Ask HN: I feel pessimistic about the future – advice? | Hacker News Reader

Ask HN: I feel pessimistic about the future – advice?

43 pointshnuu333 comments

I’m having a hard time moving with the new pace and developments in the world.

While AI is semi-interesting, I worry about the long term existential risks to humanity with respect to losing agency, reasoning ability, humanness and creativity - similar to what has happened to our attention and mental health with social media.

I’m usually quite fiscally conservative, pro technology, start-ups but in this case, I’m worried I’ve reached my “event horizon” technically speaking, but I’m barely halfway into my career.

I can see how AI can be influential I just can’t seem to get on the train technically — it just doesn’t interest me. It feels like it’s just slop ontop of slop.

It's cool to ask Claude or Cursor to code something, but I feel less emotionally invested in the craft of coding.

Yet, it feels like we’re reaching this “sink or swim” moment, similar to what Tobi from Shopify put out in his blog.

Any advice here?

View on HN

Stop listening to the hype and doomers. Both of them are wrong. Go outside, live your life, build whatever you want to build, spend time with family. Stop reading stuff by people you don't know. Everyone online is trying to sell you something. Many sell fear.

We're gonna be fine <3

Yep

You live once. Maximize your own time with yourself instead.

> Go outside.. spend time with family

These two. Go take a walk. Disable notifications on as-many-as-possible-apps (especially news & social). Look at your phone less. Look at your friends and family more. Enjoy a hobby.

go outside: everyone on their phone and you can't strike a conversation even at a bar.

Yep it's tough to be optimistic. I like to think about a micro-meteorite falling on my head. It's possible at any moment, and yet I always ask myself why I'm not worried about it.

The best advice that anyone can give you is to stop looking at social media, or reduce it by 90+%. The kind of "depression" or "despair" you express is a symptom of doom scrolling.

My next advice is a question - If you can't enjoy today, why do you think you would enjoy tomorrow, even if everything in life were perfect? "One day at a time" is really good advice. Do something tangible to mark your successes or failures on this one single day, and then rinse and repeat tomorrow. If you are "failing" too much, then recalibrate your expectations until you can mark several points of success, no matter how small. "I read my work email" or "I did zero youtube".

My non-expert prediction is that LLMs will ruin the youth's coding skills. Anyone in a deep dev discussion will have to have instinctual knowledge that LLMs can't teach, and their value will be apparent when non-trivial tasks arise. And remember, a micro-meteorite could fall on our heads at any time!

The thing I like about software dev is that as I get better at it I get better at thinking and problem solving.

I use AI rarely, but I use it to try to learn. For example, gnuplot is pretty complicated, so I'll ask it to create some examples and that's allowed me to learn way faster than trying to parse out the man page. AI is great for this, even though that's not how most people use it.

Some parts of software dev will likely be automated away (just like compilers automated conversion to binary and some parts of optimization), but I think ultimately software dev will become more important over time. Natural languages have low information density and high ambiguity. Programming languages are the opposite. So I think programming will always be a thing.

Think about music. It's not written in natural language, even though it could. We continue to use musical notation because it is the best way to communicate music. I believe the best way to have unambiguous, information-dense, logical communication with a computer is with a programming language.

I’ve been trying to contribute to a Linux kernel subsystem recently. I can program in C, but that’s not really the barrier. Most patches are pretty tiny changes to the C code, I think. At the moment I’m finding that I don’t understand enough about the subsystem and hardware - that is the main obstacle to making meaningful changes. I’m not sure if there will eventually be AI agents that work on the Linux kernel without close supervision, but the “craft of coding” was probably never the most important thing.

Something I heard on a podcast recently re. AI is: do you think computer science, machine learning, embedded systems etc. will become more important to society in a world of advanced AI or less? I think it will become more important, but I suppose that doesn’t necessarily lead to more jobs.

Keep this in mind: The future doesn't care about our feelings. It's going to happen regardless of what we think.

There is a ton of uncertainty and rapid evolution happening and it's okay to not have an opinion, reserve judgements, and generally float along until things normalize a bit more.

AI will be used against our interests. It will be weaponized against us. There's no avoiding that. Yes, the existential risk to humanity will probably come along at some point, but humanity has a vested interest in avoiding that.

I'm not saying everything is going to be okay. There's going to be some tough challenges, but at the same time, I don't think the AI genie can be put back into the bottle.

Be thankful that not everything turns out the way you want it to...it means you're not trapped by the limits of your own vision. And recognize that the future is just imagination. Stay open to the possibilities..the universe is not closing in on you, it's opening pathways you've yet to see.

Focus on yourself. Make you better today.

If you like to code and the patterns of using AI to code are bumming you out, break that down a bit and change your patterns. Use AI to teach you instead. Use it to write the mundane stuff to let you concentrate on the fun stuff. Be the architect and lead programmer and let it be your intern.

