I must be using web browsers completely wrong. Like browsing a page isn't a problem for me. I can do it at the speed of my needs.
I'm having a hard time understanding why I will tell gemini to create an account on some website for me or send an email. Those are usually just a tab away. That's why I feel like I'm missing something here.
Basically none of their examples are just "browse a page"? They're multi-step tasks combining data from multiple pages.
Like the first example in the demo carousel (the Y2K party) starts from a photo and a prompt of roughly "buy the props needed for replicating this photo from Etsy". It first analyzes the image in the current tab, identifies a bunch of things to buy, searches for them on Etsy, customizes the orders, adds them to the shopping basket, and then asks for a confirmation to actually send an order.
The second one auto-fills a form with a couple of dozen fields from the data that's in a pdf in another tab. (And in the fiction of a demo, presumably a pdf that's you already had around, not one that you made just for the purposes of using it to auto-fill the form.)
I'm not the target market for this: automating a browser with my credentials is just too scary, but I can certainly see the utility. There's a huge amount of tasks taking a minute or two are not worth creating bespoke automation for but that are also pretty mechanical processes.
Maybe I’m a curmudgeon who can’t imagine throwing an elaborate Y2K party because all my friends were alive and threw parties at the real Y2K, but… these all feel extremely contrived.
It’s as if they used AI to generate use cases for their AI tool because they weren’t really sure what it’s for…
I feel that way about IDEs too, though. My text editor has snippets, my file manager shows me what files are where, and my terminal lets me run programs. Why it's important to people that these functions to be grafted into a single window escapes me.
Maybe you're only using well-designed sites? Try making a booking with a Chinese airline and you'll quickly wish for an assistant to delegate it all to.
funny you say that, I was literally just booking a flight with air china yesterday and the UX was 10x better than the average wizzair/ryanair experience - a clear, readable UI (with a great table comparison of prices +-3 days from the selected dates), no ads, no random services getting pushed in your face, no booking tabs automatically opening in the background
Huh. Last time I tried with them (about a year ago), and more recently trying with China Eastern, I couldn't even get it to show me a flight that I knew was flying on a given day (just at a slightly higher price than the one it would show me).
I wrote my story and titled it, "My experience at work with an automated HR system". I sent it to a few friends, only a couple of them read it.
A week later, I renamed it to "The Machine Fired Me". That seemed to capture it better. The goal wasn't to make it click bait, but it was to put the spoiler, and punch line right up front. It blew up!
I had just read Life of Pi, and one thing I like about that book is that you know the punch line before you even pick up a copy. A boy is stuck with a bengal tiger in a boat. Now that the punch line is out of the way, the story has time to unfold and be interesting in its own merit. That's what I was trying to recreate with my own story.
Reminds me of Veritasium's recent videos, really driving that initial hook and maintaining the viewer's attention. He had an explanation video about it which explained how people who would be interested in something like "the Lorenz equation" probably don't know what it's called, so it might be more accurate to phrase it in terms that someone would search for or initially peak their interest.
And I think it fits neatly with making people care first. I want to learn more about the machine that fired you, that's more the start of a narrative arc. It's almost like I have more trust that you will make it interesting, since you put a little more work up front.
> The goal wasn't to make it click bait, but it was to put the spoiler, and punch line right up front.
For those who are really adverse to that kind of thing and have trouble with thinking "but it is is just making it sound like clickbait" in the comparison above: You don't have to go as far with it either. Just inserting inserting that one detail without changing the style or shortening it makes the reader's mind go from "maybe some person complaining about automated form requirements in benefits sign up or some first week onboarding program or something" bore to "fired by an automated HR!?" interest.
Also I tend to do the "Don't recommend -> I don't like this video" for those that have the thumbnails with "that face" (you know, the YouTube Thumbnail Clickbait Face, I don't even know if there's an actual term for it).
When I actually enter a video, you have my attention by default and you'll get an instant dislike for:
- "Don't forget to like and subscribe."
- Showing those like and/or subscribe buttons on screen.
- If I get suspicious that you're padding video length, talking just for the sake of stalling.
>I had just read Life of Pi, and one thing I like about that book is that you know the punch line before you even pick up a copy. A boy is stuck with a bengal tiger in a boat. Now that the punch line is out of the way, the story has time to unfold and be interesting in its own merit. That's what I was trying to recreate with my own story.
For me this is a perfect example of what I hate about clickbait.
A boy trapped in a boat with a tiger is interesting. But the rest of the story really wasn't worth the read.
We estimated a single sprint to move from our plain text passwords. Easy! Add a new field in the db for secure pass, and one to force password update. Update the api to take the new fields into account...
