Cursor IDE support hallucinates lockout policy, causes user cancellations

While some of what you say is an interesting thought experiment, I think the second half of this argument has, as you'd put it, a low symbolic coherence and low plausibility.

Recognizing the relevance of coherence and plausibility does not need to imply that other aspects are any less relevant. Redefining truth merely because coherence is important and sometimes misinterpreted is not at all reasonable.

Logically, a falsehood can validly be derived from assumptions when those assumptions are false. That simple reasoning step alone is sufficient to explain how a coherent-looking reasoning chain can result in incorrect conclusions. Also, there are other ways a coherent-looking reasoning chain can fail. What you're saying is just not a convincing argument that we need to redefine what truth is.

> Knowledge emerges from symbolic coherence, linguistic agreement, and social plausibility rather than purely from logical coherence or factual correctness.

This just seems like a redefinition of the word "knowledge" different from how it's commonly used. When most people say "knowledge" they mean beliefs that are also factually correct.

> The model genuinely believes it’s giving a correct reasoning chain, but the interpretability microscope reveals it is constructing symbolic arguments backward from a conclusion.

Sounds very human. It's quite common that we make a decision based on intuition, and the reasons we give are just post-hoc justification (for ourselves and others).

I haven’t used Cursor yet. Some colleagues have and seemed happy. I’ve had GitHub Copilot on for what feels like a couple years, a few days ago VS Code was extended to provide an agentic workflow, MCP, bring-your-own-key, it interprets instructions in a codebase. But the UX and the outputs are bad in over 3/4 of cases. It’s a nuisance to me. It injects bad code even though it has the full context. Is Cursor genuinely any better?

To me it feels like people that benefit from or at least enjoy that sort of assistance and I solve vastly different problems and code very differently.

I’ve done exhausting code reviews on juniors’ and middles’ PRs but what I’ve been feeling lately is that I’m reviewing changes introduced by a very naive poster. It doesn’t even type-check. Regardless of whether it’s Claude 3.7, o1, o3-mini, or a few models from Hugging Face.

I don’t understand how people find that useful. Yesterday I literally wasted half an hour for a test suite setup a colleague of mine introduced to the codebase that wasn’t good, and I tried delegating that fix to several of the Copilot models. All of them missed the point, some even introduced security vulnerabilities in the process invalidating JWT validation, I tried “vide coding” it till it works, until I gave up in frustration and just used an ordinary search engine, which led me to the docs, in which I immediately found the right knob. I reverted all that crap and did the simple and correct thing. So my conclusion was simple: vibe coding and LLMs made the codebase unnecessarily more complicated and wasted my time. How on earth do people code whole apps with that?

> The model genuinely believes it’s giving a correct reasoning chain

The model doesn't "genuinely believe" anything.

Yes

https://link.springer.com/article/10.1007/s10676-024-09775-5

> # ChatGPT is bullshit

> Recently, there has been considerable interest in large language models: machine learning systems which produce human-like text and dialogue. Applications of these systems have been plagued by persistent inaccuracies in their output; these are often called “AI hallucinations”. We argue that these falsehoods, and the overall activity of large language models, is better understood as bullshit in the sense explored by Frankfurt (On Bullshit, Princeton, 2005): the models are in an important way indifferent to the truth of their outputs. We distinguish two ways in which the models can be said to be bullshitters, and argue that they clearly meet at least one of these definitions. We further argue that describing AI misrepresentations as bullshit is both a more useful and more accurate way of predicting and discussing the behaviour of these systems.

Offtopic but I'm still sad that "On Bullshit" didn't go for that highest form of book titles, the single noun like "Capital", "Sapiens", etc

[flagged]

You're not supposed to trust the tool, you're supposed to review and rework the code before submitting for external review.

I use AI for rather complex tasks. It's impressive. It can make a bunch of non-trivial changes to several files, and have the code compile without warnings. But I need to iterate a few times so that the code looks like what I want.

That being said, I also lose time pretty regularly. There's a learning curve, and the tool would be much more useful if it was faster. It takes a few minutes to make changes, and there may be several iterations.

