The Nonlinear Library

The Nonlinear Library

By The Nonlinear FundEducation
Download on the App Store

All The Nonlinear Library episodes

  • AF - An Overview of Catastrophic AI Risks #2 by Dan H
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: An Overview of Catastrophic AI Risks #2, published by Dan H on June 22, 2023 on The AI Alignment Forum.
    This is the second post in a sequence of posts giving an overview of catastrophic AI risks.
    2 Malicious Use
    On the morning of March 20, 1995, five men entered the Tokyo subway system. After boarding separate subway lines, they continued for several stops before dropping the bags they were carrying and exiting. An odorless, colorless liquid inside the bags began to vaporize. Within minutes, commuters began choking and vomiting. The trains continued on toward the heart of Tokyo, with sickened passengers leaving the cars at each station. The fumes were spread at each stop, either by emanating from the tainted cars or through contact with people's clothing and shoes. By the end of the day, 13 people lay dead and 5,800 seriously injured. The group responsible for the attack was the religious cult Aum Shinrikyo [5]. Its motive for murdering innocent people? To bring about the end of the world.
    Powerful new technologies offer tremendous potential benefits, but they also carry the risk of empowering malicious actors to cause widespread harm. There will always be those with the worst of intentions, and AIs could provide them with a formidable tool to achieve their objectives. Moreover, as AI technology advances, severe malicious use could potentially destabilize society, increasing the likelihood of other risks.
    In this section, we will explore the various ways in which the malicious use of advanced AIs could pose catastrophic risks. These include engineering biochemical weapons, unleashing rogue AIs, using persuasive AIs to spread propaganda and erode consensus reality, and leveraging censorship and mass surveillance to irreversibly concentrate power. We will conclude by discussing possible strategies for mitigating the risks associated with the malicious use of AIs.
    Unilateral actors considerably increase the risks of malicious use. In instances where numerous actors have access to a powerful technology or dangerous information that could be used for harmful purposes, it only takes one individual to cause significant devastation. Malicious actors themselves are the clearest example of this, but recklessness can be equally dangerous. For example, a single research team might be excited to open source an AI system with biological research capabilities, which would speed up research and potentially save lives, but this could also increase the risk of malicious use if the AI system could be repurposed to develop bioweapons. In situations like this, the outcome may be determined by the least risk-averse research group. If only one research group thinks the benefits outweigh the risks, it could act unilaterally, deciding the outcome even if most others don't agree.
    And if they are wrong and someone does decide to develop a bioweapon, it would be too late to reverse course.
    By default, advanced AIs may increase the destructive capacity of both the most powerful and the general population. Thus, the growing potential for AIs to empower malicious actors is one of the most severe threats humanity will face in the coming decades. The examples we give in this section are only those we can foresee. It is possible that AIs could aid in the creation of dangerous new technology we cannot presently imagine, which would further increase risks from malicious use.
    2.1 Bioterrorism
    The rapid advancement of AI technology increases the risk of bioterrorism. AIs with knowledge of bioengineering could facilitate the creation of novel bioweapons and lower barriers to obtaining such agents. Engineered pandemics from AI-assisted bioweapons pose a unique challenge, as attackers have an advantage over defenders and could constitute an existential threat to humanity. We will now examine these risks and...
    33 min
  • AF - An Overview of Catastrophic AI Risks #1 by Dan H
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: An Overview of Catastrophic AI Risks #1, published by Dan H on June 22, 2023 on The AI Alignment Forum.
    This is the first post in a sequence of posts giving an overview of catastrophic AI risks.
    Abstract
    Rapid advancements in artificial intelligence (AI) have sparked growing concerns among experts, policymakers, and world leaders regarding the potential for increasingly advanced AI systems to pose catastrophic risks. Although numerous risks have been detailed separately, there is a pressing need for a systematic discussion and illustration of the potential dangers to better inform efforts to mitigate them. This paper provides an overview of the main sources of catastrophic AI risks, which we organize into four categories: malicious use, in which individuals or groups intentionally use AIs to cause harm; AI race, in which competitive environments compel actors to deploy unsafe AIs or cede control to AIs; organizational risks, highlighting how human factors and complex systems can increase the chances of catastrophic accidents; and rogue AIs, describing the inherent difficulty in controlling agents far more intelligent than humans.
