Tell HN: Y Combinator backing AI company to abuse factory workers | Hacker News Reader

Tell HN: Y Combinator backing AI company to abuse factory workers

631 pointslatexr188 comments

Optifye.ai is a dystopian company backed by Y Combinator. They’re using AI to further dehumanise and abuse individual factory workers and treat them like disposable automatons.

See a now deleted post where they show how it works:

https://hachyderm.io/@YvanDaSilva/114063748264591929

The founders look to be a couple of rich kids with little world and work experience:

> We’re CS grads from Duke and because our families run manufacturing companies (…)

They also display a profound lack of empathy by bragging about lowering stress for rich company owners, which they do by increasing the stress of everyone who works for them:

> Know any manufacturing company owners?

> Let us know at founders@optifye.ai, and we’ll help them drop their cortisol levels :)

https://www.ycombinator.com/companies/optifye-ai

This is the AI world we all know is coming, brought to you by Y Combinator investors and founders. It doesn’t “benefit humanity”, it just serves to “put you in your place”.

View on HN

Has this been manually delisted from the frontpage? It doesn't say it's flagged, but it's no longer on there.

EDIT: looks like it's #157 now, I think I just misunderstand the ranking mechanism.

It set off the flamewar detector. Moderators haven't touched it, except that we turned off the user flags on the story. Otherwise it would be displaying [flagged] at the top.

We moderate HN less, not more, when YC or a YC startup is part of the story (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...). This doesn't mean we don't moderate at all, of course; that would be too large a loophole. But we do moderate less.

For example, under normal circumstances our standard practice would be to downweight a thread like this one, because outrage posts (what we used to call riler-uppers*) are not normally on topic for Hacker News. We also might edit the title to be less sensational and more accurate and neutral. And we would leave the user flags on. In the present case, we've overriden all of that in order to follow the rule I just mentioned.

* https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

> We moderate HN less, not more, when YC or a YC startup is part of the story

I believe this because it makes sense to me.

I think a habit of tempered reactions allows the understanding - that criticism is valuable in healthy doses.

Adding this: Having HN's stance plainly said is valuable too.

I greatly value and respect your moderation work. That consideration made me temper my words in the title and the post’s description. I did not create the thread with the intent to cause a flamewar, and in fact didn’t even realise the scope of the story until the very first comment.

https://news.ycombinator.com/item?id=43170866

Everything I wrote was the conclusion of original research based entirely on what I read from the founders’ words and what I saw in the video. Apart from the Mastodon post, I saw zero opinions on the matter before posting. The more I looked at it, the stronger I felt the criticism was warranted with those specific terms.

I genuinely do not feel there is a more “accurate and neutral” title I could’ve written. From everything I found, the title looks accurate. As for neutrality, which could surely only amount in this instance to being apologetic, I don’t feel there would have been any point to making the original post if not for the criticism.

I appreciate your reply so thanks for posting it!

I'm biased of course, but when I read "$X is backing $Y to do $Z", I read $Z as what $X is specifically and intentionally backing $Y to do. In this case, that's not accurate. YC would never specifically back a startup to "abuse factory workers"—that would be absurd. Nor is the phrase neutral; it's obviously accusatory.

HN titles aren't supposed to be used to build miniature prosecution cases. On some other sites, it's fine to use titles that way—it certainly maximizes attention, which is great for those sorts of forums, but on HN titles are supposed to use neutral, descriptive language—and then people can use the comments to express their interpretation and their views.

(Edit: Actually, even on HN it's sometimes informally fine to do that too—it is a grey area but sort of light grey, if you see what I mean. That is, sometimes a submission will use a baity title, this helps the story get attention and make the front page, and then moderators will 'sand' the title back down to neutral language and the thread can stay on the front page. We do that when it's a good story for HN—if it's not, we'd be more likely to downweight the article off the front page. The trouble in the present case is that neither of those options are available to us when the story is about YC and/or a YC startup.)

I haven't studied the details so I don't know what a neutral phrase would have been in this case. Some commenters have suggested things like "production line monitoring". I'm sure it's true that technology like this can turn into a Taylorist nightmare, but there surely are also ways to use it humanely, so in this case the real story seems to be that the founders published a video that backfired badly on the internet. Hopefully they will learn from this and not only alter their marketing, but maybe even change the product to be less likely to be abused.

But that's just my superficial take and like I said, I'm biased. More importantly, I appreciate that your intentions were better than I assumed they were and am grateful that you clarified that. Even after years of doing this job, I'm constantly still learning to make fewer assumptions. It's crazy how hard that is!

> I'm biased of course, but when I read "$X is backing $Y to do $Z", I read $Z as what $X is specifically and intentionally backing $Y to do.

Fair! I agree with you and I hadn’t realised the sentence as is was accusing Y Combinator in that sense. My intention was something along the lines of “Tell HN: Y Combinator backing AI company whose product aims to abuse factory workers” but it was too long and some nuance was lost in the trimming.

> Nor is the phrase neutral; it's obviously accusatory.

Again, I agree. But as I mentioned before, I also don’t see a non-critical stance which would make the situation worth commenting on, and I do believe it should be commented on. The only reason it’s worth commenting at all is because it’s bad.

> Some commenters have suggested things like "production line monitoring".

Which is a phrase so sterile and euphemistic it immediately made me think of George Carlin’s bit on soft language.

https://youtube.com/watch?v=o25I2fzFGoY

Also, I disagree with that description because it fails to capture the point (see below).

> I'm sure it's true that technology like this can turn into a Taylorist nightmare, but there surely are also ways to use it humanely

The crux of the matter to me is that this is inhumane by design. It’s not just the video, their page on YC’s website makes it patently clear via the written word they care about factory owners but not the workers. This is neither a marketing problem nor an issue of bad clients abusing the product—the founders plainly described how they intend the product to be used, and that use is inhumane. Which is why I criticised the company and the founders above the technology.

> Even after years of doing this job, I'm constantly still learning to make fewer assumptions. It's crazy how hard that is!

Believe me, I understand. Over the years I have also had to do a lot of internet moderation myself, and you are a big influence on my approach. There is little doubt in my mind that your intentions are pure and you genuinely care for what is best for the community. If we had one of you for each internet forum, the world would be a better place. Thank you for all that you do, and for your commitment to remaining humble and fair.

