Don't die

8th October 2025

I unironically intend not to die.

This has been my goal since I was a kid - specifically, since I lost the faith as a 15yo, and finally realised - and emotionally processed - the magnitude of the problem of death.

From talking to others throughout the years, I believe almost no one goes through the actual motion of deeply internalising the fact that they might - indeed, will probably - die. Admittedly, I did not have almost any contact with death in my personal life, apart from a few of my old relatives passing away in the periphery. But any conversation I, be it with friends, family, strangers in real life or online, slides into cope. Learned helplessness of not being able to do anything to fix this is the dominating factor; making people either completely avoiding the thought, or focusing on religious escapism.

There are many obvious things to be said here, and usually anyone who seriously considered the problem has reached the same conclusions. Does dying give meaning to life? Maybe, but it's foolish to bet on this without thinking about it deeply (say, at least for a few centuries), and being smarter (say, Bertrand Russell being the lower bound). Won't it be boring? No, because there will be new possibilities in the future. Wouldn't it be better to die young when still energetic? No, because curing death probably means curing aging and all of its related symptoms. What if we are actually immortal alrady because of buddhism/quantum immortality/simulation hypothesis/christian afterlife? Great, but it's still better to have an insurance policy, and anyway, confirm that it is indeed the case.

I am now 30 years old.

My life so far has been to a greater or lesser extent steered on a macro-level by the goal of not dying. For any major life decision I took, I asked myself whether this will increase the probability of not dying. In chronological order, it has been:

• Enrolling into CS undergrad at UCL in 2014 to study the brain at Gatsby afterwards, becoming disillusioned with the academic level there, cold-emailing every good professor I could find what I should do, and dropping out to get back to study pure maths & CS back in Poland.
• Deciding that maybe ML, rather than neuroscience, is the way to go, and getting employed to work on NLP machine learning in 2015/2016 just before Transformers - later becoming disillusioned with the purely empirical chaotic trial-and-error.
• Getting more and more into mathematics to really properly understand the foundations, and finally working on logic/HoTT. I ended up doing two bachelor degrees end-to-end plus a masters in pure math, which, together with a short break in the middle from burn-out, took me a total of 7 years.
• Then, during COVID, repeatedly cold-emailing famous people working on life extension to get some career advice and either getting no answer (Sarah Constantin and others), or getting told explicitly that my skillset in math&CS is useless and I should pivot to biology and wet lab (Audrey de Grey). In 2020, things looked bleak for immortality - I thought that the default is dying almost surely. I was lost. I started studying medicine, but dropped out quickly.
• My relationships have alwasys had the undertone of being transient, because none of my partners shared the goal of not dying. From what I read, it seems that having children overrides previously-held goals, which is pretty much the worst outcome imaginable here. On the other hand, conflicting evidence shows people get more productive or motivated. I erred on the side of being cautious.
• I wanted to pursue PhD - mostly as a side goal and something I always wanted to do in life, but also to open the doors to a research career. Having missed application deadlines, I got a job at a hedge fund to have some financial cushion.
• In the meantime, things with AI have sped up in early 2020s. It seemed that maybe AI progress towards alone will bring about general scientific progress. I had some background in programming language theory and pure maths, knew about AI safety from LessWrong for quite some time, so decided that the best way forward - both maximizing probability of getting an offer, and moving towards career path I intended to take - would be to start working on probabilistic programming and AI safety at Oxford. My goals hinged mostly on capabilities, but in 2021, it seemed that safety was sufficiently small that it's Pareto-efficient to work there.

Since then... I didn't manage to do much. My idea of PPLs for AIS, who no one pursued at a time, went mainstream with UK's ARIA getting big on Safeguarded AI under davidad (who I procrastinated on contacting for two years and ultimately didn't really get involved with). My papers mostly turned out to be tangents, with no practical results. Maybe it's because I wasn't actually into AI safety that much. Maybe it was because I was depressed and burned out for quite a long time in my PhD, with a spectacularly painful breakup early on. Many reasons. In any case - last year it seemed that there will be a break-through with our team getting a huge grant from ARIA to set up a new research lab on the ML for safe AI. This had been cancelled literally last-minute.

Now, AI progress rate predictions seem to bifurcate, with either spectacular wins in the next few years, bringing about the AI safety risks, or it stalling to a halt. Prediction markets show about 50% chance someone born before 2000 will live to be 150, depending on the question.

I am not sure what to do next.

LessWrong used to be a place where people, but it got eaten alive by AI safety. I suspect old Extropian crowd, Nick Bostrom or Anders Sandberg or Robin Hanson still share the goal of not dying. But they are not very active about this. There are some scattered Twitter accounts still going on. (I intentionally dismissed Bryan Johnson's flavour of Don't Die, as it focuses on a very narrow slice of being-healthy-until-escape-velocity, which doesn't seem to be very agentic about this.)

So, what's the play here? Is it to:

• Pivot to quant, earning lots of money, hope AI progress doesn't stall, and spend it on meticulously testing for cancer?
• Work specifically on the longevity tech? How? With whom?
• Try to set up a new AI startup focusing specifically for that?
• Re-learn biology and try to contribute as a scientist?

I know there are all these prosaic, low-hanging fruits of being healthy, getting cryonics insurance and all that. But I don't know what to do in a macro-sense.