It took 6 months. Why? Well it was a legacy app, and we learned that passwords were case insensitive because the customer sent a video of him entering his password that failed. On the video, we could see a sticky note on his monitor with the password written on it.
When we made all the necessary changes, the docker file failed to build. SRE accidentally deleted the deprecated image with PHP that had reached EOL.
Your story reminded me of Bent Flyvbjerg and Dan Gardner's book "How Big Things Get Done". It's a pop science book communicating the research of Flyvbjerg & collaborators who collected data of actual schedule & budget outcomes for many large projects, along with recording the estimated schedule and budgets from the time the go/nogo decision was made to invest in the projects.
For accurate schedule and budget estimates, Flyvbjerg strongly recommends figuring out which broader class of projects your project belongs to, then going and looking at accurate historical data with actual schedule and budget outcomes for projects in that class, and using historical averages as your estimates.
There's a great table in an appendix at the back of the book, offering statistics for each category of project, ranking them by mean cost overrun.
The absolute worst category of project, for mean cost overruns, is nuclear storage, with a mean cost overrun of 238%.
IT projects are the 5th worst category of project, with a mean cost overrun of 73%, behind nuclear storage, olympic games, nuclear power and hydroelectric dams.
The table also has statistics on "what percentage of projects has a cost overrun of 50% or greater" and "of those projects with a cost overrun of 50% or greater, what is their mean cost overrun". For nuclear storage projects, 48% of them have a cost overrun of 50% or greater, and of those, the mean cost overrun is 427% (!).
For IT projects, 18% of them have a cost overrun of 50% or greater, and of those, the mean cost overrun is 447% (!!).
Some of the chapters in the book discuss some of the structural or political pressures that set projects up to fail --- e.g. in some fields its an open secret that estimates are always wildly optimistic, as if the estimates were actually realistic, no one would ever agree to fund a project.
Oof. Exactly. Simple things get complicated when you find out that there were unstated requirements, hidden booby traps, etc. And then there's random interrupts -- unrelated stuff that comes up and takes your time and focus. If you run a lean ship then interrupt management is critical, but ideally you can have a few people who are understood to take critical interrupts and thus their schedules will slip.
Not to dismiss other people's experience, but thinking improves thinking. People tend to forget that you can ask yourself questions and try to answer them. There is such thing as recursive thinking where you end up with a new thought you didn't have before you started.
Don't dismiss this superpower you have in your own head.
In my experience LLMs offer two advantages over private thinking:
1) They have access to a vast array of extremely well indexed knowledge and can tell me about things that I'd never have found before.
2) They are able to respond instantly and engagingly, while working on any topic, which helps fight fatigue, at least for me. I do not know how universal this effect is, but using them often means that I can focus for longer. I can also make them do drudgery, like refactoring 500 functions in mostly the same way that is just a little bit too complicated for deterministic tools to do, which also helps with fatigue.
Ideally, they'd also give you a more unique perspective or push-back when appropriate, but they are yes-men too much right now for that to be the case.
Lastly, I am not arguing to not do private thinking too. My argument is that LLM-involved thinking is useful as its own thing.
Re: "yes men" - critical thinking always helps. I kind of treat their responses like a random written down shower thought - malicious without scrutiny. Same with anything that you haven't gone over properly, really.
The advantages that you listed make them worth it.
The output of the prompts always needs peer review, scrutiny. The longer is the context, the longer it will deviate, like if a magnet were put nearer and nearer to a navigation compass.
This is not new, as LLMs root are statistics, data compression with losses, It is statistically indexed data with text interface.
The problem is someones are selling to people this as the artificial intelligence they watched at movies, and they are doing it deliberately, calling hallucinations to errors, calling thinking to keywords, and so on.
There is a price to pay by the society for those fast queries when people do not verify such outputs/responses, and, unfortunately, people is not doing it.
I mean, it is difficult to say. When I hear some governments are thinking in to use LLMs within the administrations I get really concerned, as I know those outputs/responses/actions will nor be revised nor questioned.
No one is arguing that thinking doesn’t improve thinking. But expressing thoughts precisely by formulating them into the formalized system of the written word adds a layer of metacognition and effort to the thinking process that simply isn’t there when 'just' thinking in your head. It’s a much more rigorous form of thinking with more depth - which improves deeper, more effortful thinking.
Exactly. As distributed systems legend Leslie Lamport puts it: “Writing is nature’s way of letting you know how sloppy your thinking is.” (He added: “Mathematics is nature’s way of letting you know how sloppy your writing is.”)