    For each category of risk, we describe specific hazards, present illustrative stories, envision ideal scenarios, and propose practical suggestions for mitigating these dangers. Our goal is to foster a comprehensive understanding of these risks and inspire collective and proactive efforts to ensure that AIs are developed and deployed in a safe manner. Ultimately, we hope this will allow us to realize the benefits of this powerful technology while minimizing the potential for catastrophic outcomes.
    Executive Summary
    Artificial intelligence (AI) has seen rapid advancements in recent years, raising concerns among AI experts, policymakers, and world leaders about the potential risks posed by advanced AIs. As with all powerful technologies, AI must be handled with great responsibility to manage the risks and harness its potential for the betterment of society. However, there is limited accessible information on how catastrophic or existential AI risks might transpire or be addressed. While numerous sources on this subject exist, they tend to be spread across various papers, often targeted toward a narrow audience or focused on specific risks. In this paper, we provide an overview of the main sources of catastrophic AI risk, which we organize into four categories:
    Malicious use. Actors could intentionally harness powerful AIs to cause widespread harm. Specific risks include bioterrorism enabled by AIs that can help humans create deadly pathogens; the deliberate dissemination of uncontrolled AI agents; and the use of AI capabilities for propaganda, censorship, and surveillance. To reduce these risks, we suggest improving biosecurity, restricting access to the most dangerous AI models, and holding AI developers legally liable for damages caused by their AI systems.
    AI race. Competition could pressure nations and corporations to rush the development of AIs and cede control to AI systems. Militaries might face pressure to develop autonomous weapons and use AIs for cyberwarfare, enabling a new kind of automated warfare where accidents can spiral out of control before humans have the chance to intervene. Corporations will face similar incentives to automate human labor and prioritize profits over safety, potentially leading to mass unemployment and dependence on AI systems. We also discuss how evolutionary dynamics might shape AIs in the long run. Natural selection among AIs may lead to selfish traits, and the advantages AIs have over humans could eventually lead to the displacement of humanity. To reduce risks from an AI race, we suggest implementing safety regulations, international coordination, and public control of general-purpose AIs.
    Organizationa...
    13 min
  • EA - Rejection thread: stories and tips by Luisa Rodriguez
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Rejection thread: stories and tips, published by Luisa Rodriguez on June 22, 2023 on The Effective Altruism Forum.
    Getting rejected from jobs can be crushing — but learning how to deal with rejection productively is an incredibly valuable skill. And hearing others' rejection stories can make us feel less alone and judged, and generally help us orient toward rejection in more productive ways.
    Let's use this thread to help each other with this.
    If you're up for it, comment and share:
    Rejection stories you might have
    Any lessons you've learned for coping with rejection
    What has helped in the past
    You can also message Lizka to share rejections that she will anonymize and add to the comments, or you can omit some details or just share tips without sharing the rejection stories themselves.
    Sharing rejections like this can be hard. Don't force yourself to do it if it stresses you out. And if you're commenting on this post, please remember to be kind.
    Luisa's experience — shared in the 80,000 Hours newsletter
    Rejection was the topic of this week's 80,000 Hours newsletter, where Luisa shared a lot about her experience and how she's learned to cope with it. (That prompted this thread!) She wrote the following:
    I've been rejected many, many times. In 2015, I applied to ten PhD programs and was rejected from nine. After doing a summer internship with GiveWell in 2016, I wasn't offered a full-time role. In 2017, I was rejected by J-PAL, IDinsight, and Founders Pledge (among others). Around the same time, I was so afraid of being rejected by Open Philanthropy, I dropped out of their hiring round.
    I now have what I consider a dream job at 80,000 Hours: I get to host a podcast about the world's most pressing problems and how to solve them. But before getting a job offer from 80,000 Hours in 2020, I got rejected by them for a role in 2018. That rejection hurt the most.
    I still remember compulsively checking my phone after my work trial to see if 80,000 Hours had made me an offer. And I still remember waking up at 5:00 AM, checking my email, and finding the kind and well-written — but devastating — rejection: "Unfortunately we don't think the role is the right fit right now."
    And I remember being so sad that I took a five-hour bus ride to stay with a friend so I wouldn't have to be alone. After a few days of wallowing, I re-read the rejection email and noticed a lot of specific feedback — and a promising path forward.
