Hacker News new | past | comments | ask | show | jobs | submit | sbszllr's comments login

As someone who has no idea what this is about bar the landing page explanation, and isn't in this space -- it would be great if the front page had examples, or links to examples.

30 seconds of clicking around and I've failed to find sth compelling.


As someone who is a curious outsider to Modelica, it piqued my interest a few years ago when I found out that this language lets you write equations more directly than most other programming languages. Take the ideal gas law, for example, PV = nRT, which has 5 identifiers. In most programming languages, you'd have to keep only one variable to the left, that is assigned to, e.g. T=PV/nR, and a similar set of equations if you want to determine any of the other variables, given the rest. In Modelica, the same equation, expressed as you would in natural math, works for determining the unknown, based on the knowns.

https://mbe.modelica.university/behavior/equations/electrica...

I don't know much beyond this.


This is generally referred to as an acausal modeling environment - you don't need to bring the causality of mathematical relationships, just bring the relationships.

There are several other tools in this space, not the least prevalent of which is Mathworks' Simscape product line [1]. Wolfram has a solution that is also very similar to Modelica [2]. Finally, I believe that you will find ModelingToolkit.jl [3] worth a look (along with Julia in general) if your interest is piqued here by Modelica. I believe MTK is a big slice of the future of acausal modeling tools.

[1] https://www.mathworks.com/products/simscape.html

[2] https://www.wolfram.com/system-modeler/

[3] https://docs.sciml.ai/ModelingToolkit/stable/examples/spring...


Wolfram SystemModeler is a Modelica implementation.


Here are some examples. Looks like a simplified framework for writing physical and electrical system simulations.

https://mbe.modelica.university/


Over 20 years ago I modeled an internal combustion engine, an automatic transmission and a multibody chassis model all in a single model. IIRC, the model had something like 250,000 equations in it and it modeled combustion, hydraulics, friction, and 3D rigid body motion. It is capable of far more than simple models.

https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&d...


That’s incredible, I had no idea this was even possible. Are most auto manufacturers modelling their vehicles this way? Seems like an amazing way to search for optimizations, predictive failures, etc.


It is pretty wide spread in automotive. I think nearly all F1 teams use it (hard to know for sure since they are quite secretive, but it is very common in my experience)


Ah! You're the author of Modelica by example?


Yes. And "Introduction to Physical Modeling with Modelica". I also built the Modelica Playground (which I deliberately didn't link to because a thundering herd of HN readers would have crashed it).


Nice to cross your path, I got the first one to kickstart myself some years ago (and still struggling with evaporation / condensation).

Would you recommend the second to introduce a colleague to OpenModelica? He is into gPROMS but will lose access to the software at retirement.

Sidenote: invest in software with freedom to operate and good knowledge reusability.


Modelica by Example is the most recent and free, so I would recommend that.


Are you aware of any books like your "Introduction to Physical Modeling with Modelica" but for readers without a background in EE, math, physics? I am looking for something for a mediocre SWE like myself. It doesn't have to be Modelica; I could try learning MatLab or Mathematica, etc.


Well if you are interested in the intersection of software engineering and technical computing, I'd recommend Julia. I'm currently working on JuliaSim which is a Modelica like system built on top of Julia. So Julia might interest you as a programming language and then you could pick and choose what aspects of things like ModelingToolkit if you are interested in the engineering, math and physics aspect or you can just stick to the software/programming aspects of Julia.


I wouldn't say simplified. You can model a lot of systems with it. It's pretty good.


The first link for “Modelica Language” includes tutorials and examples. Five seconds of clicking around got me to:

> Let us consider an extremely simple differential equation:

x = (1-X)

Looking at this equation, we see there is only one variable, x

This equation can be represented in Modelica as follows:

  model FirstOrder
    Real x;
  equation
    der(x) = 1-x;
  end FirstOrder;
This code starts with the keyword model which is used to indicate the start of the model definition.

The model keyword is followed by the model name, FirstOrder. This, in turn, is followed by a declaration of all the variables we are interested in.

[et cetera]


Modelica Language -> Modelica By Example -> https://mbe.modelica.university/

Took me under 5 seconds to find.


[flagged]


I wonder if some of those words might be terms-of-art, written with a specialist rather than a generalist audience in mind.


Accessible to whom? I work in a field that uses things like Modelica and it reads just fine to me.


It apparently sells merchandising for their own brand though.


The Modelica Association is a non-profit that publishes the specification, the standard library and all conference proceedings for free. We sell merch in part because people in the Modelica community like the language and like to show it but also as a way to fund the not for profit activities.