The fact that you don't see how that title could have been more accurate and neutral is the problem with our world today. People think their opinions are fact, and their pre-existing political leanings are neutral objective reality.

It's the word "abuse". Even in that very poorly-produced video, there was no abuse. Bad word choice at worst. There's nothing neutral about the assumption of abuse.

And "Y Combinator backing..." makes it sound like YC is on board with "abuse".

Here's a neutral, accurate title: "YC24 startup Optifye makes software to optimize factory labor that could lead to abuse."

[flagged]

I used to run a small factory (two assembly lines 10 people) and something like this would have been useful, not to force people to work harder but to optimise movements and points of friction. I would actively encourage and reward people for making suggestions and we had a process in place to test if changes made things better (and not just faster - we included easier, simpler, more enjoyable etc in the test)

Sadly it’s not about the tool in this case, it’s how it’s being promoted and positioned. The line “know who’s working and who’s not” on their website says it all sadly.

Exactly my thoughts, if this was a European demonstration they would be talking about providing training to the worker on the line to improve the efficiency.

I have no idea if that’s how they run factories in India but it definitely looks very off from Western perspective. In Europe the assumption would be that workers are trying their best, but they don’t know better. In the video, the assumption is that the workers are cheating or slacking and this tool can help you detect bad actors.

The tool itself at core seems fine though. Any privacy or abuse risks would be up to the legislature and operators to consider but it’s nothing unusual to monitor production lines.

I'm not convinced it's really that useful. If you have rows of people sewing the same piece of clothing or similar, then it's not exactly hard to track output without digital surveillance. If you have assembly lines where people perform different parts of a larger project digital surveillance is useless next to things like lean, and again, bottlenecks will show themselves extremely fast if a single station is slower than usual.

I suspect the purpose of these systems is actually to create a horrible work environment. Not because that is a good thing, but because it'll drive away anyone who has other options and leave you with the most desperate people. Who will also be the cheapest people.

The only thing truly embarrassing is that nobody at YC advised them that this ad is 1) terribly amateur and 2) has the chance to look extremely bad. Did not advise and in fact published to their feed. Weird optics, I am not familiar with what I presume to be Indian factories, other than the wild lack of safety (thanks youtube) but I cannot see how this video does anything positive.

As for the product itself I don't think it is unusual, these types of measurement systems are not new and can be helpful for a factory, like all things, it boils down to the owner/managers of said factory not the tool.

If anyone is reading from YC, I'm happy to review their submissions, I was perhaps the first who raised concerns and was on top of the thread with my comment.

(See this post I made on LinkedIn about this: https://www.linkedin.com/posts/crufter_today-y-combinator-de...)

I haven't read the details or watched the video, but it does sound like you noticed something that should not have been there.

I noticed that your page says this:

> New: the thread "Tell HN: Y Combinator backing AI company to abuse factory workers" on Hacker News got removed from the front page despite having 242 points.

The thread fell in rank because it set off a software penalty called the flamewar detector. We haven't taken any moderator action or touched it in any way. Our usual practice is to downweight shallow outrage posts (because they're not what the site is for), but we haven't done so in this case, because we have a core rule of moderating less, not more, when YC or a YC startup is in a story. (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...)

If you want to be fair, you could add that information. Otherwise, readers will assume you mean that we nefariously censor negative posts about YC, when the truth is the opposite.

Edit: after I posted this, I noticed that the thread said [flagged] at the top. That's because users flagged it—as they usually do with shallow outrage threads. However, I've turned the flags off in keeping with the rule I just mentioned.

> If you want to be fair, you could add that information. Otherwise, readers will assume you mean that we nefariously censor negative posts about YC, when the truth is the opposite.

I'll do that, sure! Dang, I have deep respect for both you and HN for many-many reasons (HN literally changed my life and made it 1000x). So don't take this as me having a grudge. I just like accountability.

> but it does sound like you noticed something that should not have been there.

I don't follow this, sorry.

Thanks! I appreciate the kind reply—a nice change from how these things usually go. And I'm grateful to hear that you feel that way about HN. (What you say about how it changed your life is true in my case too, which is one reason why I'm doing this job.)

> I don't follow this, sorry.

Oh, I just meant that it sounds like you spotted something (the video in this case, I assume) that should not have been on an official YC launch page. I was being careful with my wording because I haven't looked at any of the details and haven't myself seen the video—I'm only going by the reactions that people are having to it.

Believe it or not, founders don't run every single thing they do by the YC partners!

> Believe it or not, nobody suggested you need to run everything by a partner!

Believe it or not, nobody suggested you need to run everything by a partner! I would expect someone to at least watch a video once before posting it in a congratulatory post on YC's feed. It is not a criticism of startup culture but watching that video I genuinely question how far quality has fallen?

I don’t think it’s fallen, I think they just repost press releases and 99.9% of the time it’s uneventful.

Still, did no one watch the press release video upstream of this post? Either they are posting garbage press releases without really reviewing the content, or (worse) one or multiple people genuinely saw no problem with it.

I’d bet on the first. I don’t think it’s really that problematic, that’s just HNs reactive group political leaning that we all saw in regards to PGs last essay, but it is extremely poorly-shot.

You’d have to be pretty extreme to watch that video and come to the conclusion in the title of this thread. You would only need to have ever seen video before to be like “man these guys didn’t even just google ‘videography tips’”.

Sure, but it seems like this was posted to (and now deleted from) the @ycombinator X account with a "Congrats on the launch!" message.

> like all things, it boils down to the owner/managers of said factory not the tool.

Tools are created and optimised for a purpose. If you gouge someone’s eye out with chopsticks or a spoon, their inventors are hardly at fault. If you kill a bunch of people with an AK-47, its inventor doesn’t get a free pass.¹

These guys clearly built this tool to crack down on factory workers. They aren’t excused when managers use their tool for the thing it was made for.

¹ https://www.npr.org/sections/thetwo-way/2014/01/13/262096410...

A more apt example would have been Nobel with TNT, as it has more legitimate purposes. This software could be used legitimately.

And yet Nobel was so upset about his legacy of being the explosives guy for WW1, we still reward scientists with money and an award from him.