    "We're optimistic about your career in global prioritisation research and think you should stay in the area and build experience," they said. "We're not going anywhere, and could be a good career transition for you further down the line."
    I took their advice and accepted a job offer at Rethink Priorities, which also does global priorities research. And a year and a half later, 80,000 Hours invited me to apply for a job again.
    It's hard to say what would've happened had I not opened myself to rejection in 2018, but it seems possible I'd be in a pretty different place. While that rejection was really painful, the feedback I got was a huge help in moving my research career forward. I think there's an important lesson here.
    For me, rejection is one of the worst feelings. But whether you're like me, looking to work in global priorities research at small nonprofits, or interested to work in another potentially impactful path, getting rejected can come with unexpected benefits:
    When you get rejected from a role you thought was a good fit, you get more information about your strengths and weaknesses. It can indicate whether you need more career capital or should perhaps consider different types of roles or paths altogether.
    When applying for roles in an ecosystem you want to work in, you grow the number of people in that field who know you and who might reach out to you for futu...
    6 min
  • EA - RP’s AI Governance & Strategy team - June 2023 interim overview by MichaelA
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: RP’s AI Governance & Strategy team - June 2023 interim overview, published by MichaelA on June 22, 2023 on The Effective Altruism Forum.
    Hi! I co-lead Rethink Priorities’ AI Governance & Strategy (AIGS) team. At the suggestion of Ben West, I’m providing an update on our team.
    Caveats:
    This was quickly written and omits most of the reasoning for our choices, because that was all I had time to write and that seemed better than not providing an update at all.
    This post may not reflect the views of all members of the team, and doesn’t represent RP as an organization.
    The areas we work in are evolving rapidly, so our strategy and projects are as well.
    Comments and DMs are welcome, though I can’t guarantee a rapid or detailed reply.
    Summary
    The AIGS team works to reduce catastrophic risks related to AI by conducting research and strengthening the field of AI governance. We aim to bridge the technical and policy worlds, and we now focus on short, rapid-turnaround outputs and briefings. [read more]
    Our four key workstreams are compute governance, China, lab governance, and US regulations. [read more]
    We list some of our ongoing or completed projects. [read more]
    Please feel free to reach out if you’d like to suggest a project; if you’re open to sharing feedback, expertise, or connections with us; or if you or someone you know might be interested in working with or funding us. [read more]
    I summarize a few lessons learned and recent updates. [read more]
    Who we are
    Rethink Priorities’ AI Governance & Strategy team works to reduce catastrophic risks related to development & deployment of AI systems. We do this by producing research that grounds concrete recommendations in strategic considerations, and by strengthening coordination and talent pipelines across the AI governance field.
    We combine the intellectual independence and nonpartisanship of a think tank with the flexibility and responsiveness of a consultancy. Our work is funded solely by foundations and independent donors, giving us the freedom to pursue important questions without bias. We’re always on the lookout for unexplored high-value research questions–feel free to pitch us!
    We aim to bridge the technical and policy worlds, with expertise on foundation models and the hardware underpinning them.
    We focus on short, rapid-turnaround outputs and briefings, but also produce longer reports. Much of our work is nonpublic, but may be shareable on request.
    We have 11 staff, listed here. You can contact any of us at [email protected]
    Our four workstreams
    We recently narrowed down to four focus areas, each of which has a 1-3 person subteam working on it. Below we summarize these workstreams and link to docs that provide further information on each (e.g., about ongoing projects, public outputs, and stakeholders and paths to impact).
    Compute governance: This workstream will focus on establishing a firmer empirical and theoretical grounding for the fledgling field of compute governance, informing ongoing policy processes and debates, and developing more concrete technical and policy proposals. In particular, we will focus on understanding the impact of existing compute-related US export controls, and researching what changes to them may be feasible and beneficial.
    This workstream consists of Onni Aarne and Erich Grunewald, and we’re currently hiring a third member.
    China: This workstream’s mission is to improve decisions at the intersection of AI governance and China. We are interested in both China-West relations concerning AI, as well as AI developments within China. We are particularly focused on informing decision-makers who are concerned about catastrophic risks from AI.
    This workstream consists of Oliver Guest.
    Lab governance: This workstream identifies concrete measures frontier AI labs can adopt now and in ...
    13 min
  • LW - "textbooks are all you need" by bhauth
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: "textbooks are all you need", published by bhauth on June 21, 2023 on LessWrong.