I'd add that the deductive abilities translate to well-defined spec. I've found it does well when I know what APIs I want it to use, and what general algorithmic approaches I want (which are still sometimes brainstormed separately with an AI, but not within the codebase). I provide it a numbered outline of the desired requirements and approach to take, and it usually does a good job.

It does poorly without heavy instruction, though, especially with anything more than toy projects.

Still a valuable tool, but far from the dreamy autonomous geniuses that they often get described as.

Versioning in source control for even personal projects just got far more important.

> the very first time I allowed it to look at my codebase, it hallucinated a missing brace (my code parsed fine), "helpfully" inserted it, and then proceeded to break everything.

This is not an inherent flaw of LLMs, rather it is a flaw of a particular implementation-if you use guided sampling, so during sampling you only consider tokens allowed by the programming language grammar at that position, it becomes impossible for the LLM to generate ungrammatical output

> When it does this, I feel the hallucinations can be off the charts -- inventing APIs, function names, entire libraries,

They can use guided sampling for this too - if you know the set of function names which exist in the codebase and its dependencies, you can reject tokens that correspond to non-existent function names during sampling

Another approach, instead of or as well as guided sampling, is to use an agent with function calling - so the LLM can try compiling the modified code itself, and then attempt to recover from any errors which occur.

[flagged]

> it hallucinated a missing brace (my code parsed fine), "helpfully" inserted it, and then proceeded to break everything.

Your tone is rather hyperbolic here, making it sound like an extra brace resulted in a disaster. It didn't. It was easy to detect and easy to fix. Not a big deal.

You say slowly, but in my opinion Apple made an out of character misstep by releasing a terrible UX to everyone. Apple intelligence is a running joke now.

Yes they didn't push it as hard as, say, copilot. I still think they got in way too deep way too fast.

Investors seem to be starved for novelty right now. Web 2.0 is a given, web 3.0 is old, crypto has lost the shine, all that's left to jump on at the moment is AI.

Apple fumbled a bit with Siri, and I'm guessing they're not too keen to keep chasing everyone else, since outside of limited applications it turns out half baked at best.

Sadly, unless something shinier comes along soon, we're going to have to accept that everything everywhere else is just going to be awful. Hallucinations in your doctor's notes, legal rulings, in your coffee and laundry and everything else that hasn't yet been IoT-ified.

They already rolled out an "AI" product. Got humiliated pretty bad, and rolled it back. [0]

[0] https://www.bbc.com/news/articles/cq5ggew08eyo

Even the iOS and macOS typing correction engine has been getting worse for me over the past few OS updates. I’m now typing this on iOS, and it’s really annoying how it injects completely unrelated words, replaces minor typos with completely irrelevant words. Same in Safari on macOS. The previous release felt better than now, but still worse than a couple years ago.

>>if it ever actually will.

If they don't then I'd hope they get absolutely crucified by trade comissions everywhere, currently there are bilboards in my city advertising Apple AI even though it doesn't even exist yet - if it's never brought to the market then it's a serious case of misleading advertising.

Yet Apple has reenabled Apple Intelligence multiple times on my devices after OS updates despite me very deliberately and angrily disabling it multiple times

When you got 1-2billion users a day doing maybe 10 billion prompts a day, it’s risky

I use Scala which has arguably the best compiler/type system with Cursor.

There is no world in which a compiler or tooling will save you from the absolute mayhem it can do. I’ve had it routinely try to re-implement third party libraries, modify code unrelated to what it was asked, quietly override functions etc.

It’s like a developer who is on LSD.

Yes, most people who have an incentive in pushing AI say that hallucinations aren't a problem, since humans aren't correct all the time.

But in reality hallucinations either make people using AI lose a lot of their time trying to stuck the LLMs from dead ends or render those tools unusable.

You get some superficial checking by the compiler and test cases, but hallucinations that pass both are still an issue.

Except when the hallucinated library exists and it's malicious. This is actually happening. Without AI, by using plain google you are less likely to fall for that (so far).

Until the model injects a subtle change to your logic that does type-check and then goes haywire in production. Just takes a colleague of yours under pressure and another one to review the PR, and then you’re on call and they out sick or on vacation.

