There has been an explosion in verbose status update emails at my job recently which have all clearly been written by ChatGPT. It’s the fucking emojis though that drive me wild. It’s so hard to read the actual content when there’s an emoji for every single sentence.
And now when I see these emoji fests I instantly lose interest and trust in the content of the email. I have to spend time sifting through the fluff to find what’s actually important.
LLMs are creating an assymetric imbalance in effort to write vs effort to read. What is taking my coworkers probably a couple minutes to draft requires me 2-3x as long to decipher. That imbalance used to be the opposite.
I’ve raised the issue before at work and one response I got was to “use AI to summarize the email.” Are we really spending all this money and energy on the worlds worst compression algorithm?
I consider myself progressive and my main issue with the technology is that it was created by stealing from people who have not been compensated in any way.
I wouldn’t blame any artist that is fundamentally against this tech in every way. Good for them.
Every artist and creator of anything learned by engaging with other people's work. I see training AI as basically the same thing. Instead of training an organic mind, it's just training a neural network. If it reproduces works that are too similar to the original, that's obviously an issue, but that's the same as human artists.
For profit products are for profit products, that are required to compensate if they are derivative of other works (in this case, there would be no AI product without the upstream training data, which checks the flag that it's derivative).
If you would like to change the laws, ok. But simply breaking them and saying 'but the machine is like a person' is still... just breaking the laws and stealing.
This is a bad-faith argument, but even if I were to indulge it: human artists can/do get sued for mimicing the works of others for profit, which AI precisely does. Secondly, many of the works in question have explicit copyright terms that prohibit derivative works. They have built a multi-billion dollar industry on scaled theft. I don't see a more charitable interpretation.
You can't call something a bad-faith argument just because you disagree with it. I mean, you can, but it's not at all convincing.
As I said, if AI companies reproduce copyrighted works, they should be sued, just like a human artist would be. I haven't experienced that in my interactions with LLMs, but I've never really tried to achieve that result either. I don't really pirate anymore, but torrents are a much easier and cheaper way to do copyright infringement than using an AI tool.
LLMs don’t have to be able to mimic things. And go ahead and sue OpenAI and Anthropic! It won’t bother me at all. Fleece those guys. Take their money. It won’t stop LLMs, even if we bankrupted OpenAI and Anthropic.
It's "unauthorized use" rather than "stealing", since the original work is not moved anywhere. It's more like using your creative work to train a software system that generates similar-looking, competing works, for pennies, at industrial scale and speed.
Usually "obtaining" is just making a bunch of HTTP requests - which is kind of how the web is designed to work. The "consent" (and perhaps "desired payment" when there is no paywall) issue is the important bit and ultimately boils down to the use case. Is it a human viewing the page, a search engine updating its index, or OpenAI collecting data for training? It is annoying when things like robots.txt are simply ignored, even if they are not legally or technically binding.
The legal situation is unsurprisingly murky at the moment. Copyright law was designed for a different use case, and might not be the right tool or regulatory framework to address GenAI.
But as I think you are suggesting, it may be an example of regulatory entrepreneurship, where (AI) companies try to move forward quickly before laws and regulations catch up with them, while simultaneously trying to influence new laws and regulations in their favor.
[Copyright law itself also has many peculiarities, for example not applying to recipes, game rules, or fashion designs (hence fast fashion, knockoffs, etc.) Does it, or should it, apply to AI training and GenAI services? Time will tell.]
Accurate terminology can help to clarify discussions, including discussions of appropriate activity related to creative works.
My suspicion is also that what many artists are objecting to is less the training itself (though AI web crawlers are aggressive and annoying) and more the use of the trained model for large-scale generation of similar (and possibly competing) works ("derivative" in an artistic if not legal sense), in the artist's style, especially by large companies for commercial gain.
Many of those people have absolutely been compensated, many times over. Should they have a perpetual right to longer than their entire life, to profit off something they did 40 years ago?
You wouldn’t use MAUI to build something for only one platform. You would just use whatever it’s an abstraction over for that platform which in the case of Windows is WinUI.
Because if you can’t do that, and effectively articulate why AI media can’t be art when used as a tool by an artist to achieve their creative intent, I would claim the win on this. one.