    "Textbooks Are All You Need" was published yesterday by Microsoft Research. It's the worst-named paper I've seen recently: it's not about textbooks, it's not all you need, and gratuitously imitating the title of a paper that introduced a different type of thing is dumb. But there's a reason I'm writing about it.
    What they did was basically this:
    started with The Stack (a 3 TB collection of code) and text from StackOverflow
    used a LLM to select 6B "high-quality" tokens from (1)
    used GPT-3.5 to generate 1B tokens of text similar to textbooks
    trained a small (1.3B parameter) model ("phi-1") on (2) and (3)
    used GPT-3.5 to generate text similar to textbook exercises
    fine-tuned phi-1 on (5)
    tested phi-1 on HumanEval to evaluate its programming ability
    The results were pretty good, better than models 10x the size trained on 100x the data. So, it seems that scaling up isn't the only thing that matters, and data quality can be more important than data quantity or parameter count. (You hear that, gwern?)
    Going by the listed OpenAI API prices, running GPT-3.5 on The Stack to evaluate quality would've been maybe ~$6M. What the authors did instead was:
    Use GPT-4 to evaluate a small fraction of it.
    Use a much smaller code-specific model to generate embeddings.
    Use a classifier to predict which embeddings are from what GPT-4 evaluates as good content.
    How about if you bootstrap a model using its own evaluation for filtering? One of the authors says "I'm almost sure you can beat the teacher model" and I agree. That can give you recursive self-improvement of a type you see in both individual people and the culture of societies. People develop better taste and consume better content which makes them smarter so they develop better taste, and so on. Children hear the stories their grandfathers like, and culture develops.
    That's a weak sort of self-improvement, which tends to plateau for people. Humans do other things too, so by itself it's weaker self-improvement than it appears to be for people. This is a technique I previously spent some time thinking about, so there are some other reasons I think it tends to plateau by itself. But still - recursive self-improvement!
    Yes, in theory, if you have a much bigger model trained on a bigger dataset including the good selected data, and you can engineer prompts such that you get into a mode that models the good data specifically, then you can get the same results. In that sense, the performance reachable with this method is limited to what's possible from model scaling plus prompt engineering. The amount of scaling needed for that seems to be potentially 100x, and getting into exactly the right mode with prompt engineering might be impractical, but still, that provides some rough limits on potential here.
    Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org.
    4 min
  • EA - Lab-grown meat is cleared for sale in the United States by Ben West
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Lab-grown meat is cleared for sale in the United States, published by Ben West on June 22, 2023 on The Effective Altruism Forum.
    Upside Foods and Good Meat, two companies that make what they call “cultivated chicken,” said Wednesday that they have gotten approval from the US Department of Agriculture to start producing their cell-based proteins.
    Good Meat, which is owned by plant-based egg substitute maker Eat Just, said that production is starting immediately.
    Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org
    1 min
  • EA - Announcing the University of Chicago’s $2M Market Shaping Accelerator’s Innovation Challenge: Biosecurity, Pandemic Preparedness, and Climate Change by schethik
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Announcing the University of Chicago’s $2M Market Shaping Accelerator’s Innovation Challenge: Biosecurity, Pandemic Preparedness, and Climate Change, published by schethik on June 22, 2023 on The Effective Altruism Forum.
    The Market Shaping Accelerator (MSA) is a new initiative at the University of Chicago aiming to accelerate innovations to address pressing global challenges. It is led by Michael Kremer, Rachel Glennerster and Christopher Snyder. Our current focus areas are climate change, biosecurity and pandemic preparedness. We recently launched the MSA Innovation Challenge, which will award up to $2,000,000 total in prizes for ideas about problems to tackle with pull incentives for innovation in these areas. Pull mechanisms reward outputs and outcomes in contrast to push funding which pay for inputs (e.g. research grants).
    We want to invite members of the EA community and others to submit your ideas to the challenge. We are interested in hearing from domain experts, innovators, EA organizations as well as people just interested in a problem. You can read more about the challenge here (check out the FAQ on the bottom of this page) and view the application template here. The deadline to submit for Phase I is Friday, July 21, 2023 (12 PM CT). Submissions that meet a minimum criteria will receive $4,000. Up to $500,000 in prizes will be awarded in Phase I.