The whole subreddit is moderated poorly. I’ve seen plenty of users post on r/LocalLlama about how something negative or constructive they said on the Cursor sub was just removed.

> The best case scenario is that you lied about having people answer support. LLMs pretending to be people (you named it Sam!) and not labeled as such is clearly intended to be deceptive.

Also, illegal in the EU.

I don't know who you are but you said this so well!

I wish more people realized that virtually any subreddit for a company or product is run by the company - either directly or via a firm that specializes in 'sentiment analysis and management' or whatever the marketdroids call it these days. Even if they don't remove posts via moderation, they'll just hammer it with downvotes from sockpuppet accounts.

HN goes a step further. It has a function that allows moderators to kill or boost a post by subtracting or adding a large amount to the post's score. HN is primarily a place for Y Combinator to hype their latest venture, and a "safe" place for other startups and tech companies.

Agreed, this is what's infuriating: insistence on control.

They will utterly fail to build for a community of users if they don't have anyone on-hand who can tell them what a terrible idea that was

To the cofounder: hire someone (ideally with some thoughtful reluctance around AI, who understands what's potentially lost in using it) who will tell you your ideas around this are terrible. Hire this person before you fuck up your position in benevolent leadership of this new field

I dunno, that seems pretty reasonable to me simply for stopping the spread of misinformation. The main story will absolutely get written up by some smaller news sources, but is it really a benefit for someone facing a similar issue in the future to find an outdated and probably confusing Reddit post about it?

Sorta? I mean I want my problem fixed, regardless of it it's a person or not. Having a person listen to me complain about my problems might sooth my conscience, but I can't pay my bill or why was it so high; having those answered by a system that is contextualized to my problem sand is empowered to fix it, and not just a talking to a brick wall? I wouldn't say totally fine, but at the end of the day, if my problem is solved or my query, even if it's weird, I can't say I really needed for the voice on the other end of the pHone to come from a human. If a companies business model isn't sustainable without using AI agents, it's not really my problem that it's not, but also if I'm using their product, presumably I don't want that to go away.

huh?

this is a completely reasonable and seemingly quite transparent explaination.

if you want a conspiracy, there are better places to look.

Why in all of hades would you force your entire eng org to only use one LLM provider. It's incredibly easy to run this stuff locally on 4+ year old hardware. Why is this even something you're spending company money on? Investor funds?

Not all of them (e.g. security@)! But our support system currently is. We are standing up a much bigger team here but are behind where we should be.

you didn't get a refund because the promise of refund was also hallucinated.

Maybe wait more than an hour before implying the refunds were a lie all along.

Yeah I got asked for feedback and offered a refund when I cancelled. Never got any reply after. Guess it was AI slop

To my understanding there weren't really distinct "implementation of the code to log people off" and "enforcement" - just a bug where previous sessions were being expired when a new one was created.

That an LLM then invented a reason when asked by users why they're being logged out isn't that surprising. While not impossible, I don't think there's currently indication that they intended to change policy and are just blaming it on a hallucination as a scape goat.

There’s AI and there’s “AI”, and this whole drama would have been avoided by returning links to an FAQ rather found using embedding search rather than actually then trying to turn it into a textual answer, which — working with these systems all day — is madness

> Don't use AI. Actually care.

I agree with this. Also, whenever I care about code, I don’t use AI. So I very rarely use AI assistants for coding.

I guess this is why Cursor is interested in making AI assistants popular everywhere, they don’t want the association that “AI assisted” means careless. Even when it does, at least with today’s level of AI.

The amount paid is still pretty trivial. I wouldn’t expect much human support for most SaaS products costing $20 a month.

Not trying to defend them, but I think it’s a problem of scaling up. The user base grew very quickly and keeping up with the support inquiries must be a tough job. Therefore the first like of defense is AI support replies.

I agree with you, they should care.

They’re like a team of 10 people with thousands, if not hundreds of thousands of users. “Actually care” is not a viable path to success here.

AI companies high on their own supply, that's who. Ultralytics is (in)famous for it.

