{"slug": "arguments-for-and-against-me-dropping-out", "title": "Arguments for and against (me) dropping out", "summary": "A year into his undergraduate studies, a student is considering dropping out to focus on AI safety work, citing short timelines for transformative AI and the belief that academia is underprepared for a post-AGI world. He plans to engage in organizing, research, and building context, with specific initiatives including expanding GT AISI, supporting new organizations, founding an AI Lab Watch revitalization, and scaling field-building efforts.", "body_md": "One year ago, I was preparing for my first year of undergrad. Today, I’m considering dropping out. What changed?\n\nBefore writing this post, I attribute my decision to variety of (unordered) reasons:\n\n**Short timelines:** I believe rapid takeoff (<2 years) is likely. However, I don’t think I’ve spent enough time thinking hard about this, and am open to changing my mind.**Bandwidth:** School is a major bottleneck on my bandwidth. I think my time is more valuable spent on some subset of {organizing, research, building context, working}.**Academia is Underprepared:** I believe AI will be highly transformative. I’m doubtful that the current curriculum at my university will impart foundational skills that hold up for the rest of my life in a post-AGI (or post-ASI) world, especially as a CS major.**Peer Pressure:** A lot of the people I take seriously are planning on/recommend dropping out. This signals to me that dropping out is something I should consider seriously.\n\nThis is a pretty important decision. I’m not super confident in my motivations, so I’m writing this post to hash out exactly why (or why not) dropping out is the right decision to make. I’m posting it because 1) getting feedback from others is the fastest way to test my ideas, and 2) this could be helpful for someone in the same position as me.\n\n*Thanks to Zephy Roe, Ishan Khire, and Naren Manikandan for comments, and my sister, Anish Kallu, and Meru Gopalan for relevant discussion.*\n\n**My perspective on the different levels of dropping out**\n\nI’m somewhat against viewing university as a binary choice, i.e. either I enroll or drop out. A more intuitive perspective is measuring my engagement with school as a continuous scale, with dropping out representing 0% engagement and my maximum academic capacity as 100%.\n\nWhen I think about my academic engagement over the past academic year, my first semester fluctuated around the average student range (65%), starting above and then dropping below. My second semester was noticeably lower (~40%), as I began to prioritize organizing and engaging with AI safety over my coursework. This upcoming semester, I’m much closer to “taking 1 class” (15%), and slightly considering dropping out completely.\n\nWhy does this matter? When I compare dropping out to staying in school, I can’t assume my engagement with school will be at 100%. Rather, it's much more likely that staying in school would only mean a 10% level of engagement with my classes, which is qualitatively much different than maximal engagement.*\n\n**What am I dropping out for**?\n\nWhen I speak about dropping out, it's unclear what my concrete plan is, apart from working in a generalist capacity in the near future. I’m still working out exactly what shape I want my career to take, but for all of the generalist-adjacent roles I’m considering, dropping out is instrumentally valuable.\n\nSome of my current ideas and plans look like:\n\n- Organizing GT AISI full-time, and ambitiously expanding our organization (building out our community, fellowship, research initiatives with a center, etc.)\n- Providing operational support for newer organizations that are getting off the ground, similar to my work at\n[Second Look Research](https://secondlookresearch.com/) this summer. - Founding an org dedicated to revitalizing\n[AI Lab Watch](https://ailabwatch.org/), and acting as a third-party auditor of frontier lab safety practices. - Scaling up field-building and the talent pipeline into the field, especially for generalist and policy roles. This looks like running workshops and programs that funnel people into the field (I’m currently working on this by running\n[DCMC 2.0](http://dcminiconf.com)), as well as filling in the gaps in current hiring pipelines.\n\n**Defining the different factors / assumptions at play**\n\nI briefly touched on this at the start, but I want to clearly detail what key factors/questions are influencing my decision.\n\n**Timelines.** How fast do I think massively transformative AI (TAI) will arrive? Shorter timelines mean it makes more sense to drop out, while longer timelines enable me to invest more time into upskilling without sacrificing future impact.\n\n**Quality of education.** How robust is the current education I would receive to post-ASI worlds? Is there value in me getting an education right now, or can this be deferred? How does this change over time or if I switch degrees?\n\n**My (relative) value add.** Is my current skill set valuable to moving the needle on AI safety in some way? If (relative to other members of the field open to work) I’m incompetent, it makes more sense for me to dedicate time towards upskilling and building context, which is more compatible with staying enrolled in school.