You were going to claim the “win” on this one no matter what anyone says. That’s one of the features of being arrogant.
I have zero desire to get into a semantic argument over this. That would be very boring and is a poor refuge for anyone trying to have an honest discussion.
It’s not a coincidence that nearly all the people today with actual artistic talent universally despise AI. Meanwhile it’s all the talentless tech bros who won’t shut up about how they’re now incredible “artists” that love AI.
I don’t see prompting an AI as creating art in the same way that commissioning a painting doesn’t make you an artist. In this example, it’s the AI model that is the artist creating the “art”, but since AI models aren’t sentient (yet) then what they create isn’t art anymore than a sunset is despite being aesthetically beautiful.
Your “it’s just a tool” argument is especially ridiculous when you consider that the “tool” can create the same “art” in its entirety without you. It would be like if I googled the Mona Lisa and copy pasted it into Paint and then called myself an artist because I used Google and Paint as my tools.
In the case of AI models anything you can think to prompt is already embedded in the model so it’s not like you’re even creating anything. It’s already there. If you have infinite monkeys on a keyboard prompting AI they can generate every single possible image an AI is able to generate. Where is the artistry again?
I'll note I used no pejorative terms to describe you in order to make my points. I don't need to cast you as a villain or "unwise" individual to argue against the statements themselves.
--
Your framing of AI presumes that my proposal is prompt in >> art out. False frame.
You also state "near universal" disdain of AI from people with "actual" talent. A bold claim to make with little data, and at least for me, clear examples of being false.
I know award-winning directors, iconic creatives, and career/professional artists who are all excited by the technology, exploring ways to use it, and learning how to composite it into their work.
Perhaps you would propose that they have no talent and are not "true" artists! This is why I ask for the definition of art. It's not a "poor refuge" for an honest discussion, it's something that is the fundamental term upon which your argument is hinged. You can't gatekeep "art" without defining what the gate is.
The OP talks about Suno Studio - A DAW experience with recording, editing, generative tools, and generative restyling. This is, objectively, a different level of tool than a "prompt to music" generation experience.
Something that is composited through multiple decisions made by a human, toward some end that only they see, through an iterative process...
I think I finally see what we’re clashing on. Where do you draw the line between “this is a tool assisting me in creating my art” and “here is my AI film I stitched together with 10 Sora prompts”?
The songwriter for K-Pop Demon Hunter’s Soda Pop used AI to help them write the lyrics - fine. The hacks on X calling themselves incredible artists because they can prompt Dall-E - slop.
Yes - If you're willing to allow that AI might be a tool in the artistic process if it is used in conjunction with intent and effort, then we are at a reasonable position -- That stating "all AI is slop" is as inarticulate and insufficient as the claim of "look at what I made, I'm an artist!" coming from someone who typed in a four word prompt.
The majority of people using AI believe that production itself is art. They produce slop because they have no craft, no taste, and no innate talent.
Artists recognize that art requires the development of craft and taste - and there are many that see AI as a tool that can be used in that process.
So you must really hate collage artists. All they are doing is cutting out images and pasting them into new images. Picasso really produced some slot I guess. You seem to have a very rigid definition of "art".
I went to a concert recently and naturally they don’t let you bring alcohol in to the venue. So I very smartly decided to store the alcohol in my belly before I got there. Security had no idea how hard I bamboozled them.
I’m not sure how many people there are like me outside of this website but there’s not a single bone in my body that wants to use AI for these things.
Buying plane tickets for example. It’s not even that I don’t trust the AI or that I’m afraid it might make a mistake. I just inherently want to feel like I’m in control of these processes.
It’s the same reason I’m more afraid of flying than driving despite flying being a way safer mode of travel. When I’m flying I don’t feel like I’m in control.
I have very normie and not-so-normie friends that ask ChatGPT almost anything. My parents consistently make use of it, and they're almost 70, and not that tech-literate. There was a fun release from Anthropic about the type of queries that they're receiving, and code-gen is minority. I think we're, once again, not the average user.
I wonder how many of those “average users” will actually happily pay what the true cost is though? Are they really getting perceivable value for it or is it just more convenient than present day Google.
They all pay the price of google (micro-brainwashing by ads to buy things they don't need).
