My wife got an email from a new hire (now even a new hire yet: she's still on a trial basis), a 23 years old, where she explains that she doesn't want to use AI. That she doesn't like what AI does. On a funny sidenote: the email is obviously 99% llmish, which is hilarious.
That's one extremity: crazy people who refuse to learn a new tool.
Then on the other extremity you have the even much crazier ones: those who believe they've got an intelligent machine that is going to solve all their work problems during the day and then, at night, that is going to enlighten them by revealing them who god really is.
Where the heck are the reasonable people who use AI for what it is: a tool that can be extremely helpful at times and extremely sucky at other times but that is still, on average, a time saver?
I use LLMs daily to piece together technical reports and smooth out rough drafts. It saves me hours of time / week.
I also use it to augment my technical work, because I don't want to be out of a job one day with no marketable skills, except driving an agent harness.
There is a percentage of the population that thinks LLMs are actually intelligent and truly can't tell the difference.
I think others just want to live life as a passenger, not think, and have AI do all the work.
others just want to live life as a passenger, not think, and have AI do all the work.
Fuck, doesn't everyone? The future we were promised is one where machines did all the work leaving humans free to pursue a richer existence, whether that's creative pastimes or just laying on a beach.
Instead of democratizing the future, essentially every technological advance since the printing press has served only to increase the concentration of wealth and power at the top.
I can't imagine actually wanting to work if you didn't have to.
Lots of people enjoy doing things - no need to treat them like assholes just for that. The sci-fi version of AI frees people up - the real version sometimes (mostly?) takes away enjoyable work and reduces people's ability to make a life for their family without being miserable (which is the fault of the practical reality of our society - not inherently AI's fault).
a tool that can be extremely helpful at times and extremely sucky at other times but that is still, on average, a time saver?
I can tell you that while I use AI, it's basically a compressed version of stackoverflow to me. I don't bother with agents because I simply don't have that much code to write. The nature of code didn't change. Less code is better code, and it's a strong smell when people claim that they needed an agent to save time.
When I do have a new project, there's an extremely high chance that I can reuse existing code from another project that already has passing tests, or I am experienced enough to know what exactly needs to be done. I can probably finish my part within the sprint or faster. The agent can only introduce risk to the project timeline or code quality.
This isn't me being arrogant or "anti-AI". The hard part of software development has always been collecting consistent requirements from various stakeholders. I'm pro-AI in the sense that, when used correctly, it helps junior devs learn and otherwise shines a spotlight where the real bottlenecks are (project management and delays).
Where the heck are the reasonable people who use AI for what it is: a tool that can be extremely helpful at times and extremely sucky at other times but that is still, on average, a time saver?
They're unlikely to be vocal because they've already evaluated and decided the role AI will play in their work/life and just moved on/kept working. That's also likely to be a small pool of people relative to the number of people interacting with AI (either by force or choice).
That's also likely to be a small pool of people relative to the number of people interacting with AI
Perhaps small now. But overall this seems to be what most have done in my small social circle. Small uses of AI here and there, sometimes surprising gains/wins they find out and share if asked, otherwise really it's just another tool to learn and make use of.
Most everyone thinks it's overhyped and wedged into stupid places it doesn't make much sense, especially in consumer products. But they also can recognize that behind the scenes and in closed doors there are perhaps some exciting things happening in certain industries and use-cases.
Most are still very much in the "haven't played with it much yet, it's moving too fast and I'm too old and busy with my life" or "still evaluating as time allows" stage. This is mid-career folks in various professional roles, skewing towards tech. Some sense of "this tech is getting good fast, and I will be left behind if I do not learn it at some point in the future".
I will say most have barely ventured past the "copy/paste from ChatGPT" stage of AI use. It usually elicits comment when someone moves past that and finds out they are far more capable than they realized. Then usually some more comments a couple weeks later mentioning newfound limitations.
I count quite a few of my colleagues into this observing and must learn role. My problem is that most entry use cases that are advertised are not cutting it for me.
Write overlong emails that could have been a short hand written one? Write longer prompts to get shorter LLM email responses?
Create presentations that screw up the corporate theme and send me fixing it?
Write power automate flows that send me chasing errors?
Particularly management and leadership positions are urged to engage with LLM tools but are drowning in the noise of busy days and flood of communication. Summarising emails that then misses the crucial information is not saving any time.
Summarising emails that then misses the crucial information is not saving any time.
Ironically due to an overloaded inbox these days, I've been using a LLM as the world's most inefficient e-mail filter.
It does a decent job at this, and gives at least a nice starting point in the morning for a prioritized hitlist to power through. Gets most things pretty accurate, but I only use it to sense "importance" and not much else.
Certainly not life changing, just a nice to have.
I've had far more real gains with home IT projects. Tons of home automation integrations and "silly" projects I've been wanting to get done for a decade but never really had the energy to want to deal with. These sorts of things also allowed me some time/excuse to play around with agents to see what the fuss was about.