Downvoted but relevant...nothing says corporate bullshit like a niche programming language with logos and merch featured far more prominently than examples or explanation...


Just to be clear, there is absolutely nothing "corporate" about that web site. It is a non-profit organization.


Sorry, I guess that was rude, I don't know anything about it, just a snap judgment of the landing page.


Another day, another nondescript product name on HN for Russian roulette clicking. Dare to say anything about it, get downvoted into a smoking hole in the ground (e.g. by webdevs who assume everyone else is), yadda yadda... I'm pretty convinced by now that it will never change.


I wonder if anyone is keeping a list of the new languages that come out every year.


Modelica is nearly 30 years old, BTW.


TIL. Thanks.


I’ve been working in the space since 2018. Watermarking and fingerprinting (of models themselves and outputs) are useful tools but they have a weak adversary model.

Yet, it doesn’t stop companies from making claims like these, and what’s worse, people buying into them.


“Think of the children” is, as usual, just to get the foot in the door. They use it as a justification, because it works.

Of course CSAM is bad, shouldn’t we do everything in our power to prevent it? If you implement client-side scanning, you will catch some rookies. Some old pervs that don’t know how to use encryption manually, or use Matrix. They will use them to show how effective the system is…

with the exception that it doesn’t work against anyone who knows anything about computers. And I think the regulators know it, they aren’t dumb (imo). It’s, like I said earlier, an excuse to expand the scope of scanning later.


Yes, it's already begun, even though the directive isn't even ratified yet: https://balkaninsight.com/2023/09/29/europol-sought-unlimite...

Europol wants unfettered, unfiltered access to all scanned data, regardless if there's a crime or not.

And they want to inject all of that into their Police AI (which they also want unregulated).

It's going to be awesome future.


And of course in the released minutes the details of which idiot made which claim are redacted.

So much for the transparency and accountability they’ll no doubt promise will be there for the process of accusations (not that this makes the idea any better, useful, or more palatable), which need not apply to themselves.


This is standard acces to document request protocol across Europe. You are not going to make your staff targets of the internet mob (see Trump and the names of jurors). You can deduce these were likely actually low level staff (contrary to what the article claims) as names of actual high level staff would normally not be blacked out, although I don't know Europol, as a police body they might have different safety protocols.


Sorry this is not quality journalism and you misunderstood the message further.

1. The meeting tool place after the commission made it's proposal, meaning that contrary to the way the article sets it up, the meeting couldn't have shaped the proposal. 2. The screenshot of a meeting report states that Europol wants access to the same info as Member States for specific cases, contrary to your summary it doesn't say anything about access to all data. 3. That police agencies want to include further areas into the legislation is not unusual. That doesn't guarantee it will happen, nor does the police body speak for the executive or legislators or represent the EU views as a whole.

I do think the proposals go a bit too far, on the other side the whole tech world assumption that anything has to stay lawless is just absurd. No one can deny there is a problem with pedophile material and to say to protect the purity of free speech all such issues have to stay unaddressed is just a position blind to reality.


> No one can deny there is a problem with pedophile material and to say to protect the purity of free speech all such issues have to stay unaddressed is just a position blind to reality.

This is not about free speech at all. Free speech is about the government not censoring public speech and publication. Publishing this kind of material is already an exception of free speech and nobody disagrees with that.

This proposal is about the private communication of all citizens. Not only is it a disproportional measure, it's also ineffective. It will not reduce the demand for this material and will only stop a particular method of distribution. The bad actors will just move to other methods including non-digital or steganography, on unmonitored or hacked devices. Just like the war on drugs doesn't work, they will find a way.

But in the meantime this will deeply undermine the privacy of all people and this tooling will inevitably be used for more purposes than originally proposed. As this linked article shows.

And in order for this to work it will basically have to make FOSS operating systems illegal because as long as the user can modify their OS they can remove this scan. This is one of the reasons I consider it disproportional. Or if implemented on the messaging side only (which is easier to bypass than in the OS), it will make open messaging apps illegal.


> on the other side the whole tech world assumption that anything has to stay lawless is just absurd

This is not the whole tech world's position. Why make up an equally bad opposing position instead of just saying "this regulation is going too far"?


It's so disheartening to follow these. Time after another we hear about some insane Orwellian plot to exploit our deepest secrets. All spun so that the masses will think it's for some noble cause like protecting the children when really it's anything but. And it never stops! Tackle one and it's back a year later in some even more devious form like a fucking Hydra. I'm just so tired I wanna move into a cottage in the woods.


> shouldn’t we do everything in our power to prevent it?