A forward-thinking company that believes in the power of Innovation™.

An AI company dogfooding their own marketing. It's almost admirable in a way.

This is the future AI companies are selling. I believe they would 100%.

I worry that the tally of those who do is much higher than is prudent.

A lot of company actually, although 100% automation is still rare.

OpenAI seems to do this. I've gotten complete nonsense replies from their support for billing questions.

[dead]

Is this sarcasm? AI has been getting used to handle support requests for years without human checks. Why would they suddenly start adding human checks when the tech is way better than it was years ago?

It does say it's AI generated. This is the signature line:

    Sam
    Cursor AI Support Assistant
    cursor.com • hi@cursor.com • forum.cursor.com

Both things can be true. The AI support bot might have been trained to respond with “yup that’s the new policy”, but the unexpected shitstorm that erupted might have caused the company to backpedal by saying “official policy? Ha ha, no of course not, that was, uh, a misbehaving bot!”

> how incredibly stingy tech companies are about spending any money on support

Which is crazy. Support is part of marketing so it should get the same kind of consideration.

Why do people think Amazon is hard to beat? Price? nope. Product range? nope. Delivery time? In part. The fact if you have a problem with your product they'll handle it? Yes. After getting burned multiple times by other retailers you're gonna pay the Amazon tax so you don't have to ask 10 times for a refund or be redirected to the supplier own support or some third party repair shop.

Everyone knows it. But people are still stuck on the "support is a cost center" way of life so they keep on getting beat by the big bad Amazon.

That is because AI runs PR as well.

What if the prompt to the “support assistant” postulates that 1) everything is a user error, 2) if it’s not, it’s a policy violation, 3) if it’s not, it may be our fuckup but we are allowed? This plus the question in the email leading to a particular answer.

Given that LLMs are trained on lots of stuff and not just the policy of this company, it’s not hard to imagine how it could conjure that the policy (plausibly) is “one session per user”, and blame them of violating it.

I have a hard time being sold on “yea it’s wrong a lot, also you have to spend more time than you already do on code review.”

Getting to sit down and write the code is the most enjoyable part of the job, why would I deprive myself of that? By the time the problem has been defined well enough to explain it to an LLM sitting down and writing the code is typically very simple.

>> I use it all the time, and it has accelerated my output massively.

Like how McDonalds makes a lot of burgers fast and they are very successful so that's all we really care about?

> "and it has accelerated my output massively."

The folly of single ended metrics.

> but to avoid LLMs out of principal? That's just luddite.

Do you double check that the LLM hasn't magically recreated someone else's copyrighted code? That's just irresponsible in certain contexts.

> in case you feel it relevant.

Of course it's relevant. If a 19 year old with 1 year of driving experience tries to sell me a car using their personal anecdote as a metric I'd be suspicious. If their only salient point is that "it gets me to where I'm going faster!" I'd be doubly suspicious.

This is where education comes in. When we come cross a certain scale, we should know that O(n) comes into play, and study existing literature before trying to naively solve the problem. What would happen if the "AI" and web search didn't return anything? Would you have stuck with your implementation? What if you couldn't find a library with a usable license?

Once I had to look up a research paper to implement a computational geometry algorithm because I couldn't find it any of the typical Web sources. There were also no library to use with a license for our commercial use.

I'm not against use of "AI". But this increasing refusal of those who aspire to work in specialist domains like software development to systematically learn things is not great. That's just compounding on an already diminished capacity to process information skillfully.

Yes! This is how AI should be used. You have a question that’s quite difficult and may not score well on traditional keyword matching. An LLM can use pattern matching to point you in the right direction of well written library based on CS research and/or best practices.

I mean, even in the absence of knowledge of the existence of text searching algorithms (where I'm from we learn that in university) just a simple web search would have gotten you there as well no? Maybe would have taken a few minutes longer though.

But do you know every important detail of that library. For example, maybe that lib is not thread safe, or it allocates a lot of memory to speed thing up, or it wont work on ARM CPU because it uses some x86 hackery ASM?