I now have the world's most complex garden irrigation/sensor setup complete with automated watering and grafana panels! I could have tied everything from the hardware through software together myself, but I just was not interested in installing the 856th Ubuntu VM in my life, configuring InfluxDB from scratch, etc.
I can see the future potential due to the above, but I also see how it went off the rails easily and also produced some laughably inefficient/unmaintable stuff like Flux queries which toss what should be variables as statics directly into the grafana panels, etc. Add a single new sensor to a group and you'll be re-writing the panel queries. Just horrible ground level design only maintainable via brute force. But when you just tell it to add the thing and it does... I can see why many feel it's a superpower.
For personal use I find "AI" extremely useful and fun.
I asked Claude to create a mini stack that email me a daily digest:
Today's weather, rain forecast.
Top ten local news,US news, world news
New HN articles, bubble up articles about bicycles, crafts, Japan, exclude Rust, Zig, Python, JS, Node, Bun.
Summary of articles in my favorite subreddits.
A separate email summarizing new antique items on ShopGoodWill.com, with their starting prices, auction end dates.
Oh, and it configured Proxmox on my NUC and installed and configured the whole *arr stack.
Now, I am a techie, so I know what to tell it and could do all of the above myself with enough elbow grease, but I never would have. All of the above took under an hour.
Just this morning, I created a plugin so that Claude speaks to me as Samuel L. Jackson, or Arnold, etc. Fun fun.
Replaced 90% of my browser web searches. I don't know how many trees I burned and lakes dried.
We're right here! Some of us just prefer to stay out of these silly irrational debates because we know that neither of those two crazy camps will listen to a word we, or anyone else, will say. We might be wrong about that (I hope we're wrong about that!) but there sure are enough other people chattering that it's easy to sit back and let them have it out.
The "you'll get left behind" crowd has already made it clear that they're not worth engaging with, you're not going to find reasonable voices anywhere those threads are happening...
We're here -- I use AI for low-complexity toil, one-off scripts, search & discovery, and as a sounding board for new perspectives.
I still design and build systems almost all myself. If a snippet comes out of an LLM I'm keen to use it's filtered through my brain as well. Polished and transformed to align with my vision. After all, I'm on the hook for supporting it and bringing it to my team.
I'm sure most of us are just sick of the mania around this and choose to avoid interacting with content relating to it.
It's very useful. But irresponsible to use without recognizing and respecting its limitations.
Comments
Where are the voices of reason?
My wife got an email from a new hire (now even a new hire yet: she's still on a trial basis), a 23 years old, where she explains that she doesn't want to use AI. That she doesn't like what AI does. On a funny sidenote: the email is obviously 99% llmish, which is hilarious.
That's one extremity: crazy people who refuse to learn a new tool.
Then on the other extremity you have the even much crazier ones: those who believe they've got an intelligent machine that is going to solve all their work problems during the day and then, at night, that is going to enlighten them by revealing them who god really is.
Where the heck are the reasonable people who use AI for what it is: a tool that can be extremely helpful at times and extremely sucky at other times but that is still, on average, a time saver?
You might not be hearing from these people.
I use LLMs daily to piece together technical reports and smooth out rough drafts. It saves me hours of time / week.
I also use it to augment my technical work, because I don't want to be out of a job one day with no marketable skills, except driving an agent harness.
There is a percentage of the population that thinks LLMs are actually intelligent and truly can't tell the difference.
I think others just want to live life as a passenger, not think, and have AI do all the work.
Fuck, doesn't everyone? The future we were promised is one where machines did all the work leaving humans free to pursue a richer existence, whether that's creative pastimes or just laying on a beach.
Instead of democratizing the future, essentially every technological advance since the printing press has served only to increase the concentration of wealth and power at the top.
I can't imagine actually wanting to work if you didn't have to.
Lots of people enjoy doing things - no need to treat them like assholes just for that. The sci-fi version of AI frees people up - the real version sometimes (mostly?) takes away enjoyable work and reduces people's ability to make a life for their family without being miserable (which is the fault of the practical reality of our society - not inherently AI's fault).
Hmm, I wasn't trying to imply anyone is an asshole for enjoying things. Is that how it came off?
We're everywhere. It's extremely easy to trick yourself into thinking that somebody lamenting AI doesn't use AI. AI is an enormous category.
I can tell you that while I use AI, it's basically a compressed version of stackoverflow to me. I don't bother with agents because I simply don't have that much code to write. The nature of code didn't change. Less code is better code, and it's a strong smell when people claim that they needed an agent to save time.
When I do have a new project, there's an extremely high chance that I can reuse existing code from another project that already has passing tests, or I am experienced enough to know what exactly needs to be done. I can probably finish my part within the sprint or faster. The agent can only introduce risk to the project timeline or code quality.
