approaching this path that involve researching or otherwise advancing AI capabilities — meaning making AI systems better at some specific skills — rather than only doing things that are purely in the domain of safety. In short, this is because:

      • Capabilities work and some forms of safety work are intertwined.
      • Many available ways of learning enough about AI to contribute to safety are via capabilities-enhancing roles.

      So if you want to help prevent an AI-related catastrophe, should you be open to roles that also advance AI capabilities, or steer clear of them?

      We think this is a hard question! Capabilities-enhancing roles could be beneficial or harmful. For any role, there are a range of considerations — and reasonable people disagree on whether, and in what cases, the risks outweigh the benefits.

      So we asked the 22 people we thought would be most informed about this issue — and who we knew had a range of views —

      Continue reading →

      #138 – Sharon Hewitt Rawlette on why pleasure and pain are the only things that intrinsically matter

      I actually think that if we didn’t ever experience pleasure or pain, or any of these positive or negative qualitative states, that we wouldn’t actually have the concept of intrinsic goodness that we do in fact have and that we do use when we’re making moral decisions.

      Sharon Hewitt Rawlette

      What in the world is intrinsically good — good in itself even if it has no other effects? Over the millennia, people have offered many answers: joy, justice, equality, accomplishment, loving god, wisdom, and plenty more.

      The question is a classic that makes for great dorm-room philosophy discussion. But it’s hardly just of academic interest. The issue of what (if anything) is intrinsically valuable bears on every action we take, whether we’re looking to improve our own lives, or to help others. The wrong answer might lead us to the wrong project and render our efforts to improve the world entirely ineffective.

      Today’s guest, Sharon Hewitt Rawlette — philosopher and author of The Feeling of Value: Moral Realism Grounded in Phenomenal Consciousness — wants to resuscitate an answer to this question that is as old as philosophy itself.

      That idea, in a nutshell, is that there is only one thing of true intrinsic value: positive feelings and sensations. And similarly, there is only one thing that is intrinsically of negative value: suffering, pain, and other unpleasant sensations.

      Lots of other things are valuable too: friendship, fairness, loyalty, integrity, wealth, patience, houses, and so on. But they are only instrumentally valuable — that is to say, they’re valuable as means to the end of ensuring that all conscious beings experience more pleasure and other positive sensations, and less suffering.

      As Sharon notes, from Athens in 400 BC to Britain in 1850, the idea that only subjective experiences can be good or bad in themselves — a position known as ‘philosophical hedonism’ — has been one of the most enduringly popular ideas in ethics.

      And few will be taken aback by the notion that, all else equal, more pleasure is good and less suffering is bad. But can they really be the only intrinsically valuable things?

      Over the 20th century, philosophical hedonism became increasingly controversial in the face of some seemingly very counterintuitive implications. For this reason the famous philosopher of mind Thomas Nagel called The Feeling of Value “a radical and important philosophical contribution.”

      So what convinces Sharon that philosophical hedonism deserves another go?

      Stepping back for a moment, any answer to the question “What has intrinsic value?” faces a serious challenge: “How do we know?” It’s far from clear how something having intrinsic value can cause us to believe that it has intrinsic value. And if there’s no causal or rational connection between something being valuable and our believing that it has value, we could only get the right answer by some extraordinary coincidence. You may feel it’s intrinsically valuable to treat people fairly, but maybe there’s just no reason to trust that intuition.

      Since the 1700s, many philosophers working on so-called ‘metaethics’ — that is, the study of what ethical claims are and how we could know if they’re true — have despaired of us ever making sense of or identifying the location of ‘objective’ or ‘intrinsic’ value. They conclude that when we say things are ‘good,’ we aren’t really saying anything about their nature, but rather just expressing our own attitudes, or intentions, or something else.

      Sharon disagrees. She says the answer to all this has been right under our nose all along.

      We have a concept of value because of our experiences of positive sensations — sensations that immediately indicate to us that they are valuable and that if someone could create more of them, they ought to do so. Similarly, we have a concept of badness because of our experience of suffering — sensations that scream to us that if suffering were all there were, it would be a bad thing.