I still have a lot of my best ideas in the shower, no paper and pen, no LLM to talk to. But writing them down is the only way to iron out all the ambiguity and sort out what’s really going to work and what isn’t. LLMs are a step up from that because they give you a ready-made critical audience for your writing that can challenge your assumptions and call out gaps and fuzziness (although as I said in my other comment, make sure you tell them to be critical!)
Thinking is great. I love it. And there are advantages to not involving LLMs too early in your process. But it’s just a first step and you need to write your ideas down and submit them to external scrutiny. Best of all for that is another person who you trust to give you a careful and honest reading, but those people are busy and hard to find. LLMs are a reasonable substitute.
Agreed; also a kind of recursive placebo tends to happen in my experience:
10 You recognise your thinking (or some other desirable activity) has improved
20 You're excited about it
30 You engage more with the thinking (or other activity)
40 You get even better results
50 Even more excitement
60 GOTO 30
There is nothing to suggest LLMs will be as revolutionary as paper. The PalmPilot didn't lead to new field of science just because people had a new way to write things down.
The internet is as arguably as revolutionary as paper. And while LLMs haven’t proven to be an internet-level revolutionary technology (yet), they are closer to that than the PalmPilot.
As cognitive-offloading devices go, paper is completely neutral. It doesn't flatter you into believing you are a genius when you're not; it doesn't offer to extend your reasoning or find references in the research literature and then hallucinate and lead you astray; it will never show you advertisements for things you don't need; it will never leak your ideas and innermost thoughts to corporations owned by billionaires and totalitarian governments... I could go on but you get the drift, I'm sure. Paper wins by a mile.
? They certainly flatter you, openAI even felt compelled to give a statement on the sycophancy problem: https://openai.com/index/sycophancy-in-gpt-4o/
And South Park parodied the issue.
I use chatGPT and claude every day.
> No they don’t flatter you, try using ChatGPT once.
You're absolutely right!
On a more serious note, if it has almost infinite knowledge, is it even a cognitive-offloading tool in the same class as paper? Sounds more like something designed to stifle and make my thoughts conform to its almost infinite knowledge.
edit:
I'll admit ChatGPT is a great search engine (and also very hallucinatory depending on how much you know about the subject) and maybe it helps some people think, sure. But beyond a point I find it actually harmful as a means to develop my own ideas.
I almost entirely agree with you, but the issue is that the information you currently have might not be enough to get the answers you want through pure deduction. So how do you get more information?
I think chatbots are a very clumsy way to get information. Conversations tend to be unfocused until you, the human, take an interest in something more specific and pursue it. You're still doing all the work.
It's also too easy to believe in the hype and think it's at least better than talking to another person with more limited knowledge. The fact is talking has always sucked. It's slow, but a human is still better because they can deduce in ways LLMs never will. Deduction is not mere pattern matching or correlation. Most key insights are the result of walking a long tight rope of deductions. LLMs are best at summarizing and assisting with search when you don't know where to start.
And so we are still better off reading a book containing properly curated knowledge, thinking about it for a while, and then socializing with other humans.
No I don’t think humans have some magical metaphysical deduction capability that LLMs lack exclusively.
I have had conversations and while they don’t have the exact attentiveness of a human, they get pretty close. But what they do have an advantage in is being an expert in almost any field.
Yes, LLMs have been a very expensive philosophy lesson for many investors. Ancient epistemology debates are now front and center for everyone to see. So-called "formal epistemology" is just empiricism in disguise attempting to borrow the credibility of rationalism and failing miserably.
LLMs are Bayesian inference and come with all its baggage. We definitely know brains are way better than that, even of other animals or insects.
Ultimately, there's no point in getting a chatbot to say deceptively expert-like words that are guaranteed by design to be lower quality than the books or blogs it learned from. LLMs are at best a search tool for those sources, and investor attitude now reflects that sanity with their confidence shifting back over to Google's offerings. Agentic AI is also pretty weak since agents are as functionally limited as any traditionally written computer program, but lacking the most crucial property of repeatability.
I find it shocking how many people didn't see this whole thing as a grift from day one. What else was SV going to do during the post-covid economic slump?
I've seen people solve their own issues by asking me / telling me about something and finding the solution without me having the time to reply numerous times.
Just articulating your thoughts (and using more of your brain on them by voicing them) helps a lot.
Some talk to themselves out loud and we are starting to realize it actually helps.
Just like how writing helps memorisation. Our brains are efficient, they only do what they have to do. Just like you won't build much muscles from using forklifts.
I've seen multiple cases of... inception. Someone going all in with ChatGPT and what not to create their strategy. When asked _anything_ about it, they defended it as if they came up with it, but could barely reason about it. Almost as if they were convinced it was their idea, but it really wasn't. Weird times.