I think the average person will happily pay the same price to OpenAI (being micro-brainwashed by the AI to buy things they don't need, i.e. ads). I feel confident OpenAI will be able to charge even more for ads than Google since OpenAI will be able to influence people even more strongly, and hide the ads even better.
When you say "true cost" I've interpreted that to mean the down-sides of using an AI ChatBot as a primary information source, so my reply below is in context of that interpretation:
There is a sizeable chunk of people who (perhaps foolishly) trust ChatGPT despite knowing it can produce errors. They use it because it does the "research" for them, and does so quickly. This presents a type of tech-agility that they themselves do not possess. So on balance they may be more tech-empowered by using a flawed AI ChatBot than they are by manually reading news, websites and blogs.
There is also an issue of trust. A novice reading the top 5 search results has no real idea if the information being presented is biased, error free, or even factual. Google's work to blend paid and organic placement also presents the flaw of dollars over quality. ChatGPT on the other hand presents a known level of trust to them.
A similar scenario plays out with the way novices are more trusting of apps that appear on a curated store rather than seeking out software via web searches.
I think that users on HN take for granted that they have outsized experience and skill in developing trust in the tech landscape, and have a mental list of news, websites and software providers that they deem trustworthy. This can lead to not understanding the motivation for relying on an AI ChatBot, or compartmentalising people who use those services as some kind simpleton.
I don’t see why you wouldn’t book a flight using an AI assistant. No one’s saying it should do it completely unsupervised (maybe that’ll come much later), but having something that can research the best routes based on my criteria and show me several options — with a single click to purchase the one I find most convenient — is something I’d love.
It could even work against the dynamic pricing algorithms airlines use to maximize revenue: if I have a tireless assistant exploring every possible combination to find the cheapest ticket, it’ll probably do a much better job than I ever could.
There's probably little danger to the savvy user who understands how manipulative technology like this can be.
The problems come when vulnerable users are targeted using dark patterns. How AI dark patterns will evolve is very uncertain [1] however I suspect they will be extremely subtle and very effective.
What's the worst that can happen if someone vulnerable is persuaded to buy a flight by an AI. I don't know, maybe depression and bad credit after the chatbot's promises weren't met. If they're persuaded to buy a weapon, that's a different matter.
At least current advertising is somewhat public, although that's increasingly less true as ads get more targeted.
This is new territory where ads will be so extremely private it will be only known by the user (maybe they won't even notice) and someone reading the subpoenaed chat logs after a user does something terrible. Those chat logs will likely be inconclusive anyway.
I suppose you just have to trust that it's incentivized to find you the best route and not only offer you 3 options which it says are the best, but are actually paid promotions.
It depends too on what you value. I’d be more than happy to pay a premium if it meant the time for me looking for a flight and having a seat booked is drastically reduced.
We used to get that through the services of a travel agency. Maybe we will soon have that luxury again?
I would try using AI to book flights - then double check if I can't get a better offer. Do this a couple of times and when I see AI is as good or even better at getting me flights, then sure, why not use it.
Extrapolating from my experience testing it for coding tasks the result is not reliable even if it was right a couple of times. A risk I'm not willing to take. And I can't say that AI powered chat assistants on web pages have been much help either.
You can even automate this kind of testing in the AI model. I think the Google ADK has a built-in system for tests you use to confirm the reply quality.
I feel the same, but Airline and big hotel websites have way too many dark patterns made to confuse the user and force them to pay extra.
Booking an emergency flight last time I had a family issue was a mind-fucking experience. I had to go through 10 screens trying to sell me stuff and constantly hiding the skip button in different places. Maybe HN will say that I "shouldn't have had a family emergency in the first place" but reality is realty.
And honestly it's not just booking websites, it's anything tech that they do. For example, the last checkin kiosk I used also had an incredibly convoluted path for the case where someone else booked my luggage but it was a different size.
I’m with you. My elderly parents always ask me to book a ticket for them every time they need to fly because the airline websites are so full of dark patterns, it drives them anxious that they’ve missed something or spent money on something they don’t need.
This is sadly prevalent in some niches (e.g. low cost travel), but I don't think LLMs would be able to navigate those dark patterns better than humans would.