Yes, that's my point

Your point is mooted. I’m both cases, an invention has far more potential negative applications than positive ones.

How is it moot? TNT was a massive safety boost for the mining industry. The negative uses were developed/adopted after it was invented.

"an invention has far more potential negative applications than positive ones."

Emphasis mine. Do you mean to say all inventions have more potential negative uses? Even say a toothbrush? I suppose that could be true, but then you have to look at magnitude of impact and rate of occurrence to see that the potential negative uses like stabbing someone is low.

No, I was referring to dynamite and to the startup at the center of this controversy.

> This software could be used legitimately.

But it wasn’t designed to, which is the point. The creators had a specific goal in mind, and that goal is exploiting factory workers to the benefit of the owners. This is clear from their written words and video. Like with any tool, the goal influences the design and the design influences how people use it.

Seems like management could use the tool to reward high performers. It's all about how it's used.

saw someone point out how they could have framed it as a way to reward good workers, but instead they showed how they can go after people underperforming. Just bad framing and shows how out of touch they are

If we're being real here: it's excellent framing for their customers who very likely does this but slower (and with an equal amount of empathy). The framing is excellent in terms of the basics of pitching: know your audience identify bottlenecks, present a solution to remove bottlenecks, and clearly demonstrate how.

The optics of presenting this in a public domain full of people who relate more to the factory worker than the middle manager is indeed tonedeaf, though.

Are factory managers that dumb that they can't figure out what its actually for?

Out of touch? The people offended by this are not their target customers. I imagine there are far fewer managers sitting around thinking, "I need a way to identify my good workers so I can reward them" than managers thinking, "I need a way to tell when my workers are not working."

I think people are thrown off by the Indian work culture shown in the video, which is very hierarchical and frequently involves parent-like scolding. That said, I think that culture is horrible and should be eliminated.

I worked at an Indian consulting firm as the only American and experienced this firsthand.

It’s pretty wild when you witness that stuff after working in “traditional American tech companies” your entire career. That and the absolute abuse of metrics like story points for all kinds of things ranging from financial reporting to employee performance evaluation.

It’s fucking wild. Makes you really appreciate how good some places can be.

Why do you consider the use of story points for those 2 cases "abuse of metrics"?

Many parts of India have a much more chilled out approach to work. The work experience that you describe is much common in North Indian companies or where North Indians are managers. Looking at the names of these kids it seems to be the case as well.

Tbh this isn’t that crazy. If you hire someone to do their job outputting 10 items per hour and that number is reasonable because a bunch of other workers you hired for the same job are doing it and 1 guy hits 1 per hour then that guys shouldn’t be doing that job.

The outrage should be focused on the absolute meme of their ad video cuz they were like “lets literally have a convo with an individual but refer to them as a workspace and have them say human painful responses but then just shit on them anyway impersonally”

The product is not crazy. The video is wild.

Continued rant :

It’s kinda like a ruler. If you measure workers so that one’s doing 10x less/worse output than the average that’s good.

If you compare workers down to the .01% difference in output that’s stupid and inhumane.

Yes, but many business owners/managers are stupid and inhumane.

Yeah. I did new car prep for a dealership as a teen.

Ask me how many cars I did per day: totally reasonable. Make me clock in and out for every car I work on: bad.

> Tbh this isn’t that crazy.

Yep, seems like a bog standard accountability / performance management.

> The product is not crazy. The video is wild.

This is how it all starts. Sane solutions wielded by madmen.

Because this will definitely be used only to innocently tell off people doing 1/10 the work of everyone else, and not micromanage and hound people to increasingly unrealistic standards in already desperate conditions.

Safe to say you aren't in any position where every move you make will be watched by AI and analysed for faults so that your boss can scream at you more efficiently whenever you don't meet standards for their pitiful wages.

Do you really think this tool is making folks micromanage and abuse employees or perhaps they already would be doing that and this tool helps it?

There can be real value in these types of tools, its ultimately up to the implementation and I don't believe this tool will somehow make a happy work environment into an abusive one, the abuse will have most likely already existed.

Absolutely I expect it to be used to micromanage and abuse. Yes those behaviours already exist that’s why I know a tool that enables them will amphifly them

> and this tool helps it

And that is bad.

> the abuse will have most likely already existed.

The abuse will get worse. The correct ethical answer to “the conditions are bad” is “improve the conditions” not “make them worse”.

picture this: corporate buys something (like O365), and is reluctant to end licensing for the bundle. So... if they're locked into a contract that includes management-abuse-as-a-service, enabling bad behaviors, do you think they'll back out of enabling that one abusive manager out of five? how will that impact the workforce?

Maybe the implementation could include an option to show the worker's name.

>or perhaps they already would be doing that and this tool helps it?

Yes. I don't think we should ethically encourage the abuse of workers. And that official lens of marketing can and will shape who reaches out, even if the tool can indeed be used ethically. Framing is key.

As a tame example: think of the graphic design and marketing of red bull vs Monster. They have the same basic ingredients and purpose but that simple red bull design vs the in-your-face punk-esque vibe of Monster will change who buys it, how they identify with it, and even alter the perception of how it tastes.

The incentive is almost universally to micromanage and abuse employees. Reducing the friction will increase the incidence.

It's also dumb from a factory prospective. Our factories did time studies to understand things. What we learned:

Certain lines are primarily made up of barely functioning older people. No one else sticks around in those jobs. Think barely functioning alcoholic or recovering alcoholics that have nothing. However we would also get a few 18 year olds with no idea how jobs/work works and or zero accountability (they just ghost jobs).

From the numbers we should want to build our processes around the high performers. But we can't expand our base of high performers AND they are the most likely to just disappear and not easily have their productivity replaced.

So yes, it was correct that 10% of our people outperformed by 10X, and yes, it was smart to not try to improve that but to understand reality.

> From the numbers we should want to build our processes around the high performers. But we can't expand our base of high performers AND they are the most likely to just disappear and not easily have their productivity replaced.

You're failing to retain high performers? Are there perhaps methods for retaining high performers that you have not tried?

AI for Executive performance monitoring would be an interesting social experiment.