\n\n**Current responsibilities.** Some responsibilities require me to be at school, at least physically. For example, it would be very hard to co-direct GT AISI without being a student (or actively in Atlanta at the very least). Can I soundly forfeit these responsibilities and trust that they will be taken care of, or is it important for me to maintain them.\n\n**Bandwidth.** Do I have capacity to take on meaningful work, or am I bottlenecked in some ways (time, location, mental effort)? Does dropping out of school meaningfully reduce this bottleneck?\n\n**Maturity.** Even if I’m competent enough, I might not be mature enough. There are things I just don’t know yet (i.e. I haven’t lived enough life) and maybe going to college is the best way for me to do this. This seems pretty hard to measure, but could have meaningful implications for my quality of work.\n\n**External pressures.** While it is ultimately my decision, it doesn’t always feel that way. I don’t want to disappoint my family, who strongly value receiving a proper education and earning a degree. At the same time, I feel the pull from my AIS friends who are planning on dropping out, and the excitement and promise of impact “the real world” holds.\n\n**Happiness.** I think it's important that I’m happy and satisfied with whatever decision I make!\n\n**Money. **I consider myself extremely privileged, and could comfortably complete college if I wanted to – however, it's not exactly free. I’m unclear how the economics of the future will look, so I’m hesitant on spending money on a degree I may eventually drop out of.\n\n**Arguments for dropping out**\n\nHere are my current arguments for dropping out, derived from the factors I listed above. The biggest drivers pushing me to drop out are {Short Timelines, Quality of Education, Relative Competence, Bandwidth}.\n\n**Short timelines mean major action is required ASAP**. I generally subscribe to shorter timelines for massively transformative AI, which means there is a seemingly small window of time where I can act and make a difference in the world. In this case, I would likely contribute in a generalist capacity, helping run events, support new organizations, and generally make things happen, which typically doesn’t prerequisite a college degree.\n\nAt any given time, I’ll likely have multiple projects/responsibilities I’m actively leading or contributing to. Throwing school into the mix heavily constrains my time, and adds unnecessary complications (e.g. going to required sections, studying for exams, completing psets) that limit my capacity for high-quality work.\n\nAdditionally, the AI safety community largely convenes around Berkeley and DC, meaning being in Atlanta could be a major downside. Thus, dropping out maximizes the time and resources I have to work on AI safety and minimizes external distractions. **This first reason feeds into most of my other rationales for dropping out of school.**\n\n**AI Safety is ****generalist ****constrained****. **It seems to me that one of the biggest (current) bottlenecks for impactful safety work is the lack of generalist talent in the field. I’ve built the relevant skills by helping organize [various](https://www.aisi.dev/) [organizations](https://secondlookresearch.com/) and [programs](http://dcminiconf.com), and could meaningfully contribute to neglected work without a college degree.\n\n**My university’s current curriculum is poorly prepared for ASI**. At the macro scale, I’m concerned that my university’s current approach to CS education (which is very application based) won’t scale well to post-ASI worlds. Completing my degree right now will only partially contribute to a strong, foundational understanding of CS, and would generally teach me less valuable skills than what I could learn while working on AI safety directly.\n\nI would probably get a better CS education by deferring my degree a few years into the future, with advances in education helping me learn more (this is a weak guess, not a strong prediction).\n\n**Personal Motivations.** I’m personally biased towards being in California since it allows me to be closer to home and spend time with my family and friends, while maximizing my impact in AI safety. I don’t find much enjoyment being in school (apart from my friends), but this doesn’t mean working on AI safety full time will bring me the enjoyment I need - this could be completely orthogonal to my career/environment.\n\n**Arguments against dropping out**\n\nNow some strong reasons for staying enrolled in school. The biggest drivers pushing me to stay enrolled are {Maturity, Relative Competence, Current Responsibilities}.\n\n**I’m not intellectually mature enough to meaningfully contribute to AI safety. **While I currently believe in short timelines, upon further reflection I may see more merit in longer timelines, or at least start hedging with longer bets. I think I have a reasonable grasp on the core arguments of AI safety, and some of the key technical agendas and policy interventions – however, I wouldn’t say I’ve mastered them, and I think I definitely have a lot to learn.