The worst thing any of us can do is give up our agency and hope for tomorrow.

You should unplug more and vary your news diet.

I always feel at my best when I spend some time outside, going for a walk, sitting on my back porch sipping my morning coffee, reading a good book, or getting in a good workout. If you're like most people, you're spending way too much time online and looking at a screen.

It's also important to consume news from a variety of biases and view points. Unfortunately completely unbiased news reporting is very difficult to come by, so the best alternative I've found is to balance things out. There's a lot of news stories that each side suppresses or misleads on to fit their pet agendas. You'll find that a lot of media hype is meant to scare you into continuing to follow their big stories and giving them your eyeballs.

Finally, control what you can control and make your peace with everything else that you can't. Once you come to terms with the fact that nothing is permanent, you can find comfort in that.

I don't know but it seems like LLMs are just adding small incremental improvements in the polish/iterating phase.

Some new revolutionary thing will probably happen in the future but it might be decades away. Heck AGI might be even harder to solve than fusion which has been 30 years away since even before I was born.

We're probably close to an AI winter. There's been a huge investment (over $1T iirc) and not that much to show for it. The euphoria of the past couple of years is definitely over.

I do think there are some very useful applications for LLMs... just not the ground breaking cancer solving thing we've been told again and again that would justify that much money thrown at it.

> I don't know but it seems like LLMs are just adding small incremental improvements in the polish/iterating phase.

A couple days ago I was searching my hard drive for <something> and found this java file I downloaded a long time ago. This long, lost project idea from days gone by I never got around to doing anything with though I do remember making an attempt.

So, I got the robots to work. They analyzed the code and answered a whole slew of questions about how it compares to 'modern' implementations. They converted it to C, made improvements (which they also suggested) and added functionality (which I suggested) to the original code then wrapped it all up into a gimp plugin. If I ever get around to installing gimp-devel (or whatever the Fedora package is) I can get the bugs shaken out and upload it to the gimp plugin registry -- as was also suggested by the robots.

And this was all over the course of an hour or two until, quite honestly, I got hungry and went foraging for food. I would never spend the days to learn how gimp plugins work and, most likely, would just have just let the file sit on my hard drive for another decade before even looking at it again.

I have to disagree they are only good at small, incremental improvements. There's a bunch of papers I've collected over they years which only have the 'algorithmic code' I plan on letting them loose on which I either tried (and failed) to convert into actual code or didn't even try at all.

You're picking probably one of the best use cases for LLMs though.

At this point you'd think anything related to language is a solved problem and yet my wife is a translator who spends way too much time correcting machine translated text.

And what about other realms outside of language like self driving? How long until it is generally available?

Have you heard the pedestrian crossing recordings? It's a spoof, but the advice is spot on: "you don't need to worry because there's absolutely nothing you can do to stop it."

If you're fiscally conservative, you should be fine; just keep on keeping on for now. The sky has been falling for a long time, and maybe one day it will, but take reasonable actions given the information you have, and accept that you might look back with more information and see that there were other actions you could have taken that would have played out differently. c'est la vie.

You don't need to board every hype train. If it works out, you'll be able to catch up later.

I've felt this way for some time since Covid. I've been waiting for the proverbial other shoe to drop, and I think we're seeing that now.

There's public acknowledgement now of a housing bubble, tacit admissions we're probably in a recession, the "AI" Bubble, and more.

Pair this with the current political instability in the States, and it's gonna get ugly.

I feel less hopeful about the near and long-term future than I have in a long time.

> I worry about the long term existential risks to humanity with respect to losing agency, reasoning ability, humanness and creativity

You are going to be dead long before this happens.

You've probably got around 2350 weeks left to live (give or take) if you are very lucky.

Do you really want to spend them worrying about something that you have very little personal control over?

Figure out what you do want to do with your time and do that. The time is going to pass anyway.

My general advice is to be aware when you're in the grief cycle and get out of there as fast as possible so that you can think clearly. A lot of people in the comments are in denial and bargaining. You sound like you're at depression... almost out of it!

But things will continue to happen. Some people take years to get out of denial or anger. In this period, they do really dumb things. Things will get worse. Which causes more grief.

Catch yourself. Let the emotions flow, but be aware of the process. Maybe you choose to fight it, but try to get into the Acceptance phase first, where your anger can strike like a hammer, instead of a river.

A lot of the pro-AI people have been there too. Kent Beck wrote a nice article on his experience: https://tidyfirst.substack.com/p/90-of-my-skills-are-now-wor...

What you need to do is unsubscribe entirely from all social media and stop reading news first thing in the morning. You have to remember that "Tobi" can't tell a pointer from a hole in the ground, and neither can Sam Altman. Nor can they tell you how any of the stuff they "build" actually works, or what its real capabilities are. Put simply, they're talking out of their ass.