Shouldn't the manager of the 'bunch of workers' notice the guy is underperforming and understand why ? Maybe that manager is the one that shouldn't be doing that job

And they do that by either looking over your shoulder (1 person at a time) or collecting metrics on the entire team and the output. Both of these have different downsides.

The biggest issue is leadership or managers always wanting the number to go up from the individual. "We need 12 widgets per hour instead of 10 for just this one quarter bro" but then that becomes the new norm and eventually "We need 14/16/18/20 widgets per hour..."

It's boiling frog management that makes people distrust managers doing any kind of performance monitoring

yah this is exactly what labor law says in some countries: a manager standing behind your desk? ok. a machine surveilling you? not ok.

Yeah it’s a matter of scale. The manager can’t stand behind your desk and watch you all day every day. AI can.

But that would increase their cortisol, according to the founders.

It's clear you never worked in a factory and you have just as much empathy as these CS grads. This kind of thinking why I hate capitalism so much.

I worked in a factory multiple times and I can tell from experience nobody needs a stupid performance measurement like this. Your manager will make sure you work you ass off. Or you work with a big dangerous machine so you have to pay very much attention all day. Of course not every factory is the same, but putting even more pressure to factory workers like this is just inhumane and the most capitalist move I can imagine. Next step is to put robotic whips next to the lines and when their productivity goes below a specific value hit them automatically... Literal slavery.

Looking for 10x discrepancies is not how this will be used, and you know it. Adoption of this sort of tech is going to lead to Amazon "peeing in bottles" situations. It's wild how much faith people have in the ethics of business owners, especially the ultra-wealthy.

Your example sounds reasonable but it's not realistic: The actual use of this type of tools is to intimidate those workers who have outputted 9.8 items instead of the average 9.9 over the past week.

This is how our society ended up making Amazon delivery workers urinating in fucking bottles inside their trucks.

You've spent 32.8 seconds urinating this shift, 1.3 seconds longer than average. Your bonus has been reduced until performance meets standards. At Amazon, the customer comes first!

Congratulations! This week your urination schedule has improved. Enjoy a Kindle credit on us, good for any ebook license under $2.99. At Amazon, our partners come first!

I wish entrepreneurs would stop trying to make Manna[1] a reality. It's supposed to be a dystopia, people. Stop trying to create it!

1: https://marshallbrain.com/manna1

Fun story, thanks for sharing! Seems pretty inevitable... the Elon Musks of the world would love this. Luxury living for the 1%, fully automated indentured servitude for the rest of us.

But at least we'll have nice text to speech in the headsets, hopefully.

Kind of like the movie Elysium but without the space station. Somehow, making dystopias into reality seems like a sort of weird moral imperative to a lot of tech people. I guess techies (wrongly) think that they'll somehow end up on the side of the 1%.

As long as we have hierarchical social systems, there'll always be people happy to step on the 90% if that means they can be part of the 10%. And many tech workers do fall into that bucket :/ As a group, we're not exactly known for our ethics or social benevolence...

The product is shit. At some point we have to put human decency first, and not lock people in AI-surveilled setups.

If you're worried about some guy slacking, then hire some supervisors, ask them to chat with the employees, understand what their issues come from, if they had a rough week and they need some slack, etc.

But doing it impersonally through an AI is inhumane.

Imagine if you have kids and we replace all their teachers with just a camera and an AI, that "teaches them to write" and then nag them if they're not "as good as their classmates" or whatever... that would be insane. Kids are meant to be loved and grow with care.

Well, guess what, grown ups too. Job or no job. We're not here "in order to make money for the bosses", but just to all contribute in a just and useful manner to society. And so workers need to do their part of the job, but bosses need also to respect the humanity and decency of workers. It goes both ways.

Yet this is exactly how educational curricula are being developed right now - AI generates lesson plan based upon standards (easier for teachers), teacher feeds in reading assignment, etc., AI generates quiz and grades responses. Maybe there's human oversight, maybe not, but the education industry is largely embracing this AI control layer without much hesitation because it reduces cortisol for everyone except students.

When the only way you’re able to see education is through the lens of job training and value production, you won’t give a shit about the experience of students. How can one measure the students’ suffering in terms of dollar bills? Same attitude used everywhere, but more egregious because it’s done to kids.

So you’d be ok with this AI being applied to other industries? Maybe lines of code?

I mean, if you're gonna make this kind of software to track individuals, you could, at bare minimum, throw a name in there so that the supervisor doesn't have start the chat with "hey number seventeen".

If literally having your identity reduced to a number in the process of getting yelled at for not being productive doesn't seem bad/crazy to you, I'm guessing there isn't anything I could say to move you.

This is idealistic. In reality, we have more control over tools than how they are used - in general we have more control over changes earlier in the pipeline. Precious little control, but more control.

We already advanced beyond Taylorism's myopic time-motion glorification. This isn't a tool to improve the process, but to push the employee towards the meat grinder rather than look for a more intelligent approach (matching an employee's tasks and abilities rather than just giving him a red rating and ugly performance numbers)

This is normal Indian work culture. You can also guess in most cases, the factory workers are a different caste from the managers.

Is it dystopian, or is it just real-time performance monitoring poorly marketed by inexperienced founders?

There are tools like this for tracking git commits and velocity (that I’ve been on the receiving end of). It probably makes less sense in that context, but if your job is a repetitive task, I don’t think it’s necessarily abuse or dystopian to track it.

Monitoring bottlenecks isn’t a bad thing. They probably could have chosen an example where the solution to the bottleneck didn’t involve berating a low performer (e.g. adjusting the line to add another station or similar)

You mean if they marketed it differently, managers would magically stop using it exactly the way it was advertised?

Assuming that this comment is made in good faith — I doubt that they have much usage currently. Better marketing would show how managers could use this tool to better support their staff rather than scolding them.

Perhaps real-time performance monitoring of people is wrong as well and marketing just made it sound less bad as it is?

If your management tracks your commits/velocity, look for a better job. The market easily allows you to do this. It doesn't make sense anyway and anyone with some technical know-how knows this.

This will escalate into something bad like publication metrics did to science. You end up with superfluous drivel that just pads real discovery. Although for code it is even more trivial.

Bottlenecks are something entirely different. If you have a critical-path, there need to be resources to tackle the task at hand. Real-time monitoring is entirely useless here as well.