I'm more concerned about the original abuse. The pictures are obviously an issue as they create a market _for_ abuse, but if you're not targeting the original crime, I don't think you stand a chance of actually improving the world by destroying rights.


the solution, regrettably, already in motion, is (obviously) to make sure less people know anything about computers.

by these two actions combined this anti-freedom garbage (further consolidating and centralizing powers) will work effectively


I very much agree, plus:

Are they thinking of the children when they raid dad's home because a picture of a kids genitals went to a physician for tele-medicine?

Are they thinking of the kids when they come for dad when dad really doesn't like his pictures scanned and self-hosts his infra and uses a Linux based phone?


I don’t know CCPA laws that well, but I do the EU ones. As it stands “the right to erasure / to be forgotten” is extremely vague on this, and there doesn’t seem to be a wide precedent. In general the law is applicable to raw data records and not to aggregate data/metrics, neither to models. However, models in this context refers to one particular ruling w.r.t. to insurance or credit scoring industry (don’t remember exactly which one).

I want to point out that the model doesn’t need to “spit out” removed data. It can be a classifier, or regression model, and not a generative model, and ideally, it would not be trained on your data.

Worth noting that from the technical standpoint, it’s difficult too. Say, a model costs X-large amount of dollars. Normally, I would retrain it e.g. every 6 months. But now I have erasure requests coming in on a regular basis —- retraining often to comply with those is too expensive. There’s a line of research on “machine unlearning” on how to do it efficiently but it’s quite underwhelming so far.


Twitter’s real value was its “global RSS” nature. You could get short, quick updates about local news, published papers, sports events, political happenings, and more.

It made it easy to source info from your followers (or their followers) quickly. Think Ask HN but broader.

Activity on Twitter is down now, mastodon and threads aren’t cutting it. There’s more activity on LinkedIn but it’s quite phoney.

I genuinely feel that Twitter provided me with useful information from people of interest.


Why aren't mastodon and threads cutting it? They provide exactly what you claim you want.

Leads me to believe for a lot of people, the allure of Twitter isn't actually anything intentional, but just the combination of being at the right place at the right time - that feeling you got, the relationships you fostered. Just like if you were on Tumblr during its independent peak.


Lots of people I know have half-migrated to Bluesky or totally moved to Mastodon, so my feed is much emptier than it used to be.

I'm pretty optimistic about the fediverse in the medium term, certainly for many communities. But right now twitter.com is still the default for organizations large and small, and I'm not sure that changes. It's gonna be a mess.


OP is right in that his change would make the softmax in the attention output zero if it "has nothing to add" (QuietAttention, as he said).

Buuut, it's missing the forest for the trees. The goal of the last step of attention (ref., Fig. 2, left in https://arxiv.org/abs/1706.03762) is not to add/say anything (as the author is saying) but to compute the relationship between the tokens (QK^T) and V -- in layman terms, simplifying, which tokens are related to each other. The softmax is there because it gives a representation that is nicer to work with, it gives probabilities, instead of unscaled matrix multiplication.

TLDR; author isn't wrong but he isn't right, practically speaking, either.


What's wrong with unscaled matrix multiplication? Softmax has some kind of intuition in the context, but why not layer norm or something else instead (if anything is needed at all)?


The family of sigmoid functions has nice gradient properties with theoretical backing. Good starting read: https://stats.stackexchange.com/questions/162988/why-sigmoid...


Doesn't really explain why not more typical ReLU + layer norm or some alternative.


Source: I work in the field.

This is a current limitation, and an artifact of the data+method but not something that should be relied upon.

If we do some adversary modelling, we can find two ways to work around this:

1) actively generate and search for such data; perhaps expensive for small actors but not well equipped malicious ones.

2) wait for deep learning to catch up, e.g. by extending NERFs (neural radiance fields) to faces; matter of time.

Now, if your company/government is on the bleeding edge of ML-based deception, they can have such policy, and they will update it 12-18-24 months (or whenever (1) or (2) materialises). However, I don't know one organisation that doesn't have some outdated security guideline that they cling to, e.g. old school password rules and rotations.

Will "turning sideways to spot a deepfake" be a valid test in 5 years? Prolly no, so don't base your secops around this.


> This is a current limitation

The thing with any AI/ML tech is that current limitations are always underplayed by proponents. Self-driving cars will come out next year, every year.

I'd say that until the tech actually exists, this is a great way to detect live deepfakes. Not using the technique just because maybe sometime in the future it won't work isn't very sound.

For an extreme opponent you may need additional steps. So this sideways trick probably isn't enough for CIA or whatnot, but that's about as fringe as you can get and very little generic advice applies anyway.