    Ideas that are selected for entry into Phase II will benefit from support and guidance of the MSA team as well as domain specialists to help turn their ideas into fully worked up contracts. Top ideas will also gain the MSA’s support in fundraising for the multi-millions or billions of dollars needed to back their pull mechanism.
    ITN Framework
    The MSA Innovation Challenge is partly informed by the ITN Framework – we are seeking to surface major problems (importance), that can plausibly be addressed by innovation (tractability), but where innovation is under-incentivized by markets (neglectedness).
    We would welcome both technologically close and technologically distant targets – we think pull mechanisms can accelerate innovation and scale up for both.
    What is pull funding?
    Pull mechanisms reward outputs and outcomes rather than fund inputs. They create an incentive for the private sector to invest in R&D and bring solutions to market. Advance Market Commitments (AMCs) are an example of a pull mechanism. AMCs involve promising, in advance, to purchase or subsidize the purchase of a large quantity of an innovative product if it is invented. The $1.5 billion Advance Market Commitment for the Pneumococcal Vaccine was launched in 2009. Since then three vaccines for the strains of pneumococcus common in low- and middle-income countries have been developed, hundreds of millions of doses delivered, and an estimated 700,000 lives saved. The rate of vaccine coverage for the pneumococcal vaccine in GAVI countries converged to the global rate five years faster than for the rotavirus vaccine which GAVI supported without an AMC.
    Pull mechanisms have important advantages:
    They can be designed to be firm and solution-agnostic. The funder does not have to choose a particular firm or technological path in advance, they can just commit to rewarding an effective solution.
    The funder does not have to pay unless the targets are met. Payment can be linked to scale and take-up.
    They reduce demand uncertainty – they can signal to firms there will be demand for socially useful innovations.
    They can incentivize solutions that appeal to consumers. The funder can provide a matching subsidy to a consumer purchase. This incentivizes firms to develop products that consumers will actually use.
    Further reading
    Advance Market Commitments: Insights from Theory and Experience
    The Case for More Pull Financing
    Making Markets for Vaccines: Ideas to Action...
    5 min
  • LW - The Hubinger lectures on AGI safety: an introductory lecture series by evhub
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: The Hubinger lectures on AGI safety: an introductory lecture series, published by evhub on June 22, 2023 on LessWrong.
    In early 2023, I (Evan Hubinger) gave a series of recorded lectures to SERI MATS fellows with the goal of building up a series of lectures that could serve as foundational introductory material to a variety of topics in AGI safety. Those lectures have now been edited and are available on YouTube for anyone who would like to watch them.
    The basic goal of this lecture series is to serve as longform, in-depth video content for people who are new to AGI safety, but interested enough to be willing to spend a great deal of time engaging with longform content, and who prefer video content to written content. Though we already have good introductory shortform video content and good introductory longform written content, the idea of this lecture series is to bridge the gap between those two.
    Note that the topics I chose to include are highly opinionated: though this is introductory material, it is not intended to introduce the listener to every topic in AI safety—rather, it is focused on the topics that I personally think are most important to understand. This is intentional: in my opinion, I think it is far more valuable to have some specific gears-level model of how to think about AI safety, rather than a shallow overview of many different possible ways of thinking about AI safety. The former allows you to actually start operationalizing that model to work on interventions that would be valuable under it, something the latter doesn't do.
    The lecture series is composed of six lectures, each around 2 hours long, covering the topics:
    Machine learning + instrumental convergence
    Risks from learned optimization
    Deceptive alignment
    How to evaluate alignment proposals
    LLMs + predictive models
    Overview of alignment proposals
    Each lecture features a good deal of audience questions both in the middle and at the end, the idea being to hopefully pre-empt any questions or confusions the listener might have.
    The full slide deck for all the talks is available here.
    Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org.
    3 min
  • AF - The Hubinger lectures on AGI safety: an introductory lecture series by Evan Hubinger
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: The Hubinger lectures on AGI safety: an introductory lecture series, published by Evan Hubinger on June 22, 2023 on The AI Alignment Forum.
    In early 2023, I (Evan Hubinger) gave a series of recorded lectures to SERI MATS fellows with the goal of building up a series of lectures that could serve as foundational introductory material to a variety of topics in AGI safety. Those lectures have now been edited and are available on YouTube for anyone who would like to watch them.