Flexibility is the advantage. In a chat interface, you can type literally whatever you want and ChatGPT will do its best to serve you. In a website like Skyscanner, you are inherently limited by their UI design.
If all you want is the cheapest flight on a specific day, Skyscanner is really great. But what if you need to book a bus at the other end of your flight? Skyscanner is not going to help you with that, but ChatGPT might! It could search up different bus providers in your destination and cross-reference them against the available flights.
How much you trust ChatGPT to actually do this well is up to you. But I suspect a lot of people will trust it, and I would probably be willing to use it for low-stakes tasks at least.
I would argue Skyscanner or whatever other company is better in offering additional services (hotels, taxis, buses) than ChatGPT, because it’s specialized.
I think if you really exactly know what you want the input in AI might be faster "book me on the flight tomorrow at 1pm from x to y on airline xyz -y" This I could imagine being faster, but it would still require verification by me to actually pay. I wonder if AI is faster in doing that given the added latency compared to me visiting airline xyz and doing the search manually (even perceive loading time taking in consideration) as it will be perceived less time if you are active.
It will just iframe whatever page/app you would have been browsing anyway but potentially with ChatGPT directly being able to operate on the App state.
So if configured, I guess ChatGPT will be just a handy middle layer to your usual interfaces.
It baffles me that people seem to think that chat is limited to text and text only. We're not there yet but the moment chats get excellent embedded interfaces is when we see this tech really come to fruition - at least from a consumer point of view.
Here's somewhat of a counter example. At work our llm project can schedule you time off. Workday already has a dedicated UI for this, so text interface can't be better right? Well it's a very popular feature, people use it all the time. In my opinion it's not better than a dedicated UI, but for some people it's good enough and more convenient (our site loads much faster than workday, they are likely already using it throughout their day, etc.)
Ahh great point thank you very much this makes a lot of sense.
I see my mistake now. I evaluate based on how it could be useful for me. As a heavy computer user, familiar with shortcuts and user interfaces, interacting with UX works very good.
But for a lot of users text will be more natural and easier. I might be able to get the flight I want easiest with Skyscanner, but other users might not be and will come to a better result with texts.
It’s the same as I prefer documentations over Youtube tutorials, but it’s different on different stages.
That was my intention, they've completely dropped the ball lately and I don't know if its incompetence or just they through there would never be another option so felt the need to optimize for another metric than search success.
The majority of americans are more concerned with AI. Only like 22% are optimistic. And why would they be optimistic that it'll result in a better life for them
If (when) companies want their things to be present in ChatGPT replies, they need to provide an AI-compatible way to get it. Just shoving a full-ass web page at it is inefficient and error-prone.
They have to either build a version of their site that's AI-accessible or provide an API (or MCP) for it to access the data.
Now that the API is built and the cost is paid, we can use it for non-AI uses.
In India, it is pretty common to call a travel agent and book tickets, in fact it is the preferred method for those who can afford it. It is super convenient, everything including the transfer of funds is taken care of by the agent.
This experience is 10x better than online alternatives. AI agents can replicate this at marginal cost.
you don't have to go so far as it buying the tickets for you if you don't trust it enough to do that. I built a deep research agent and one of the tasks that i found it very useful for was taking complex requirements and building a report for me to review and make decisions based off of. I live in one city, my travel partner lives in another, and we each want flights to get to a city around the same time, options for airbnbs, and travel activities. I may not trust ai to do this without human intervention but i certainly trust it to assemble this information for me with options and i can make decisions based on that
Same here. Neither do I trust these tools to be working accurately, nor do I have the patience to wait for them to complete the given task when I ca do that manually 10x faster already.
I have use a system similar to this guy and TickTick is perfect. I even use shared lists with my girlfriend to track chores which is something we implemented recently and works great.
And now when I see these emoji fests I instantly lose interest and trust in the content of the email. I have to spend time sifting through the fluff to find what’s actually important.
LLMs are creating an assymetric imbalance in effort to write vs effort to read. What is taking my coworkers probably a couple minutes to draft requires me 2-3x as long to decipher. That imbalance used to be the opposite.
I’ve raised the issue before at work and one response I got was to “use AI to summarize the email.” Are we really spending all this money and energy on the worlds worst compression algorithm?