It sounded to me like the parent poster wasn't saying not to use it, but simply that it cannot be relied upon. In other words, a deepfake could fail a 'turn sideways' test and that would be useful, but you shouldn't rely on a 'passing' test.


Another way to think of it might be that it can be relied on - until it can't. Be ready and wary of that happening, but until then you have what's probably a good mitigation of the problem.


I think the concern is complacency, and the inertia that existing security practices leads to security gaps in the future. "However, I don't know one organisation that doesn't have some outdated security guideline that they cling to, e.g. old school password rules and rotations."

Or put another way, humans can't be ready and wary, constantly and indefinitely. At some point, fatigue sets in. People move in and out of the organization. Periodic reviews of security practices don't always catch everything. Why something was implemented was forgotten by institutional memory. And then there's the cost for retraining people.


The flip side of that is people feeling/assuming there's nothing they can really do with the resources they have therefore they choose to do nothing.

Also, those that are actively using mitigations that are going to be outdated at some point are probably far more likely to be aware of how close they are to being outdated by encountering more ambiguous cases, as seeing the state of the art progress right in front of them.

As for people sticking to outdated security practices? That's a problem of people and organizations being introspective and examining themselves, and is not linked to any one thing. We all have that problem to a lesser or greater degree in all aspects of what we do, so either you have systems in place to mitigate it or you don't.


Therefore, developing and customizing a proper framework for security and privacy starts by accurately assessing statutory, regulatory, and contractual obligations, and the organization's appetite for risks in balance with the organization's mission and vision, before developing the policies and and specific practices that organizational members should be doing.

To use a Go (the game, not the language) metaphor, skilled players always assess the whole board rather than automatically make a local move in response to a local threat. What's right for one organization is not going to be right for another. Asking the caller to turn sideways to protect against deepfakes should be considered within the organization's own framework, along with the various risks involved with deepfakes, and many other risks aside from deep fake video calls.


Asking the caller to turn sideways is also a cheap countermeasure without serious side-effects. So there's low risk to adopting it.


If that is conclusion that is considered within the organization’s custom security and privacy framework, sure.

If there is no such framework, this is no different than yoloing lines of code in a production app by a team that does not have at least some grasp of the architectural principles and constraints at play. Or worse, not understanding the “job to be done” and building the wrong product and solving for the wrong problem.


How do you find out that it doesn't work?


Exactly. Even the article gave a couple cases of convincing profile deepfakes. Admittedly they’re exceptional cases, but in general progress tends to be made.


The self driving car of next year arrives just in time for the Iranian atomic bomb :-D which is ready in two years for about as long as I'm around. https://www.theatlantic.com/international/archive/2015/04/ir...

If all the money on self driving cars would have been put into public transport (driverless on rails is a solved issue) and pushing shared car ownership instead, we might actually get somewhere towards congestion-free cities.


We can already have congestion free-cities today, no new technology nor public transport required. We had the technology for quite a while now: congestion charging.

It works really well in Singapore to control congestion, and also worked well in London when they adopted it afterwards.

Public transport also works quite well in many places around the world.

It also used to work really well in North America in the past. A past when the continent was much poorer. (I'm mostly talking about USA plus Canada here.)

Public transport only works when after you step off the bus or train, you can get to your destination on foot. Density is outlawed in much of the USA and Canada.

https://www.youtube.com/watch?v=MnyeRlMsTgI&t=416s starts a good section about Fake London, Ontario. At great expense, they built a new train line. But approximately no one uses it, because you can't get anywhere when leaving the stations. The video shows an example of a station where the closest other building is about 150m away. And that's just a single building. The next ones are even further.

Land use restrictions and minimum parking requirements are a major roadblock. And just throwing money at public transit directly won't solve those.

Shared car ownership is an interesting idea. Uber can be seen as one implementation of this concept. It can be done profitably, but I'm not sure it has much impact on the shape of cities?

In the grand scheme of things, there's not much money being put into self-driving cars so far. A quick Googling gives a Forbes article that suggests about 200 billion USD.


In terms of this particular tech previous obvious limitation, namely no blinking, worked for something like a quarter from discovery.

Venn diagram of people who someone wants to trick by this particular tech, those who read any security guidelines and those worthy of applying this kind of approach to in the first place is however pretty narrow for the foreseeable future. It's more of a narrative framing device to talk about 'what to do to uncover deepfake video call' as a way to present interesting current tech limitations - not that I particularly mind it.