      How do we know that pleasure is valuable, and that suffering is the opposite of valuable? Directly!

      While I might be mistaken that a painting I’m looking at is in real life as it appears to me, I can’t be mistaken about the nature of my perception of it. If it looks red to me, it may or may not be red, but it’s definitely the case that I am perceiving redness. Similarly, while I might be mistaken that a painting is intrinsically valuable, I can’t be mistaken about the pleasurable sensations I’m feeling when I look at it, and the fact that among other qualities those sensations have the property of goodness.

      While intuitive on some level, this arguably implies some very strange things. Most famously, the philosopher Robert Nozick challenged it with the idea of an ‘experience machine’: if you could enter into a simulated world and enjoy a life far more pleasurable than the one you experience now, should you do so, even if it would mean none of your accomplishments or relationships would be ‘real’? Nozick and many of his colleagues thought not.

      The idea has also been challenged for failing to value human freedom and autonomy for its own sake. Would it really be OK to kill one person to use their organs to save the lives of five others, if doing so would generate more pleasure and less suffering? Few believe so.

      In today’s interview, Sharon explains the case for a theory of value grounded in subjective experiences, and why she believes these counterarguments are misguided. A philosophical hedonist shouldn’t get in an experience machine, nor override an individual’s autonomy, except in situations so different from the classic thought experiments that it no longer seems strange they would do so.

      Host Rob Wiblin and Sharon cover all that, as well as:

      • The essential need to disentangle intrinsic, instrumental, and other sorts of value
      • Why Sharon’s arguments lead to hedonistic utilitarianism rather than hedonistic egoism (in which we only care about our own feelings)
      • How do people react to the ‘experience machine’ thought experiment when surveyed?
      • Why hedonism recommends often thinking and acting as though it were false
      • Whether it’s crazy to think that relationships are only useful because of their effects on our subjective experiences
      • Whether it will ever be possible to eliminate pain, and whether doing so would be desirable
      • If we didn’t have positive or negative experiences, whether that would cause us to simply never talk about goodness and badness
      • Whether the plausibility of hedonism is affected by our theory of mind
      • And plenty more

      Get this episode by subscribing to our podcast on the world’s most pressing problems and how to solve them: type ‘80,000 Hours’ into your podcasting app. Or read the transcript below.

      Producer: Keiran Harris
      Audio mastering: Ryan Kessler
      Transcriptions: Katy Moore

      Continue reading →

      Kuhan Jeyapragasan on effective altruism university groups

      I was just thinking about what a unique situation university is.

      Where and when else do you see such a large concentration of really caring, dedicated, driven, talented, ambitious people who are figuring out their values, and are actively trying to figure out what they want to do with the rest of their life?

      Kuhan Jeyapragasan

      In this episode of 80k After Hours, Rob Wiblin interviews Kuhan Jeyapragasan about effective altruism university groups.

      From 2015 to 2020, Kuhan did an undergrad and then a master’s in maths and computer science at Stanford — and did a lot to organise and improve the EA group on campus.

      Rob and Kuhan cover:

      • The challenges of making a group appealing and accepting of everyone
      • The concrete things Kuhan did to grow the successful Stanford EA group
      • Whether local groups are turning off some people who should be interested in effective altruism, and what they could do differently
      • Lessons Kuhan learned from Stanford EA
      • The Stanford Existential Risks Initiative (SERI)

      Who this episode is for:

      • People already involved in EA university groups
      • People interested in getting involved in EA university groups

      Who this episode isn’t for:

      • People who’ve never heard of ‘effective altruism groups’
      • People who’ve never heard of ‘effective altruism’
      • People who’ve never heard of ‘university’

      Get this episode by subscribing to our more experimental podcast on the world’s most pressing problems and how to solve them: type ’80k After Hours’ into your podcasting app. Or read the transcript below.

      Producer: Keiran Harris
      Audio mastering: Ryan Kessler
      Transcriptions: Katy Moore

      Gershwin – Rhapsody in Blue, original 1924 version” by Jason Weinberger is licensed under creative commons

      Continue reading →