Sounds like a job for silver/ripgrep and possibly stack exchange. Might take another minute to get it rolling but has other benefits like cost and privacy.

> I could have asked my knowledgeable friends and coworkers, but not at 11PM on a Saturday.

Get friends with weirder daily schedules. :-)

> never look at the code, just the output, and only talk to the LLM about what I wanted

Sir, you have just passed vibe coding exam. Certified Vibe Coder printout is in the making but AI has difficulty finding a printer. /s

How do you define code quality in this case and what is your stack?

Good employee, you get cookie and 1h extra pto

I had a professor who used `ed` to write his code. He said only bring able to see one line at a time forces you to think more about what you're doing.

Anyways, Cursor generates all my code now.

Actually, I'm kind of cheating because I use https://github.com/davidhalter/jedi-vim for that purpose: allows me to jump to definitions with <leader>d ;) Excellent plugin, and doesn't require an LSP.

I am unironically with you. I think people should start to learn from computer architecture and assembly and only then, after demonstrating proper skill, graduate to C, and after demonstrating skill there graduate to managed-memory languages.

Relevant XKCD: https://xkcd.com/378/

That’s a pretty mean spirited way to approach this subject.

I think the creators of Redis and Django are very capable and self-respecting software developers.

Job market for knowledge jobs isn’t even that good anymore and plenty of people expect it to get worse regardless of their stance on AI. What makes you so sure that LLM users have a bank to laugh all the way to? Already there are many like you, the money you’d make is peanuts

Are you going to the bank to take out a loan? You're claiming you've outcompeted other programmers by using... optimizing compilers?

what is it about reddit that causes this behavior, when they otherwise are skeptical only of whatever the "official story" is at all costs? it is fascinating behavior.

The reason many prefer Cursor over VSCode + GitHub Copilot is because of how much faster Cursor is for tab completion. They use some smaller models that are latency optimized specifically to make the tab completion feel as fast as possible.

Copilot's tab completion is significantly worse than cursor's in my experience (only tried free copilot)

That's the wrong IDE to compare it to though, Cursor's AI features are 10x better than VSCode's. I tried Zed last month and while the editing was great, the AI features were too half-baked so I ended up going back to Cursor. Hopefully it gets better fast!

I find that if you turn off telemetry (i.e. turn on privacy) the resource hogging slows down a lot

I have heard Roo Code is a fork of Cline that is better. I have never used either so far.

https://github.com/RooVetGit/Roo-Code

I found the Windsurf agent to be relatively less capable, but their inline tool (and the “Tab” they’re promoting so much) has been extremely underwhelming, compared to Cursor.

The only one in this class to be even worse in my experience is Github Copilot.

And the agent beta is looking pretty good, so far, too.

I have never rolled my eyes harder.

Apparently they use AI to read emails. So the future of email will be like phone support now, where you keep writing LIVE AGENT until you get a human responding.

Don't be an ass

Slack does this as well. It's a genius idea from a business perspective. Normally IT admins have to go around asking users if they need the service (or more likely you have to request a license for yourself), regularly monitor usage, deactivate stale users etc., all to make sure the company isn't wasting money. Slack comes along and says - don't worry, just onboard every user at the company. If they don't log in and send at least N messages we won't bill them for that month.

That's a fun one. It could be interpreted as a generous implementation of a monthly subscription, or a hostile implementation of a metered plan.

Wow, I wish more services did that.

Kagi should take it a step further and just charge per search

surely the first thing you do when you subscribe to Kagi is set your default browser search to Kagi.

Really? Brilliant idea.

This would only be complete in aggregate. We don’t know how many people signed up as a result.

This is true but Cursor’s problems are a bit worse than a normal paywalled service.

Cursor allows users to get free credits without a credit card and this forced them to change their VSCode fork on how it handles identification so they can stop users from spawning new accounts.

Another is that normally, companies have a cost for each free user. For Cursor, this cost is so sporadic since it doesn’t charge per million context, they use credits. Free users get 50 credits but 1 credit could be 200k+ context each so it could be $40-50 per free user per month. And these users get 50 credits every month.