> If your management tracks your commits/velocity, look for a better job.

Don't most version-control systems support/automate this?

It also tracks LOCs, but people learned that it is a bad metric. Velocity is basically the same, just with advanced stupidity.

But no, I have never used an environment where velocity is tracked. Of course my source control does provide data how often and what people do commit and you can use that data to create a meaningless number.

Some systems like Jira tracks something like Sprint velocity. Not much of a fan here either, but if used correctly, it is oriented towards a goal. That is the metric you should really track to see if you are still on course.

The product should be marketed as a tool for monitoring and highlighting bottlenecks in the manufacturing production line in order to help maintain peak output. This is a completely reasonable thing to want, it's no different to monitoring micro-services and their latencies/loads.

The video they made however where they berate and meanly put-down an individual employee is so far from acceptable. That's not how personnel performance issues should be managed in the real world, completely void of human empathy. It shows the founders (and did YC view and approve this?) are lacking in areas

This sort of performance management is unfortunately necessary. The problem is that we need tools for it to be built by people who can empathise with those subjected to them, and who want to do the right thing, and not these sorts of folks who are too immature and inexperienced to get it right.

My previous company ran a warehouse and there was a clear bell curve of productivity. Most people were fine, some were excellent, but some were below the level that was realistically achievable. We did careful and considerate analysis and it helped improve productivity.

When done badly however you end up with management using productivity tracking as a lever to increase productivity across the curve. Amazon driver delivery quotas are a great example – people urinating in bottles is clearly a symptom of the quota being too high. Unfortunately software built naively to help bring up the bottom 10% can too easily be used to force up the productivity of the other 90%.

Necessary according to who?

Colleagues, the business. The important things are that performance management is a) used to identify those performing far below what is possible, and b) that there is a good faith effort to support those people to perform better.

Both of these things are mis-handled by many companies, who will use it to encourage more performance out of those who are already as productive as practically possible, and that the tools are used in bad faith or with no intention of helping people, only as an excuse to fire them.

You don’t need constant monitoring to find the bottom 10% though. People performing that badly are easy to pick out with spot checks.

It’s the same with software. Your direct manager should know if their employee is in the bottom 10%. If they don’t, they aren’t doing their job or they have too many reports.

What constant automated monitoring allows is cutting back on the number of managers and like you said, pushing high performers even further.

That depends what you consider monitoring to be. Managers don't just know these things, they aren't walking the floor watching people, they're busy doing things. Their source of knowing these things is often looking at data for the number of things processed, or in some cases failure rates.

You're right that this doesn't need to be constant or even that automated though.

Investors.

The video shows two people whose full-time job is watching an automated dashboard and who have failed to spot an issue for at least a month.

Quite apart from anything else, this isn't a good argument that this type of performance management is necessary, it's a great argument for firing all the managers in that factory.

We have a production line without individual performance tracking in a country with very high living standards and working costs.

If a line doesn't perform, you can solve it without it. Every single time.

Sure, there is not as much market pressure than in general logistics, but I still don't believe you would need these metrics at all.

Brought to you by the VC famous for InstallMonetizer? Make no mistake, it’ll basically back anything that makes money, there’s no moral high ground. And like it or not, this kind of AI (or should I say A-eye) is here to stay.

Did YC fund (or otherwise back) InstallMonetizer? The way I understood it, they applied to YC with a different idea and worked on IM earlier/on the side/later. (In which case, the controversy was more whether YC works with people who make things like InstallMonetizer.)

> Did YC fund (or otherwise back) InstallMonetizer?

Yes. Quoting Paul Graham:

> Last week there was some controversy online about a company we funded called InstallMonetizer.

https://news.ycombinator.com/item?id=5092711

Here's his clarification that followed, saying they didn't fund the product InstallMonetizer:

> They're working on something new, and all the office hours I had with them were about that. They're not even in our database of companies as InstallMonetizer but as the new thing. (I'm not sure if I can say the name because it may not be launched yet.) I knew they had some previous product that was called a Windows installer, but I don't think we ever talked about what it did. https://news.ycombinator.com/item?id=5093047

That's deeply untrue and unfair, and the fact that you've reached back 12 years to find an example, when YC has funded thousands of startups since then, is evidence against what you're saying.

YC has turned down many startups because they didn't think what the product was doing was right, and it has defunded and banned startups for ethical reasons over the years. It's not your fault that you didn't know this, since no one makes a big deal of it, but I think it's proper to bring it up when you say that the people at YC will "back anything that makes money".

I deeply respect your work here on HN (and I semi-frequently email you so you know I do), but I disagree here. I don’t have to go back 12 years, I think YC funded its fair share of crypto scams like other VCs (of course what’s considered a crypto scam is different for everyone so no point in arguing), but I picked a very well known example that most old users probably know. I should probably mention I commented on a Launch HN thread this month that’s clearly illegal by design: https://news.ycombinator.com/item?id=42886294.

I don’t have much against VCs investing in whatever makes money, to be honest, given how immoral the top dogs of this industry are to begin with, and I can’t claim I haven’t worked in an at least somewhat immoral company.

Edit: I only saw the second paragraph after a refresh. Sure, what I wrote about “back anything that makes money” was inaccurate so I should retract that, but I stand by the “no moral high ground” point.

Ok, I appreciate the reply and am sorry about editing while you were replying. (Sometimes I leave a tag in my comments while I'm still editing them, but I didn't do that this time.)

No problem!

YC is funding weapon manufacturers lol. Where are the ethics there?

People have different views about what is ethical. The HN community itself is divided on the ethics of things like weapons manufacturing (a.k.a. "defense companies") or crytpocurrency (mentioned elsewhere in this thread). Having different views about ethics is not the same thing as having no ethics.

This is a strange view.

Weapons development is at best helpful as it hopefully provides a deterrent to nefarious actors and at worst protects our interest as Americans if someone caused us to move beyond weapons as a "deterrent."

Pretending like national defense and development of weapons is a bad thing is hiding your head in sand and pretending that there's nobody that wants to hurt you.

> protects our interest as Americans

Whose? Not mine.

To me, yours is the strange viewpoint. We have nukes and a massive air force. That is the bulk of the deterrence. Use good statesmanship and diplomacy and that covers the rest. Anything else being developed is for political suppression or fueling regional wars and imperial bullshit.