I've been in this career for almost 30 years and there was no point in my career at which developers weren't going to "go away" real soon now (tm), replaced by "no code", or "visual programming", or those UML diagrams, or XML (remember that one?), or whatever the fuck. And yet we're still here, in greater numbers, and we make the most money you can make outside of Big Law and Wall Street. Something isn't lining up? You bet. You're probably just too young to have perspective.

This isn't the only field in which the claims of demise have been greatly exaggerated by people who have zero clue. The people who create this hype hope it will land on the younger generation which is less immune to it and/or has a shorter attention span. Preferably, on people who haven't been through a single major hype cycle before. Folks who have been around for a while eventually learn to smell at least some kinds of bullshit from pretty far away.

Full disclosure: I'm a researcher in this field (though not in coding assistance per se). Your profession is not in danger in the foreseeable future, assuming you do something that requires actual intelligence. Your tooling will get a lot better though, so you'll have to type a lot less boilerplate. But I'm assuming that's not the part you take any particular pride in anyway. I certainly don't, so I use ChatGPT quite liberally for things I'd previously use Stack Overflow for.

Stop reading the news 24/7/365. You can't change what premier/president "John Smith" is doing.

Save cash. Invest (every month) a 'small' fixed amount. So when it goes high, yeah!, when it goes low, you will buy low and lower your average cost of acquisition.

Speaking of acquisition, read the Rules of Acquisition (https://projectsanctuary.com/the_complete_ferengi_rules_of_a...)

Read "The Daily Stoic: 366 Meditations on Wisdom, Perseverance, and the Art of Living". Read the 'entry of the day', think about it, then journal in the EoD. THIS will change you (to the better imho).

Don't worry this wave of hype will calm down just like all others, and we'll end up with few new useful tools in our belt.

You shouldn't have been invested in the _coding_ in the first place. Engineering is the real craft no LLM is capable of. Spitting out compilable letters is just one of many tools.

With all due respect Tobi is not an engineer.

My expectations of AI in the next 12-24 months:

FAANG and other large companies are integrating or will integrate AI vertically. A couple of possible scenarios:

1) Executive asks AI to give him some insights, AI talks to him, shows him a few possible clarifications, executive picks one. AI then creates tickets for downstream teams with descriptions and deadlines.

2) Some senior data analyst receives the ticket, and asks AI to build a query. With a bit of discussion they build the query and AI scheduled automatically with a dashboard attached. Analyst click OK and AI closed the ticket, tagged the executive and send him the link

I don't know what's going to happen after 24 months. Maybe AI stays at the same level and FAANG team quits with big bonus and bring the experience to other companies. Maybe it levels up again. Either way, bad for you and me.

I do believe the next two decades are going to be more chaotic and more jungle like.

> but I feel less emotionally invested in the craft of coding.

I don't think caring about the craft of coding and using AI tools are mutually exclusive options. I think you can totally do both! In a way, some of these tools actually allow you more space and time to "care about the craft" because they can take care of all of the mundane boilerplate for you.

A man once said "Bad programmers worry about the code. Good programmers worry about data structures and their relationships."[1]. Well IMO these tools allow programmers the time/space to care about data structures and their relationships -vs syntax and such.

[1] https://news.ycombinator.com/item?id=25501427

>It's cool to ask Claude or Cursor to code something, but I feel less emotionally invested in the craft of coding.

So don't. Just write code yourself. I promise you some of us are still doing it.

Be aware that no matter how the AI stuff ends up playing out, there will be a strong demand for "old school" work for a very long time. Likely long enough to support the remainder of your career.

If you're not interested in a particular segment of the field, go to a segment that you are interested in. Depending on what your personal goals and desires are, it's not necessary to follow along with whatever the current hotness happens to be.

like mcdow said: stop listening. everyone has an agent specially those in power.

- focus on yourself - believe, whether you want it or not, is a powerful tool... therefore so it's denial. - believe in things you find useful for you - focus on the present, not the history, not the mystery

open your mouth, wider, wider, good, now say ahhhhhhgg, good, close you right eye, yes yes, loook to the left, good, now close the left eye, good yes look to the right, fine, well, not so fine but not so bad....,you seem to be suffering from immanence, which as a male you are not imune from, women, born with the phsycological wetware to deal with child birth can handle any amount of immanence, but men must stay busy, and actualy initiate and do stuff, or they will get infected with external soonness, and loose there timing and initiative. It quite litteraly does not matter what you do, anything at all is good, pick lint, build an airplane, get a degree, walk a lot, irritate people on the inter.,er,,er...ok not that, as it's immpossible not to do, but you get the idea? right

Build the future you'd want to see

these things take about a decade to play out - reminds me of the dotcom bubble .. the opportunities there for those who persist.