\n\nThis goes beyond just AI safety knowledge too. I’ve only completed my first year of undergrad, without taking highly challenging courses. There’s an argument that AI safety is a really hard problem, and that I haven’t developed intellectually enough to meaningfully contribute in the long-term. College is a great environment for learning the fundamentals and cultivating a deep understanding of anything, and it could be valuable to spend time studying the core questions of the field, rather than rushing into short-term work that may or may not lead to long-term impact.\n\n**My current responsibilities require ****me**** to be responsible for them. **Dropping out means leaving more than school behind - it also requires me to give up organizing at GT, which I perceive as very important and high impact, as well as an opportunity for me to grow. The only situation to give this up in is if the potential for growth and impact is greater at other opportunities, and if I trust my successor to do as good of a job at maintaining the organization without me. GT is a somewhat experienced organizer constrained at the moment, so I value my presence at the club and think it's definitely worth my time to organize this semester.\n\n**Life experience**. University teaches you more than just an academic curriculum, but rather instills important life lessons and teaches you things you can’t necessarily learn elsewhere. I’m uncertain whether or not I’m actually mature enough as a person (I’m only about to turn 19) to actually lead my life outside of university, and whether or not this will be critical for actually contributing to AI safety. Of course, I can learn a lot of things outside of university - however, I’m uncertain about how much time I’ll have to really focus on my own personal growth over my work.\n\n**Credibility.** While dropping out to work immediately seems reasonable in short-timeline worlds, we might be living in moderate-timeline worlds where spending a couple of years investing in a degree is critical for maximal impact down the line. While the AI safety community as a whole is highly meritocratic, the same can’t be said about the entire world, and some of the highest impact careers (such as working policy) fall outside the sphere of pure AI safety work. Graduating holds weight, and this is important to consider in worlds where timelines are longer.\n\n**Personal motivations**. I care about receiving a higher education - this is personally important to me, and something I eventually want to achieve. I enjoy being in college sometimes, especially when I’m just spending time with my friends. My parents would be happy that I’m working towards my degree. There are benefits, which include having a dining plan and a gym and working space, that just don’t come without overhead in the real world - of course this costs money, just like in real life, but with much more simplicity.\n\n**My broader conflict between upskilling and taking action**\n\nI often feel the need to act quickly, do things, be agentic, etc. in order to be impactful - timelines are short, and we don’t have the time to sit around and think about what to do. Rather, we need to act ASAP!!!\n\nOn the other hand, I feel the need to upskill - actually understand the arguments, motivations, threat models, and make sure I get what I’m actually working towards. I suspect I have a fuzzy sense of this, but I don’t know if it is too fuzzy or if I am even a good judge of this.\n\nI recognize that these two aren’t mutually exclusive, but there's probably a tradeoff at play where I can’t invest fully into one without losing a major part of the other. In some ways, I think my internal debate on whether or not to drop out of school is a manifestation of this same conflict.\n\n**Uncertainty**\n\nI’m really uncertain about a lot of the writing in this post, and would really like to hear your thoughts. This is a really important decision for me to make, and I don’t want to take it lightly. Post-writing, I still feel confused about what to do, but I think I have much more clarity on what factors are really at play here. My current plan is to come up with a plan by 2027 - by the end of this semester, I’m confident I’ll know what decision to take.\n\n[Discuss](https://www.lesswrong.com/posts/CfqLNt6jgb9xrixf4/arguments-for-and-against-me-dropping-out#comments)", "url": "https://wpnews.pro/news/arguments-for-and-against-me-dropping-out", "canonical_source": "https://www.lesswrong.com/posts/CfqLNt6jgb9xrixf4/arguments-for-and-against-me-dropping-out", "published_at": "2026-08-12 01:12:57+00:00", "updated_at": "2026-08-12 01:39:17.745213+00:00", "lang": "en", "topics": ["ai-safety", "ai-policy", "ai-research", "ai-startups"], "entities": ["GT AISI", "Second Look Research", "AI Lab Watch", "DCMC 2.0", "Zephy Roe", "Ishan Khire", "Naren Manikandan", "Anish Kallu"], "alternates": {"html": "https://wpnews.pro/news/arguments-for-and-against-me-dropping-out", "markdown": "https://wpnews.pro/news/arguments-for-and-against-me-dropping-out.md", "text": "https://wpnews.pro/news/arguments-for-and-against-me-dropping-out.txt", "jsonld": "https://wpnews.pro/news/arguments-for-and-against-me-dropping-out.jsonld"}}