Exactly! Our SecOps includes seeing people regularly. Until deep fakes can fake accents, tone, body language and jokes, we're safe. :)


This may be like a proof of work cryptography issue, except the burden of work is on the deep fake. Just ask a battery of questions, just like out of a Bladerunner scene or whatever. This is still the problem with AI. It depends on tons of datasets and connectivity. Human data and human code are kind of the same. Even individually, we can start with jackshit and still come up with an answer, whether right or wrong. Ah, Lisp.


> Self-driving cars will come out next year, every year.

"Come out" could mean different things in different contexts. Deepfake defence context is analogous to something like: there are cars on public roads with no driver at the wheel. And this is already true in multiple places in the world.


Waymo in Arizona is an example


I think it's odd we don't think of other limitations of products the same way. Put another way, why don't we just say it can't do it.

Example, we don't say a jet ski has a current speed limitation of 80 mph, we say it can go 80, but not 81. It's a simple fact. No promise that it will be faster tomorrow, because that's not what it is, it's not its future self.

It's like they're combining startup it will always be better after you invest more money with the reality of what "is" means.


One thing that I haven't seen mentioned is that many of the recent articles I've seen misuse the phrase "deep fake" and usually mean "face-swap algorithm" or "look-alike". The former, I believe has been able to defeat this test for 10 years at least and the latter has always been able defeat this trick.


> The thing with any AI/ML tech is that current limitations are always underplayed by proponents

if you don't worry about deepfakes, ok. But if you worry about deepfakes, you should not be reassured that this glitch is going to save you.

I'm not a proponent, just think your argument in this context doesn't work.


Self-driving cars are a million times harder than this, this is a terrible comparison.

Getting a model to work with images turned sideways is a few lines of code (just turn image sideways at training time).


>> images turned sideways

Instead of pictures of faces, now they're just vertical lines.


The technique can in principle be defeated today so it should not be employed as a single test, but rather another arrow in the quiver.


The only person who is promising self driving cars next year (and has done so every year for the past 5 years) is Elon Musk. Most respectable self-driving car companies are both further along than Tesla and more realistic about their timelines.


Let's take a look at some of those realistic timelines. A quick googling gave me a very helpful listicle by VentureBeat from 2017, titled Self-driving car timeline for 11 top automakers. [1]

Some examples:

Ford - Level 4 vehicle in 2021, no gas pedal, no steering wheel, and the passenger will never need to take control of the vehicle in a predefined area.

Honda - production vehicles with automated driving capabilities on highways sometime around 2020

Toyta - Self-driving on the highway by 2020

Renault-Nissan - 2020 for the autonomous car in urban conditions, probably 2025 for the driverless car

Volvo - It’s our ambition to have a car that can drive fully autonomously on the highway by 2021.

Hyundai - We are targeting for the highway in 2020 and urban driving in 2030.

Daimler - large-scale commercial production to take off between 2020 and 2025

BMW - highly and fully automated driving into series production by 2021

Tesla - End of 2017

It certainly wasn't just Tesla who was promising self-driving cars any second now. Tesla was definitely the most agressive, but failed to meet its goals just like every other manufacturer.

--

[1] https://venturebeat.com/2017/06/04/self-driving-car-timeline...


There was definitely a period when everyone (for certain values of same) felt they needed to get into a game of topper with increasingly outlandish claims. Because if they didn't people on, say, forums like this one (and more importantly the stock market) would see them as hopelessly behind.


Wow they all really got suckered by the AI grifters didn't they?


Self-driving cars are common in Europe for decades. We just use the less cool term "subway" for them.

Sorry, I couldn't resist. :)


Subways are common worldwide.

In fact, the first (practical) one was in Boston; not in Europe.

Sorry, I couldn't resist. ;)


London and Budapest had subways before Boston did. So did some other cities depending on which list you look at.

So what made Boston’s later entry the first “practical” one?

[Edit] Or do you mean self-driving subways? Does Boston have one already? A quick Googling suggests the opposite:

https://whdh.com/news/mbta-officials-considering-self-drivin...


Sure, but are they self driving?

A number of european capitals seem to have managed to do driverless high capacity underground trains. Here in the UK, we've got a number of automated trains but for union reasons they still have drivers in the cab who press go at each station.

In the US, it looks like Detroit has a self driving line, and there are a bunch of airport shuttles. Presumably you are hitting the same union issues as us?


Let's not dismiss the point that self-driving cars are the "stone soup" of machine learning industry. Like the monk who claimed he could make soup with just a stone, machine learning claimed that with two cameras, two microphones, and steering/brake/accelerator control, a machine would someday soon drive just like a human can with that hardware equivalent.

Then it turned out well, we actually need a lot more cameras. Now we need high res microphones. Now we need magnets embedded in the road. Now we need highly accurate GPS maps. Now we need high power LIDAR that damages other cameras on the road. Now we need....