This isn't me being arrogant or "anti-AI". The hard part of software development has always been collecting consistent requirements from various stakeholders. I'm pro-AI in the sense that, when used correctly, it helps junior devs learn and otherwise shines a spotlight where the real bottlenecks are (project management and delays).
They're unlikely to be vocal because they've already evaluated and decided the role AI will play in their work/life and just moved on/kept working. That's also likely to be a small pool of people relative to the number of people interacting with AI (either by force or choice).
Perhaps small now. But overall this seems to be what most have done in my small social circle. Small uses of AI here and there, sometimes surprising gains/wins they find out and share if asked, otherwise really it's just another tool to learn and make use of.
Most everyone thinks it's overhyped and wedged into stupid places it doesn't make much sense, especially in consumer products. But they also can recognize that behind the scenes and in closed doors there are perhaps some exciting things happening in certain industries and use-cases.
Most are still very much in the "haven't played with it much yet, it's moving too fast and I'm too old and busy with my life" or "still evaluating as time allows" stage. This is mid-career folks in various professional roles, skewing towards tech. Some sense of "this tech is getting good fast, and I will be left behind if I do not learn it at some point in the future".
I will say most have barely ventured past the "copy/paste from ChatGPT" stage of AI use. It usually elicits comment when someone moves past that and finds out they are far more capable than they realized. Then usually some more comments a couple weeks later mentioning newfound limitations.
I count quite a few of my colleagues into this observing and must learn role. My problem is that most entry use cases that are advertised are not cutting it for me. Write overlong emails that could have been a short hand written one? Write longer prompts to get shorter LLM email responses? Create presentations that screw up the corporate theme and send me fixing it? Write power automate flows that send me chasing errors? Particularly management and leadership positions are urged to engage with LLM tools but are drowning in the noise of busy days and flood of communication. Summarising emails that then misses the crucial information is not saving any time.
Ironically due to an overloaded inbox these days, I've been using a LLM as the world's most inefficient e-mail filter.
It does a decent job at this, and gives at least a nice starting point in the morning for a prioritized hitlist to power through. Gets most things pretty accurate, but I only use it to sense "importance" and not much else.
Certainly not life changing, just a nice to have.
I've had far more real gains with home IT projects. Tons of home automation integrations and "silly" projects I've been wanting to get done for a decade but never really had the energy to want to deal with. These sorts of things also allowed me some time/excuse to play around with agents to see what the fuss was about.
I now have the world's most complex garden irrigation/sensor setup complete with automated watering and grafana panels! I could have tied everything from the hardware through software together myself, but I just was not interested in installing the 856th Ubuntu VM in my life, configuring InfluxDB from scratch, etc.
I can see the future potential due to the above, but I also see how it went off the rails easily and also produced some laughably inefficient/unmaintable stuff like Flux queries which toss what should be variables as statics directly into the grafana panels, etc. Add a single new sensor to a group and you'll be re-writing the panel queries. Just horrible ground level design only maintainable via brute force. But when you just tell it to add the thing and it does... I can see why many feel it's a superpower.
For personal use I find "AI" extremely useful and fun.
I asked Claude to create a mini stack that email me a daily digest:
Today's weather, rain forecast. Top ten local news,US news, world news New HN articles, bubble up articles about bicycles, crafts, Japan, exclude Rust, Zig, Python, JS, Node, Bun. Summary of articles in my favorite subreddits.
A separate email summarizing new antique items on ShopGoodWill.com, with their starting prices, auction end dates.
Oh, and it configured Proxmox on my NUC and installed and configured the whole *arr stack.
Now, I am a techie, so I know what to tell it and could do all of the above myself with enough elbow grease, but I never would have. All of the above took under an hour.
Just this morning, I created a plugin so that Claude speaks to me as Samuel L. Jackson, or Arnold, etc. Fun fun.
Replaced 90% of my browser web searches. I don't know how many trees I burned and lakes dried.
We're right here! Some of us just prefer to stay out of these silly irrational debates because we know that neither of those two crazy camps will listen to a word we, or anyone else, will say. We might be wrong about that (I hope we're wrong about that!) but there sure are enough other people chattering that it's easy to sit back and let them have it out.
The "you'll get left behind" crowd has already made it clear that they're not worth engaging with, you're not going to find reasonable voices anywhere those threads are happening...
We're here -- I use AI for low-complexity toil, one-off scripts, search & discovery, and as a sounding board for new perspectives.
I still design and build systems almost all myself. If a snippet comes out of an LLM I'm keen to use it's filtered through my brain as well. Polished and transformed to align with my vision. After all, I'm on the hook for supporting it and bringing it to my team.
I'm sure most of us are just sick of the mania around this and choose to avoid interacting with content relating to it.
It's very useful. But irresponsible to use without recognizing and respecting its limitations.
I have > 10yoe, fwiw.
Everywhere, presumably, not vocal, doing other things.