Do you feel good about your mental health? No need to answer, just consider that such issues could be the cause for the anxiety about the future. In that case, therapy is recommended.

My take on the whole thing is quite different than yours. I can't relate, therefore I cannot give heartfelt advice.

Also, take a break. HN is full of noise.

I noticed an existential risk for a software development career around 2015. I saw then that software development was becoming commoditized and that the market was becoming saturated with “good enough” developers that could do the CRUD/framework development that most companies needed. AI has accelerated the trend.

Back then my concerns wasn’t getting a job, it was compensation plateauing at best or being a race to the bottom at worse. I was right.

Before anyone mentions the eye popping compensation that the BigTech and adjacent companies are paying, those positions are just a small subset of companies out there and most of the 2.8 million developers in the US will probably never make over $200K inflation adjusted in their entire career doing “enterprise dev”. (Yes I did a stint at BigTech).

I see I wasn’t wrong, in 2023 and 2024, when I was looking for your bog standard enterprise dev jobs with AWS experience remotely as a plan B, “senior” [1] dev jobs were still offering $150K - $160k in most major cities and “architect” roles were going for $175K - $180K. These are the same numbers I’ve seen since 2016-2018.

What’s even worse, every job I applied to had hundreds of applications (LinkedIn show you) and I heard crickets.

I pivoted into working for cloud consulting companies full time (and before that working in the consulting department at BigTech) mid 2020 and found a full time job at a 3rd party consulting company.

That being said, projects that I would have scoped as a two person project with my leading it and adding a mid level person to do some of the grunt work coding, I now know that I have the bandwidth to do myself with the help of AI.

That also means that the company as a whole needs fewer developers. But still needs just as many customer facing people who can fly out to customers site and be on calls gathering requirements and pushing things through the sales cycle.

The solution? Quickly move up and become a true “senior” as defined by every tech company with leveling guidelines.

https://www.levels.fyi/blog/swe-level-framework.html

On a side note, you don’t need to “learn AI” as far as how it works under the hood, save that for the PhD’s. You need to learn how to leverage it for coding, writing, and how to create features and products that add business value. I have my own niche within AWS that I had before LLMs were publicly available where I use it in for now novel ways when I need to.

I don't have any advice, either technologically or socially. I have what I think solutions are but society doesn't like my solutions, and we live in a society.

1. Maybe read some Pessimist philosophers. I find solace in cosmic entropy. I don't believe in a literal God, but Mainlander's idea of pandeism where god killed itself and the cosmos is his "rotting body" helps me gain a cosmic spin on things.

2. The older I get the more I realize humans are really dumb people. Capable of some amazing things, and yes, "wisdom of the crowd" can be right at times, but it can be very very very wrong. And "really smart people" who lead the charge can really fuck shit up. Technologically. Socially. And more.

3. If it helps, I literally just added a filter to ublock for hacker news to strip out Startup and AI keywords, it's not the best filter (I could probably remove some non-capitalized words)

this might help reduce at least having to see it, and get back to what you WANT to see.

I want to see actual coding and language theory and os stuff. But 1/3 of the content here is AI now, and it's a huge turn off. It's noise to me, so having a filter to get what I WANT out of this site, is better than me turning it off completely.

I wish I had an answer for you in terms of mental state, I don't think there's a way out of this rut, and I think it's only going to get worse from here, and the pollyannas wanting you to pretend like nothings wrong... Well. IMO that's why we're here. Ignorant of history (actual history not the pop history everyone gets in school) and I don't mean this from one perspective or another. I've read Fascist literature, Capitalist Literature and Communist Literature. I've seen scenes play out on all sides that I despair from.

I don't mean to bring you down, but the best I've got is the above (and maybe microdosing psilocybin which helped me yesterday. I was stuck in a mental pit, and 666 mg psilocybin chocolate bars helped lift me just enough so I was able to see with clarity, give me the serotonin boost I needed and step back and observe the "snakes"/selves all writhing - negative aspects of myself, why they exist, how I anchor myself to angst, because it is a grounding. If I have no angst, I have nothing in this world but chaos.

Angst ist meine heimat. But being able to step outside and see what it is and why and how... I don't say you should do that, psychedelics are not for everyone, and they can fuck some people up, and some people get the wrong ideas from them (looks at the hippie->fascist pipeline).

------

Speaking of coding, and finding reasons to do it - I'm continually fighting my own ADHD tendencies to make some really dumb stupid little projects that don't matter to anyone, but to me, to say "I COULD DO THIS". I will never get rich off it. I will never even make money off it, but to know I was creative. That is about the best I got.

One little ember flickering before it dies off with the rest of the charred husk of this planet.

Sorry so long.

Reply on news.ycombinator.com