Each little ingredient in the soup "made only with a stone." Machine learning has utterly failed to deliver on this original promise of learning to operate a vehicle like a person, with no more sensors than a person.


"Machine learning has utterly failed to deliver on this original promise of learning to operate a vehicle like a person, with no more sensors than a person."

I am not aware of anyone except Musk making that claim. "Machine learning" as in the statements of the main researchers, certainly did not promise anything like it.


The problem for self driving cars is the risk tolerance. No one cares if a deep fake tool fails once every 100,000 hours because it results in a sub standard video instead of someone dying.


What about reflections? When I worked on media forensics, the reflection discrepancy detector worked extremely well, but was very situational, as pictures were not guaranteed to have enough of a reflection to analyze.

Asking the subject to hold up a mirror and move it around pushes the matte and inpainting problems to a whole nother level (though it may require automated analysis to detect the discrepancies).

I think that too might be spoofable given enough time and data. Maybe we could have complex optical trains (reflection, distortion, chromatic aberration), possibly even one that modulates in real time...this kind of just devolves into a Byzantine generals problem. Data coming from an untrusted pipe just fundamentally isn't trustable.


I wonder how good the deepfake would be for things it didn't have training data on. For example, making an extreme grimace. Or have the caller insert a ping pong ball in his cheek to continue, or pull his face with his fingers.

One thing I notice with colorized movies is the color of the actor's teeth tends to flicker between grey and ivory. I wonder if there are similar artifacts with deep fakes.


Years and years of having to do increasingly more insane things to log into banking apps until we’re fully doing karaoke in our living rooms or stripping nude to reveal our brand tattoos


Plenty of new content for the banks' TikTok followers to enjoy :D


Please drink a verification can to continue, caller.



If I remember correctly, the context was that Microsoft had made the Kinect mandatory for the Xbox One which wouldn't function without it. And the Kinect was being used for some silly voice/motion control crap.

The extreme reaction and copypastas like this probably lead to microsoft scrapping that idea a few years later.


A can of Ubik please


"Please put one finger behind each ear and flap them at me."


I had to laugh with tears at this one. :)


Shoe on head?


This was my first thought. Ask the person to turn sideways and then put a shoe on their head. Or put a shoe on their head and then turn sideways.


Last time I applied for a credit card online, they asked me to take a video of myself and turn my head from side to side.


This sounds like a great way to get sufficient images/video of you to create a deepfake that could pass this test. Hmmm...


New mandatory security rule: Employees must never turn their heads side to side in a meeting.


Microsoft Teams developed a feature when if you’re using a background and turn sideways, your nose and the back of your head are automatically cut off.

Bug closed, no longer an issue, overcome by events.


Interesting that you bring that up. The most egregiously invasive student and employee monitoring software requires that the subject always face the camera. That seems most ripe for bypassing with the current state of deepfakes. https://www.wired.com/story/student-monitoring-software-priv...


I work as a Digital Gardener[1] and we’re trained to NEVER use our real name.

- [1] https://youtu.be/XQLdhVpLBVE


My bank does a much better system where they ask for a photo of you holding your ID and a bit of paper with a number the support person gave you for authorizing larger transactions. It's still not bullet proof but since you already have to be logged in to the app to do this, I'd say it is sufficient.


Interesting anecdata! Do you think the photo is authenticated/validated automatically (by software), or by a human, or combination (assistance)?

And, if you are willing to share, what country/bank?


This case I was on the bank text support requesting to make a transaction of $100,000 in one go which the app would not let me do. So it was a real person on the other side. Bank was in Australia called Up.


This sounds like a good thing. An extra step in a $100,000 transaction to prevent accidents or crimes definitely feels justified if the accounts not marked as normally moving heaps of money like a billionaire or something.


Yeah this is quite common with fintech (stock brokers and crypto IME) KYC nowadays I've noticed.


May I ask what card/Institution? This would be an immediate no for me.


I'd trust the data with a (real, not online) bank more than most other companies like Google.

I'd be more worried about people hacking into networked security camera DVRs at stores and cafes and extracting image data from there. Multiple angles. Movement. Some are very high resolution these days. Sometimes they're mounted right on the POS, in your face. Sometimes they're actually in the top bezel of the beverage coolers.

Banks are the hardest way to get this data, not the easiest one.


> Banks are the hardest way to get this data, not the easiest one.

Is this statement based on data or a hunch? A quick google turns up a lot of bank data breaches.


A quick google turns up a lot of bank data breaches.

Because banks have to report data breaches. Do you think every neighborhood Gas-N-Blow is publicizing, or even knows, that it's been hacked?