Lastly, the cursor vip free repo has trended on GitHub many times and users who do pay might stop and use this repo instead.

The Cursor vip free creator is well within his rights to do what they want and get “free” access. This unfortunately hurts paying customers since Cursor has to stop these “hacks.”

This is why Cursor should just move to a VSCode extension. I’ve used Augment and other VSCode extensions and the feature set is close to Cursor so it’s possible for them just to be an extension. The other would be to remove free accounts but allow users to bring their own keys. To use Composer/Agent, you can’t bring your own keys.

This will allow Cursor to stop maintaining a VSCode fork, helps them stop caring if users create new accounts (since all users are paying) and lets users bring their own keys if they don’t want to pay. Hell, if they charge a lifetime fee to bring our own keys for Agent, that would bring in revenue too. But as I see now, Roo and Cline’s agent features are catching up and Cursor won’t have a moat soon.

The AI support triage agent must have deemed it unworthy.

> I don't understand why they don't fix this.

It makes number go up and to the right

Cursor was one of the first AI editors and there was no way to add those features to VSCode thru extensions at that time. Another issue was that VSCode made special exceptions and opened APIs for CoPilot, Microsoft’s product, but not for other players. Not sure if this has changed now.

Cursor took the best course of action at the time by forking but needs to come back into the fold. If VSCode is restricting access to APIs to CoPilot, forking it publicly and putting that in the Readme, “We forked VSCode since they give preferential treatment to CoPilot” would get a lot of community support.

Tradeoff with slowing down user acquisition

Currently, there are 59 comments after the post was deleted and the thread was locked.

It is worth mentioning that the comments that remain are valuable, as they highlight the captured market size and express concern about the impending deterioration of the situation.

No: an LLM that doesn't confabulate will certainly get things wrong in some of the same ways that honest humans do - being misinformed, confusing similar things, "brain" damage from bad programming or hardware errors. But LLM confabulations like the one we're discussing only occur in humans when they're being sociopathically dishonest. A lawyer who makes up a court case is not a "human being wrong," it's a human lying, intentionally trying to deceive. When an LLM does it, it's because it is not capable of understanding that court cases are real events that actually happened.

Cursor's AI agent simply autocompleted a bunch of words that looked like a standard TOU agreement, presumably based on the thousands of such agreements in its training data. It is not actually capable of recognizing that it made a mistake, though I'm sure if you pointed it out directly it would say "you're right, I made a mistake." If a human did this, making up TOU explanations without bothering to check the actual agreement, the explanation would be that they were unbelievably cynical and lazy.

It is very depressing that ChatGPT has been out for nearly three years and we're still having this discussion.

Tested the free chat. The chat bot gave slightly incorrect answer, and trustworthiness gave it score of 0.749 and said the answer is completely incorrect, which was not actually the case. Seems more confusing with two answers that are somewhat wrong.

There’s a bubble, in the sense of the dot-com bubble. There’s a lot of money being thrown at the AI equivalents of Pets.com, and that will pop eventually. But the internet persisted and transformed every aspect of the society even after the bubble. I think AI will be similar.

AI has historically gone through bubbles, AI Winter is a term exactly because of these cycles of hype.

Maybe this time is different. Maybe not.

Who knows, the topic might be an omen.

Railroads revolutionized transport and yet railway mania was undeniably a bubble. Something can be both very useful and yet also overvalued and overhyped leading to significant malinvestment (sometimes, everyone wins to the detriment of the investors, sometimes just everyone loses out because a huge amount of effort was spent on not useful stuff, usually somewhere in between).

I'm shocked by this perspective, and I'm deep into the LLM game (shipped 7 figure products using LLMs). I don't feel like anything has been revolutionized around communication - I can spot AI generated emails pretty easily (just send the prompt, people). On the learning front I do find LLMs to be more capable search engines for many tasks, so they're helpful absolutely.

I'm an AI fan, but there's clearly a desperate attempt by just about every tech company to integrate AI at the cost of genuinely productive developments in the space, in a manner that one might describe as a "bubble." Microsoft's gotta buy enough GPUs to handle Copilot in Windows Notepad, after all...