And that's fine, but they're solely yours. 57% of US citizens support expanding the military.

Further you have no evidence to back up your claim that anything outside of the Air Force and nukes are all that are required to properly deter threats.

I would imagine that you also have no credible evidence that "Anything else being developed is for political suppression or fueling regional wars and imperial bullshit."

If you don't want to be involved in developing weapons then that's fine, nobody is forcing you to do so. To claim someone as anti-ethical because they chose to do so in a legal way may be your opinion but surely not a fact.

> We have nukes and a massive air force.

> YC is funding weapon manufacturers lol. Where are the ethics there?

Circular reasoning? The reason we have a massive and more importantly effective air force is because weapons development occurred. It would not take very long at all for China to surpass the US in terms of air effectiveness if the US stopped working on defense projects.

We have far more carriers though

There is no circular reasoning here. I said funding weapons development is unethical, guy responded saying we need weapon development for deterrence, I point out that we have had plenty of effective weapons and the new ones will not be used for peer wars but for gladio/terrorism or regional wars.

Further weapon development just fuels arms races and puts the world in a worse place. That is why I find it unethical to fund it.

Also, btw, China already surpassed us despite all our spending. They are defining what a 6th gen fighter is, after all.

dude, YC funded stablecoin scams, including one that lost everyone's money in Terra-Luna.

The response here would be significantly different if this was about measuring the performance of software engineers in wealthy countries.

I mean… yes, obviously? There’s a lot less room for exploitation of white collar workers in wealthy countries than sweatshops. Though performance quantifying software still deserves a lot of scrutiny for removing humanity from work in any context.

It's obvious to you becase you seem like a decent human being, judging from this comment.

As someone who grew up in a 3rd world country and whose mother owned a clothing factory, this product seems...fine? The response is an indication of how little people know about how their t-shirts and shoes are made.

It's nuanced. If it allows you to find outliers (low performers to manage and high performers to praise), that's fine. If you try to push everyone further and further to their breaking point and make them trade the same amount of money for more of their time and more importantly health, it's certainly not fine.

Ah ok, my bad then, it's a perfect product and we shouldn't change anything at all then. Let's continue to treat humans as machines, especially in 3rd world countries, who cares right ? Even they say it's fine.

TBH I'd rather buy shirts where I know that people doing them have been well-treated

The people who worked in these companies are usually incredibly grateful for the opportunity to do seated work indoors for reasonable wages.

Applying Western labour practices to third world countries would prevent them from ever developing...hurting the very people we all want to help.

Applying those practices to western employment will mean no one can afford what those grateful people make.

But the lack of the practices shouldn't give the owners free reign to better themselves more while depleting others in the name of "wanting to help".

Looool "those people should thank us for colonizing their countries, we brought them road"...

No, we know, and we're not okay with it either. I understand it's often an improvement over other options for employment. That makes it understandable and even supportable to some extent, not okay.

Here in the US, we spent centuries fighting and dying for better options. Tools like this are used to launder the dismantling of the results of all that work through a fantasy of objective metrics.

I watched my mother spend 30 years building her company...it's hard enough to build a manufacturing company in 3rd world country. Applying Western labour standards would make it impossible.

The way out of poverty is to through. You need to create enough value to be able to afford the airconed offices where everyone sits on an Aeron with a macbook pro.

I'm sitting on a $70 ikea chair, in a flat reconverted to an office, right under the roof, with no AC and I have 1st gen base model m1

FAANG is an exception, even in the west

Yeah, I was going to say "through to what?"

Enough people have to have enough money to be able to buy the things made in a factory someone's mom spent 30 years on for that factory to exist. These tools are being used to dismantle the "what" people strive for. It's precarity as a service.

>As someone who grew up in a 3rd world country and whose mother owned a clothing factory, this product seems...fine?

People in western countries find things like sweatshops to be objectionable.

>The response is an indication of how little people know about how their t-shirts and shoes are made.

People in western countries are well aware of how their shirts are made, and don't like it, and try to avoid it when possible specifically because they find sweatshop conditions objectionable.

I'm not sure what data you're using to prove this but it seems like many of the largest brands in the US are still using sweatshop labor [1]. Many still have no idea of what's going on in Xinjiang [2]

1. https://yoursustainableguide.com/brands-that-use-sweatshops/ 2. https://whatishappeninginxinjiang.com/brands-linked-to-xinji...

Are you trying to make the point that people don’t care or aren’t aware of the fact that their clothes are made in sweatshops by linking to a western blog extensively detailing the brands that use sweatshops with the implied point that you should avoid them because they are bad?

Sweatshops are bad. Westerns know this. They don’t want sweatshops.

Seeing YC back a sweatshop management application that uses AI to help managers harass their workers is sad. It would be similar to them investing in faster slave ships in the 1850s.

I'm saying that people _say_ they have issues with sweatshops but when push comes to shove they're willing to buy the Nikes.

I'm not trying to make a decision here on right vs. wrong just pointing out that most people _say_ they care but they're not really willing to do much about it.

I wonder if you'd think differently if your mother was a worker rather than an owner?

I worked in a factory during university- made playground equipment out of metal. Every task had a reporting system and the company knew exactly who was producing what. This was a long time ago - they still used desktop calculators with paper tape to do calculations. I was only there for a relatively short time, I just wanted to work and make money for school. We were promised productivity bonuses - my section of the factory achieved 147% of expected production. Another had done almost 180%, but when the bosses did the bonus talk they added in the trucking division and showed a 30% productivity due to excessive “dead heads” ( empty trucks ) returning from equipment delivery. It was a management issue- they were not very good at finding return loads for those trucks. So the production bonus was calculated on the total score of the enterprise- so we got almost nothing on the factory floor for bonus. When my time there was done a friend in accounting told me that I had rated as the most productive in the factory, he had seen the numbers as there was a ranking sheet that was calculated every week. ( remember paper tape calculators ) no one ever told me at the company, no real reward, no approach about coming back to the factory. The point of this post is that this type of thing - monitor the employees- has been a thing for a very long time. It is the humans running the factory that decide to reward, encourage , promote those who work hard - or not.