Good point. I’m still wary of just assuming (if that’s what we’re doing here?) that old established organizations you’d expect to be secure are in fact secure. For example I would have expected credit rating agencies to be secure…

Mandatory reporting certainly helps IMO. Reporting should be mandatory for anyone handling PII.


No bank is going to run such a system in house. It will be a contracted service whose data is one breach away from giving fraudsters a firehose of data to exploit their victims.


You would? You would trust a random number to call you and talk to you about your bank account?

(That's what Chase's fraud department tells you to do.. no joke)


"I trust you more than Google" is a pretty low bar in terms of personal data.


I want to know so that I can forward this to lawyers that specialize in biometric privacy law (in IL).

Fuck these biometeric data farmers.


Yes I believe this sideways turning thing is mandatory when doing online identifications


What is an "online identification"? In what context would such a thing occur?


And now that scan could eventually end up out there someplace.


Agreed. Now they have the data to deep fake you turning your head.

I hope they delete the data immediately after use.


Frankly, of all the personally identifying data I share with my bank, a low resolution phone video of the side of my head is the least worrying. It's like worrying the government knows my mum's maiden name!

In the eventuality that robust deepfake technology to provide fluid real-time animation of my head from limited data sources exists and someone actually wants to use it against me, they can probably find video content involving the side of my head from some freely available social network anyway.


I've been looking to rent housing and get a new job the last few months. The amount of info I've sent strangers always worries me.

At least with housing they don't ask me to input the information I've already sent them into their crappy website.


And, if deepfake technology becomes so easy to use, video of your face will no longer serve to identify you.


The implementation I’ve seen only stores a hash based on the image analysis


As far as I can see, secops is an eternal cat-and-mouse game.


Some see secops as futile until the tools are here. So we're making those tools instead.


Literally an arms race.


Face race?


[flagged]


Would you please stop posting unsubstantive and/or flamebait comments to HN? You've been doing it repeatedly, it's against the site guidelines, we end up banning such accounts, and we've had to warn you more than once before.

If you'd please review https://news.ycombinator.com/newsguidelines.html and stick to the rules when posting here, we'd appreciate it.

Note this one, just to pick one example:

"Eschew flamebait. Avoid unrelated controversies and generic tangents."

That's one of the most important rules for avoiding what turns internet threads dumb and nasty.


Not sure what to say to this one. Women can get sensitive, if requests of them can be seen in an unpleasant light. Women have also been historically tricked into posing for cameras, had their images misused, and are often quite sensitive about it.

I thought my comment was legit, and on topic. If one is going to implement a policy where people have to slowly move their camera around their body, there may be severe misunderstandings ... and an inability to clarify if a bad response runs-away on twitter and such.

Support persons should be carefully coached on how to handle this.

I guess all I can say here is, I didn't mean this to be so controversial.

Sorry.


The greater problem is that you've been breaking the site guidelines repeatedly.

https://news.ycombinator.com/item?id=32364710

https://news.ycombinator.com/item?id=32364583

https://news.ycombinator.com/item?id=32299889

https://news.ycombinator.com/item?id=32299875

https://news.ycombinator.com/item?id=32295137

https://news.ycombinator.com/item?id=32292576

https://news.ycombinator.com/item?id=32292440

https://news.ycombinator.com/item?id=32290426

https://news.ycombinator.com/item?id=32289581

We ban accounts that keep doing that, so would you please review https://news.ycombinator.com/newsguidelines.html and take the intended spirit of curious conversation more to heart? Tedious flamewar, especially the back-and-forth, tit-for-tat kind, is exactly what we don't want here.´


Listen, I'm willing to try to adhere more closely. I see how some of the above posts click with what you're complaining about.

One suggestion ; some news articles are just, almost, entrapment. I feel like HN having an article about racism, is going to entice all sorts of comments which break the site guidelines.

Take my posts above ; I legitimately am concerned that by simply labeling those we are strongly opposed to, eg racists, we fail. If we just label, if we therefore misunderstand their motivations, any attempts at correction become flawed, and just plain don't work.

I want positive corrective action to fix things, not divisive posturing.

Obviously in hindsight I should not have waded in, I should have realised how my post, my motivations could be misunderstood. I agree, my bad. 100%. But once there, I'm left in this horrid position of suddenly feeling as if I'm being labelled as a racist sympathizer or some such. A reputation is a hard thing to leave on the ground, with no response!

Perhaps I should find better ways to wade out.

Anyhow, are we OK for now?


job security


Indeed (:


It saddens me how many smart people are working in such an unethical field.


Interesting to bump in to somebody that works in this field.