Indeed, people trying to write prompts for the chatbots and continuously iterating on making their prompts clearer / more effective at conveying their needs is an exercise many haven't done since highschool. Who would've thought that working on your writing and reading proficiency may improve your thinking.
I can imagine for some it's quite a challenge to deviate from short-form shitposting they normally do and formulate thoughts in complete sentences for the LLMs.
Recursive self-questioning predates external tools and is already well known. What is new is broad access to a low cost, non retaliatory dialogic interface that removes many social, sexual, and status pressures. LLMs do not make people think. They reduce interpersonal distortions that often interfere with thinking. That reduction in specific social biases (while introducing model encoded priors) is what can materially improve cognition for reflective and exploratory tasks.
Simply, when thinking hits a wall, we can now consult a machine via conversation interface lacking conventional human social biases. That is a new superpower.
Unfortunately we do neglect more and more of our own innate talents. Imagine sitting there just thinking, without even a reMarkable to keep notes? Do people even trust their memory beyond their immediate working memory?
It's also absolute awesome how every person's brain works the same way. It makes it some much more convenient that what works for one person works for every person.
When I was a kid people told me I needed no Chess Computer - You can play chess in your head, you know ? I really tried, no luck. Got a mediocre device for Christmas, couldn't beat it for a while, couldn't lose against it soon after. Won some tournaments in my age group and beyond. Thought there must be more interesting problems to solve, got degrees in Math, Law and went into politics for a while. Friends from College call on your birthday, invite you to their weddings, they work on problems in medicine, economics, niches of math you've never heard of - you listen, a couple of days later, you wake up from a weird dream and wonder, ask Opus 4.5/Gemini 3.0 deepthink some questions, call them back: "did you try X ?" they tell you, that they always considered you a genius. You feel good about yourself for a moment before you remember that Von Neumann needed no LLMs and that José Raúl Capablanca died over half a decade before Turing wrote down the first algorithm for a Chess Computer. An Email from a client pops up, he isn't gonna pay your bill unless you make one more modification to that CRUD app. You want to eat and get back to work. Can't help but think about Eratosthenes who needed neither glasses nor telescopes to figure out the earths circumference. Would he have marvelled at the achievements of Newton and his successors at NASA or made fun of those nerds that needed polished pieces of glass not only to figure out the mysteries of the Universe but even for basic literacy.
The way most people think is by talking to each other but writing is a stronger way to think and writing to an LLM or with the help of an LLM has some of the benefits of talking with someone. Also, writing and sketchingon a piece of paper have unique advantages.
I suggest you start converting your writing into short digestible Tiktok dance moves...
Joking aside, paper is resilient. Share your digital writings everywhere, then make paper copies that you can donate to libraries. If this fails, that's fine. You won't be around to see it.
You've made my day. On the other hand it's quite bold to say that Tiktok dances will last 100 years. The videos there are 30-40 seconds long matching viewers attention span, so using simple math we can extrapolate that in 100 years they will last 0.03ns at max
My main concern is that there isn't a reliable way to know your information is securely stored[0].
> A few years ago, I received a letter in the mail addressed to my then-toddler. It was from a company I had never heard of. Apparently, there had been a breach and some customer information had been stolen. They offered a year of credit monitoring and other services. I had to read through every single word in that barrage of text to find out that this was a subcontractor with the hospital where my kids were born. So my kid's information was stolen before he could talk. Interestingly, they didn't send any letter about his twin brother. I'm pretty sure his name was right there next to his brother's in the database.
> Here was a company that I had no interaction with, that I had never done business with, that somehow managed to lose our private information to criminals. That's the problem with online identity. If I upload my ID online for verification, it has to go through the wires. Once it reaches someone else's server, I can never get it back, and I have no control over what they do with it.
All those parties are copying and transferring your information, and it's only a matter of time before it leaks.
Exactly. Everything "private" that you post online will become public eventually.
Everyone says "we only store the data temporarily and it's deleted right after" including everyone who didn't do that and got hacked.
But I think we're far too late into this issue by now.
It's 2026 and we still don't have a way to know if our passwords are being stored in a secure way in their databases. What hope do we have to know about how our photos are being handled?
Honestly that main concern should be two main concerns.
You/your kid/your wife goes to hàckernews.com and is prompted for age verification again, evidently the other information has expired based on the message. So they submit their details. Oops, that was typosquatting and now who the hell knows has your information. Good luck.
Slight tangent but I've had my own Pavlovian moment when I taught a loud neighbor how to keep the volume down [0]. It all started when our RF tv remote interfered. He probably thought there was a ghost that turned off the tv whenever his volume went above 15.
reply