And the humans are also machines, purpose built over time to maximize efficiency and ignore anything else. Good employers are evolutionarily suboptimal, thats the issue.

> Boost your assembly line efficiency by up to 30%

Ethics of this aside the above claim must be dubious I would think the majority of manufacturing inefficiencies are due to down time as a result of raw material shipping delays or machine break down… of course I’m in no position to offer an informed opinion but just based on the product website I have a hard time taking this stuff seriously.

Monitoring of factory workers isn’t hard to do with current surveillance and 1 or 2 humans in the loop

Software can not abuse workers. Managers can, with our without software.

We've had automated KPI measuring tools since punch clocks. Nowadays it's OK in some companies to install remote access software to monitor employees' screens. It's nothing new. It's just collecting data. Question is, what will bosses do with this data, will they abuse or develop.

I have no hate towards those guys. No love also. It's just business.

The famous it's ok if it's done with algorithm.

> Software can not abuse workers. Managers can, with our without software.

And guns cannot kill people. Other people can, with or without guns. It shouldn’t be hard to understand tools facilitate the task you’re using them for.

> Nowadays it's OK in some companies to install remote access software to monitor employees' screens.

No, it is not OK. It is done, but that doesn’t mean it is right or ethical.

> It's nothing new.

That’s not an excuse to not do better and revise practices.

> It's just business.

See the Nuremberg defence.

That post was pulled partly because of my comment. I commented this:

"While I see the economical usefullness, this sounds like the worst possible application of AI.

Using AI to surveil is building hell on earth. AI should be used to help people work less/easier, not whip them into working more."

Which ended up on the top of the thread. Was surprised to wake up this morning and see it gone.

LinkedIn post I made about this:

https://www.linkedin.com/posts/crufter_today-y-combinator-de...

It's clear there's a large fragment of AI founder types that totally had nobody tell them and I hope they faceplant, hard.

I find the marketing interesting considering that this product already exists in other continents ... and it is NOT deployed in factories. It is deployed in office settings. If this our future? Lots of video evidence: https://www.youtube.com/results?search_query=+Using+AI+to+mo... Expect dishonest marketing by those aspiring to build these surveillance and anti-freedom systems in our western countries.

Why not deploy it across every moment of everyone's life, with algorithmic prediction of economically unproductive deviance and BadThink?

Maybe I'll pitch that to someone with money.

Larry Ellison is already pitching this.

https://arstechnica.com/information-technology/2024/09/omnip...

That's why we fight for social and worker's rights, try to pull that shit in germany

But for that you need strong unions and I heard it basically is communism, and I heard communism basically is hell on earth, I guess it makes whatever this software is doing heaven

Just like how workplace software analyze white-collar work, such as this Microsoft product [1] (and many of those "bossware" which we saw during Covid), I don't see how this tool is any different but from a manual labor analysis standpoint? Their marketing might have been off, but I mean as long as the camera recordings are securely stored, what could be the issue? This solution is more like an automatic supervisor, and this could help as it could heavily reduce the need for supervisor roles (which also can come with biases and potential workplace-politics), and thus directly increase salaries for the workers actually performing the manual assembly. Also, there are already cameras in assemblies for monitoring the environment and also modern solutions to analyze workers and worker safety such as Invisible AI [2] and Rolloos [3].

[1] https://www.microsoft.com/en-us/microsoft-viva/workplace-ana...

[2] https://www.invisible.ai/

[3] https://www.rolloos.com/en/about-us/mission-vision/

Update: I just saw the part where the founder in the video refers to the person as "Number 17", this is pretty low even considering they might have wanted to anonymize a person's "John Doe" name for the example. I don't think such empathy-lacking individuals deserve any serious funding to build this out.

Thank you, I’ve been seeing the reaction to the announcement but hadn’t yet found the announcement itself.

Can anyone share the vision models they might be using and how they might be tuning them?

Having been in this industry for a long time now, I see a disappointing trend in tech, and US society in general: viewing everything at zero sum.

I’m probably naive, but I remember in the past tech focusing on innovation that would generate enough gains for everyone to get a share (or at least the gain to the tech company did not come at the expense of someone else)

Now, more and more I see business plans that are zero sum. Using tech to take from someone else, not growing the pie.

This matches a general trend in public life is the US to view everything as a zero sum contest

There was a time when US tech people would use this idea as a moral parable.

https://marshallbrain.com/manna1

But now we have current tech businesses basing their future growth strategies as much on better dark patterns as on actual value creation. Imagine a society using their best minds to figure out 'better' dark patterns, and counting the outcome from those patterns as GDP growth. That society is royally f'd on every level.

It seems like there’s been a rash of these instances lately where someone does something, says something, or builds something like this that’s not just offensive but unconscionable. And there’s always the predictable outcry and then usually that’s the end of it.

It’s given me pause to think about why that pattern has been established and I think the simple answer is that there are no consequences. The people we see in the news doing horrific things for attention are doing it because there is no mechanism to hold them accountable. Product launches like this - where it’s explicit purpose is to degrade and exploit humans - happen, meme video and all, because these people will not face any consequences for it (and the potential benefit is massive to them if it takes off).

Yelling and screaming about how horrible it is doesn’t really do anything and it’s not an effective use of time or energy. I wonder what society could do - not conceptually, but practically - to establish consequences for someone who launches a product like this.

The secondary school the founder went to is a dead giveaway. Ofcourse yc would fund them. Anyone would fund them infact.

Overpaid software engineers here please spend a moment to talk about your 3pm massage and free omakase

Or just go work a real widget business and you will realize that optimizing worker efficiency is critical and useful

Imagine the features you could add to this. Like a robot that walks around behind the workers and gives well-timed corrective communications with a whip.

The whip is for the programmers.

https://sierraclassicgaming.com/game/space-quest-iii/sq3_pro...

So it appears that automating managers is easier than automating "unskilled workers"

An abhorrent product that no doubt will be have abhorrent customers and maybe one or two that will use it for good.

It’s not the only company with low morale that YC is sponsoring. There have been a dozens of copycats shamelessly copying other products, some of them open source.

It is disappointing to see YC going to new levels of bottom without any proper accountability, just greed.

No comment really, just came here to remind everyone about, "The Yes Men."