    The basic goal of this lecture series is to serve as longform, in-depth video content for people who are new to AGI safety, but interested enough to be willing to spend a great deal of time engaging with longform content, and who prefer video content to written content. Though we already have good introductory shortform video content and good introductory longform written content, the idea of this lecture series is to bridge the gap between those two.
    Note that the topics I chose to include are highly opinionated: though this is introductory material, it is not intended to introduce the listener to every topic in AI safety—rather, it is focused on the topics that I personally think are most important to understand. This is intentional: in my opinion, I think it is far more valuable to have some specific gears-level model of how to think about AI safety, rather than a shallow overview of many different possible ways of thinking about AI safety. The former allows you to actually start operationalizing that model to work on interventions that would be valuable under it, something the latter doesn't do.
    The lecture series is composed of six lectures, each around 2 hours long, covering the topics:
    Machine learning + instrumental convergence
    Risks from learned optimization
    Deceptive alignment
    How to evaluate alignment proposals
    LLMs + predictive models
    Overview of alignment proposals
    Each lecture features a good deal of audience questions both in the middle and at the end, the idea being to hopefully pre-empt any questions or confusions the listener might have.
    The full slide deck for all the talks is available here.
    Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org.
    3 min
  • LW - Which personality traits are real? Stress-testing the lexical hypothesis by tailcalled
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Which personality traits are real? Stress-testing the lexical hypothesis, published by tailcalled on June 21, 2023 on LessWrong.
    This post is also available on my Substack. Thanks to Justis Mills for proofreading and feedback!
    Most scientific personality models are, directly or indirectly, based on the lexical hypothesis, which roughly speaking states that there is a correspondence between important personality traits and abstract behavior-descriptive adjectives. For example, the Big Five was created by having people rate themselves using words like "outgoing", "hard-working" and "kind", and finding patterns in these. It is neat that one can create models in this way, but the large amount of abstraction involved by using abstract adjectives raises huge questions about how "real" the personality traits are.
    I have created a new personality test, currently named Targeted Personality Test. I have multiple goals with this test, but one of them is to investigate which personality traits are “real” without relying on the lexical hypothesis. I do this mainly by assessing lots of specific narrow behaviors, rather than abstract vague adjectives.
    By the end of this blog post, I hope to have introduced some concepts that makes my approach make sense, and thereby enable you to understand this diagram I made summarizing my results:
    The semi-formal understanding of what is going on in this chart is very long, so before we proceed, let me give a brief, vague indication of what you will be informed about:
    Trait impact: a measure of how strongly the personality trait influences the various behaviors and thoughts that we would expect it to.
    Factor model loss: a measure of how much the personality trait conflates different unrelated things together.
    Correlation with lexical notion: a measure of how well-labelled the personality trait is. (You can mostly ignore this variable as all of the personality traits performed reasonably well on it.)
    Easy-Goingness: An example
    A conventional personality test such as the SPI-81-27&5 might measure your personality traits such as Easy-Goingness by asking you how well a few abstract statements describe you, e.g.:
    I like to take it easy
    I like a leisurely lifestyle
    I have a slow pace to my life
    It seems plausible that someone who agrees that such statements describe them would be Easy-Going in some sense, and indeed I bet this sort of measure can pass all sorts of criteria used by psychologists to evaluate the quality of the test.
    So if it is probably valid by the standard criteria, what could go wrong that the standard criteria don’t test for?
    Well, let’s imagine the sort of person who is Easy-Going. They probably tend to relax in their free time, e.g. watching TV, and they probably don’t get worked up about controversial stuff, and they probably don’t go above and beyond at work. Basically, a relaxed person who doesn’t get too stressed or excited about things.
    When we call the person above “Easy-Going”, is this just a convenient label we use for someone who happens to have a constellation of traits like the above? Or are we saying that there is some underlying factor, like a motive to take it easy, which causes them to have these sorts of characteristics? Or maybe there are some underlying factors, but they are heterogeneous and “Easy-Goingness” lumps them together? These are the sorts of questions I tried to investigate.
    My first step was to come up with a more concrete characterization of Easy-Goingness than abstract statements like “I like to take it easy” or “I have a slow pace to my life”. I did this by giving the SPI-81-27&5 test to a bunch of people, and then asking the people who score high and low in Easy-Goingness to describe an example of how they could be said to be Easy-Going. To give you a taste for the answers, one person who scored h...
    16 min