Even if they generously compensate locked out users, they'll probably end up ahead using the AI bot.

No, I get that. But that obvious screw-up is being conflated with assessments of "is Cursor any good?" "is Cursor still ahead of the pack?" "does LLM-assisted coding good/useful?" et cetera.

This one failure is kitchen-sinking into a broad list of grievances, perceived slights, or unrelated problems all at once. I don't understand that, since I'm finding Cursor to be strong, helpful, and constantly improving. My experience is entirely different from this list of grievances.

at no point did they say anything that even comes close to saying “no bugs in human written code.”

if you’re willing to come down off your defensive AI position, because your response is a common one from people who are bought into the tech, i’ll try explain what they were saying (if not, stop reading now, save yourself some time).

maybe you’ll learn something, who knows :shrug:

> Programming was one of the few remaining places where you had to be precise, where it was harder to fool yourself. Where you had to understand it to program it.

they are talking about the approach, motivations and attitudes involved in “the craft”.

we strive for perfection, knowing we will never reach it. we, as programmers/hackers/engineers must see past our own bullshit/delusions to find our way to the fabled “solution”.

they are lamenting how those attitudes have shifted towards “fuck it, that’ll do, who cares if the code reads good, LLM made it work”.

where in the “vibe coding” feedback loop is there a place for me, a human being, to realise i have completely misunderstood a concept for the last five years and suddenly realise “oh shit, THATS HOW THAT WORKS!? HOW HAVE I NOT REALISED THAT FOR FIVE YEARS.” ?

where in “just ask chatgpt for a summary about a topic” is my journey where i learn about a documentation rendering library that i never even knew existed until i actually started reading the docs site for a library?

maybe we were thinking about transferring our docs off confluence onto a public site to document our API? asking chatGpt removes that opportunity for accidental learning and growth.

in essence, they’re lamenting the sacrifice people seem to be willing to make for convenience, at the price of continually growing and learning as a human being.

at least that’s my take on it. probably wrong — but if i am at least i get to learn something new and grow as a person and see past my own bullshit and delusions!

I don't want to get into a semantics argument but that's not accountability. That's just one more behavior prompt/indication but you can't fire the LLM. It might still do the wrong thing.

and the best thing about it is that the base model is going to be trained on reddit posts so expect SupportBot3000 to be even more confident about this fact in the future!

I mean to be fair, I like that they're putting their money where their mouth is so to speak - if you want to sell a product based on the idea that AI can handle complex tasks, you should probably have AI doing what should be simple, frontline support.

Does "it" use the official extensions, or does it only allow "you" the user to use them?

Slack was a similar thing. IPO'd at 10 million users with 400M in revenue but got destroyed by Microsoft thanks to Teams.

Cursor is at a worse position and at greater risk of ending up like Slack very quickly and Microsoft will do the exact same thing they did to Slack.

This time by extinguishing (EEE) them by racing prices of VSCode + Copilot close to zero, until it is free.

The best thing Cursor should do is for OpenAI to buy them at a $10B valuation.

Yes, I’m complaining about complainers. It’s the circle of life on Hacker News, someone always has to play the food chain’s top predator: the mildly annoyed power user.

I probably should have added sarcasm tags to my post. My very firm opinion is that AI should only make suggestions to humans and not decisions for humans.

I'd argue that humans also more easily learn from huge mistakes. Typically, we need only one training sample to avoid a whole class of errors in the future (also because we are being held accountable).

As annoying as it is when the human support tech is wrong about something, I'm not hoping they'll lose their job as a result. I want them to have better training/docs so it doesn't happen again in the future, just like I'm sure they'll do with this AI bot.

How is this not an example of humans being held accountable? What would be the difference here if a help center article contained incorrect information? Would you go after the technical writer instead of the founders or Cursor employees responding on Reddit?

You are aggressive and rude. They didn’t sell you a house. It’s a product that costs 20 usd a month. I know all other products and prefer Claude Code. But it is expensive. I burn 20-40 USD in a single session. And I find Cursor pretty good. Better than the alternatives, minus Claude Code.