NSFW warning: https://www.youtube.com/watch?v=8_bWAF-XxQM

The video features a tear-away business suit to reveal a gold lame stretch suit with a giant, inflatable phallus that has a monitor at the end and controls to let a manager electrocute under performing factory workers from the beach.

This was presented live and in person to an unwitting, but credulous breakout session at the WTO in 2003 to awkward applause.

My first full-time job out of high school was unloading trucks of coffee and sugar, by hand.

Trust me when I say that the motto on the loading dock is 'fuck me, fuck you'.

I can't call this slavery but their attitude towards their workers who make $1 an hour is sadly extremely common for among wealthy Indian people. They just don't see the non-English speaking poor in India to be human, and very likely the people below their status in America to be human. It's disgusting and the reason why I can't identify with my ancestry

It's too bad the AI can't just do the sewing.

I don't predict much success for their venture

Science sans conscience... we haven't progressed one bit in the last 500 years

This seems like a tempest in a tea pot. You're automating doing what factory floor managers already do? Hitler! From the title I thought they had developed a robot that flogs 6 year old workers or something.

If I were YC, though, I'd probably have a rule about startups not using "backed by Y Combinator" logos on their homepage like Optifye does. YC's a pretty low touch investor at the seed round level, their startups could do lots of things YC didn't expect, didn't know about, and couldn't prevent.

I think people are scared this will increase the power of the companies over workers. But a worker’s negotiating position is based on their indispensability and productivity, and on average this won’t change that. I know in reality it's often the fog of war that makes the work bearable. And many good managers more or less consciously keep it that way. But fundamentally, how can more insight and more truth be bad? My resources are limited and I'd rather spend them on a good worker than a slightly worse one. I thought that was the whole point of meritocracy.

This kind of product is really shameful, and peak capitalism... looking at people as mere robots to serve your, disgusting

Apparently on their website they're asking for comments/feedback, so... you can tell them yourself what you think of their disgusting tech

> Let us know at founders@optifye.ai, and we’ll help them drop their cortisol levels :)

Peak capitalism would have the employees holding their own value and not being crushed in trash power dynamics to allow this kind of stuff. Capitalism is about nonviolent voluntary exchanges between two parties, when one party has a power dynamic skewed in such a way they can use tools like this that employees hate, then that’s not capitalism anymore.

That's approaching "real communism has never been tried".

apologists everywhere are all the same

so this is more like peak feudalism huh?

back to 1800's I suppose.

There is no seperation b/w private entities , the state and the church , all trying to exploit the middle class / lower class was one of the gists that I think when I recall feudalism sounds familiar ? Guess what ? We are living at one right now.

Funny that what you describe will never exist without heavy (extreme?) regulation and progressive taxation, which are anathema to most advocates of capitalism. Have you considered that maybe your definition of capitalism doesn't agree with the definition society has agreed upon?

yes yes true capitalism has never been tried. You realize you're doing the same thing hardline communists do right?

Nobody looks at people as robots. Robots are cheaper, do not require food or sleep, and do not have to be murdered when they attempt to unionize.

Robots are far superior.

But then if everyone is out of their job and unemployed ? who will buy the stuff if noone has the money.

I think there is a balance.

Otherwise its going to be 1984 in more than one way (the spying part is already there) (it would also do of that the countries are ready to produce things as much but they won't and limit it to create that constant mood of war to make people not question them / make them weak.)

I think capitalism has fallen. Capitalism is a good system but to its degrees. If you push the accelerator too hard , you get fuedalism.

and we are at feudalism. I am not sure if we can undo this. Let this sink in, the american dream , all our thinkng that capitalism being good and communism being bad fundamentally doesn't matter because we have entered a system where the lines of division are so blurry that they are practically nonexistent.

CGP Grey did a video on this topic about 10 years ago. Curious if the video aged well. [1]

[1] - https://www.youtube.com/watch?v=7Pq-S557XQU [video][15 mins][Humans Need Not Apply]

> Curious if the video aged well.

Just watched it, and I would say that it did age very well.

> But then if everyone is out of their job and unemployed ? who will buy the stuff if noone has the money.

Last time I tried to say something like that I got plenty comments calling me for reading too many sci-fi books... I guess some people just lack imagination and experience with exponentials.

Having a "job" is not some universal good. What people need is resources/needs met. Refusing to buy factory shoes because it puts cobblers out of business is a losing strategy. Better to support the former cobblers needs than to keep forcing them to make shoes.

Except I just want to say that we don't support the cobblers ?

Would they even be cobblers if they are out of business / are heavily impacted by it (like it wipes out 90% cobblers , making their lives actively worse , forcing them to go into the factory lives which actively sucks for them)

I am not sure this is a nuanced topic.

Not yet for many tasks. Humans are more flexible, easier to replace (nothing to install etc) and one fte in this type of work is 10 years of robot. Hope it will change soon, but it's not there yet.

Hm, how do you explain that humans are still working in factories all around the world then ?

Because there are still a lot of things robots simply can't do, like picking items and packaging them. Seems incredibly simple, right? It's not. It's also the reason we won't have self driving vehicles before the collapse of western society.

the fact that they’re backing this is ridiculous.

This product should be banned and these guys should be forced to work a year at a textile mill for trying to build it

Yeah it has terrible optics, yet it's clearly going to be normalized and come. The question is who does it and what is the organization of it. If this company doesn't do it, the next will.

In certain roles, AI micromanagement clearly will create higher performance. Add the marketplace of capitalism and it'll all compete away.

There are certain roles, like artists, where this is the wrong solution wholly: monitoring whether an artist is at her desk will create badly performing artists, and this will show. In these roles, these tools won't apply.

> Yeah it has terrible optics, yet it's clearly going to be normalized and come. The question is who does it and what is the organization of it. If this company doesn't do it, the next will.

Where have we seen this before..

In these roles, these tools won't apply.

There will be companies that will apply them regardless, even in roles where they'll make things worse. The incentive for managers to show 'a bias for action' often results in managers doing any action that they can think of rather than the right action backed by data.

In the US? Sure. In more developed parts of the world? Doubtful. European labor laws are already much, much stronger than their US counterparts, and most countries outright ban using cameras to monitor employees.

Reply on news.ycombinator.com