What do you do in this field?

What's the direction of travel on it?

What makes it worth pursuing at a commercial level? In other words - how is this tech going to be abused/monetized?


1) actively generate and search for such data

What about doing a bunch of video calls, and asking for callers to show their profile, "to guard against deepfakes?"


The other thing is, why is this even important, when you shouldn't be basing decisions off the other person's race or face in general?

Base everything off the work they do, not how they look. Embracing deepfakes is accepting that you don't discriminate on appearances.

Hell, everyone should systematically deepfake themselves into white males for interviews so that there is assured to be zero racial/gender bias in the interview process.


But currently, it's pretty much a guarantee that you can pick out a deepfake with this method as there is no way for current methods to account for it that are in use.

As with any interaction with more than one adversary, there is an infinite escalation and evolution with time. And similarly then something will come up then that is unaccounted for and so on, and so on.


Asking for entropy that’s easy for a real human to comply with and difficult for a prebuilt AI is at least a short term measure. Such-as show me the back of your head sideways then go from head to feet without cutting the feed.

Easy for a human, difficult for ML/AI


>Will "turning sideways to spot a deepfake" be a valid test in 5 years? Prolly no, so don't base your secops around this.

We'll just ask them to do "the Linda Blair". If they can turn their head 360 degrees, prolly a deepfake ;P


> so don't base your secops around this.

If it's a high-threat context I don't think live video should be relied on regardless of deep fakes. Bribing or coercing the person is always an alternative when the stakes are high.


What if the real person draws something on his face? Does the deepfake algorithm removes it from the resulting image? Can you ask the caller to draw a line on his face with a pen as a test?


> Can you ask the caller to draw a line on his face with a pen as a test?

I think if the caller did this without objection that would be a bigger indication that it is a deep fake than the alternative. What real person is going to comply with this?


I was thinking of those two cases. Stuff like this is always a cat and mouse game.


> Will "turning sideways to spot a deepfake" be a valid test in 5 years? Prolly no, so don't base your secops around this.

couldn't the same thing be said about passwords, 2FA with SMS or asymmetric cryptography?

meanwhile real IDs have been easy to replicate for decades, but are still good enough for the job.


OK, you passed the yokogao test. Now take a crayon and draw an X on your cheek.


“The Impossible Mission Force has the technical capabilities to copy anyone’s face and imitate their voice, so don’t base your secops around someone’s appearance.”

… yes, because that worked well.


> … yes, because that worked well.

Just to be clear, Mission Impossible is not a documentary.


It is however, a lower bound on whether it is the case that something is a reasonably forseeable/precedented area of research.

After all, if the artist can imagine and build a story around it, there'll be an engineer somewhere who'll go "Ah, what the hell, I could do that."

*By Golblum/Finnagle's Law, it is guaranteed said Engineer will not contemplate whether they should before implementing it, and distributing it to the world.

This another example of why we can't have nice things.


Show HN is about showing your own work, not just sth you think is interesting. Doesn't matter if you missed it, still shouldn't have posted it.


I'd say "Shouldn't have put Show HN in the title", otherwise it's a good submission.


Disclaimer I am the author (not OP). I guess the user re-posted by simply copying my original one...


My main gripe was that it was postet by OP as a "Show HN" ;)

Nice work btw always astounded how you keep developing it for so long nearly solo. I was many times intrigued enough to try to switch to OneDev just to try it, but my personal git setup is working too well right now. Never switch a running system runs too deep in me.

Anyway there seem to be discussions taking place so I will cut the noise and stop whining.


Nothing that you said is wrong but it doesn't make the situation better.

1) As many people pointed out, this doesn't prevent OCR, it just prevents copying strings (e.g. with crawlers). 2) Majority of OCR doesn't deal with PDFs produced from a text source but either from a) jpg-scans of documents b) pdfs produced from those jpg-scans. 3) The first thing I tried, was OCR with my iPhone and it obviously worked. As someone else said, there're solutions that let you batch process many documents.

Don't get me wrong, your stuff works for what you designed it to. However, it provides <false sense of security> by <falsely> claiming that it prevents OCR; which in turn, can lead to more harm[1].

[1] - e.g., it may convince people to share stuff that they wouldn't otherwise.


It's interesting what/when/if will unsettle Python. And what the adoption would be like.

Julia has been <designed> to unsettle Python in the data space but to no avail.

¯\_(ツ)_/¯


Julia was designed to unsettle a creeping set of python in the domain of Perl, C++, and Matlab in scientific computing and various quant analysis that’s pretty distant, IMO, from what industry data engineering, ML, and analytics work was doing.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: