Hacker News Reader

Comment by beepbooptheory | Hacker News Reader

To at least attempt to think about this critically, and to try to ignore the amount of frankly shocking things people are saying here, can I ask, maybe from people in the field, is this level of industrial/scientific espionage even a critical factor in our current world? We live in a global world where scientists in various institutions are constantly trading papers, findings. Why are we so worried that the U.S. might be doing something so different from China, that this level espionage would even be viable? Surely, everyone is working on different things here and there, but this idea that there is such critical intellectual property the US that needs to be guarded so intensely in this like James Bond scenario, just feels a little wacky.

Are you all really here in 2022, being like "we can't risk a Chinese national from even seeing our American semiconductor research, they will steal it and use it against us?" Are actual scientists concerned? Or just the state, and apparently most of the fellow commenters?

View on HN

> Are you all really here in 2022, being like "we can't risk a Chinese national from even seeing our American semiconductor research, they will steal it and use it against us?"

There are a number of issues at play here and I can speak to several of them, having either helped the federal government directly or having knowledge of industrial espionage in my industry at the time (semiconductors).

Some technology is export-controlled in the United States, full stop. Semiconductor research may absolutely be export-controlled and thus may not leave the United States or be shown to foreign nationals. Any researcher's talk, paper, research, materials, etc. must undergo an export controls review first before it leaves the United States or is shown to a foreign national. This is SOP at every research lab in the United States.

If you do not disclose information when asked by the federal government as part of an investigation impacting national security, then woe be unto you. I don't know what else to say there. It appears that this may be what caused this gentleman the issues outlined by the NY Times.

Industrial espionage is absolutely a thing in the semiconductor industry. I was once told how Hynix pilfered fab designs from Samsung. The loss was estimated to be $1 billion. I am personally aware of other lesser IP loss through the hands of careless engineers.

I hope this helps answer your question. It is absolutely a thing in 2022.

> The estimated loss was estimated to be $1 billion.

I always give side eye to the accounting that considers lost sales opportunities to be losses.

Not saying Hynix did nothing wrong, only that that number assumes a lot of factors that are in no way guaranteed (stability of the market, no third player independently developing innovations that would create an attractive alternative for customers, the ability of the stealing company to implement the design they've acquired, etc.).

> I always give side eye to the accounting that considers lost sales opportunities to be losses.

That's a perfectly valid but also somewhat trivial point. Some number has to be assigned as human nature requires it.

Maybe the best representation of the situation is to present it as a question. When fabs cost $xx billion and take almost a decade to reach fruition from R&D to the point where they start shipping a viable product, if you were able to shave two to four years off that process, what does that number look like to you?

It looks like saving a billion dollars in redundant work, which is a net societal gain. Now consumers have two companies to choose chips from instead of one functional monopoly due to a trade secret moat... Win for consumers. ;)

... Hence my unease with the way the question is phrased. I'm not saying we should either mandate corporate espionage or enforce some kind of punishments for not publicizing designs, but my point is it doesn't take a lot of reframing of the question to completely change the numbers, and that makes the numbers suspect for being anything real.

It's akin to the sleight of mind that allowed us to put the burden of fraud on consumers with the concept of "identity theft," instead of asking why it should be the responsibility of the person who's information was used that some third party used that information to trick a company into giving away resources. What if the mindset were instead "it's not Alice in Topeka's responsibility that Visa allowed Bob Hacker to scam Walmart out of a bunch of free stuff by pretending to be Alice, and Alice should be required to do exactly nothing to fix Walmart and Visa's problem?" What if instead of having to contract credit monitoring agencies to look out for identity theft, the government held credit reporting agencies responsible with statutory damages for defamation when they incorrectly declare a person has done something they have not?

A slight reframing of the problem completely changes the proposed solutions and the value calculus of what was gained and what was lost.

> Now consumers have two companies to choose chips from instead of one functional monopoly due to a trade secret moat... Win for consumers.

If we were considering this purely from the lens of consumer choice, allowing IP theft seems like a no-brainer. But it's pretty clear from the last few years how important supply chains are to the countries they reside in (and how their absence is strongly felt). To phrase it another way, suppose IP theft makes an industry outright unprofitable in a country and those jobs disappear in that country. It can take decades to spin that stuff up again and the consequences can be disastrous. From that perspective, it's understandable why governments would want to tamp down on IP theft. That said, the story described in the OP is tragic and reflects a perverse situation where someone who came to America seeking those things we promise to the world was considered guilty for chasing the American dream.

It's not a gain even purely from the lens of consumer choice unless there is only one iteration of chip improvements ever. Why? Because if this is permitted, there will be no company A to do the research to create the next generation of chips and innovation basically halts.

How are those even remotely similar?

On one hand you have a company that has invested billions of dollars to develop a technology on the assumption that it will give them an advantage that will allow them to recoup their losses. Then someone steals it and they can't recoup because it's no longer an advantage. Yes, in this one instance it's a benefit to consumers, but it's clearly not a benefit to the people who worked hard to develop the technology or the investors that took on the risk of trying.

It feels like your next argument will be that patents shouldn't be enforceable because it's better for society that everyone have access to everything.

> It looks like saving a billion dollars in redundant work, which is a net societal gain. Now consumers have two companies to choose chips from instead of one functional monopoly due to a trade secret moat... Win for consumers. ;)

But looking at it from the company point of view.

- Company A spends 5 billion $ researching a new fab, from which better chips can be made.

- Company A expects to sell those chips as <cost> + <margin> + <an amount extra to pay off the 5 billion that was spent>

- Company B steals the plans for the fab

- Company B sells the chips for <cost> + <margin>

At this point, it is impossible for Company A to recoup the money they spent researching the technology, because people will buy from Company B if it costs less from them. That's a pretty huge loss.

For Company A it's a loss, for Company B it's a gain, and for the world it's a gain, because now there are more of those better chips than there would have been. The problem is that if Company A expects this to happen, they may not spend the US$5 billion in the first place, which is an even bigger loss for the world.

Nortel comes to mind. Think about all the highly paid engineers that lost their jobs. An entire domestic sector wiped out.

Nortel was sunk by mismanagement before the espionage episode, not by the espionage itself. Quite possibly a lot of Nortel's technology would have been lost in the confusion of bankruptcy if the spies hadn't saved it.

Do you have any info or reading on this? I tried to Google "nortel mismanagement", but it's unclear any of the sources put it in light of the espionage.

There are some comments at https://news.ycombinator.com/item?id=30114607, but I admit I haven't verified them. I'm interested to hear what you end up concluding if you do dig into the sources.

Sales isn't the only part of it. Hynix could have reverse engineered the chip and made a functional copy that wouldn't violate any copyright or patent laws (this last is a maybe, depending on if the interface is patented or just the implementation), but that would have cost millions in engineering hours. They also could have hired engineers to build the fabs, again at the cost of a lot of engineering hours.

The time to bring technology to market is also worth lots of money.

Do you not think Intel is at a loss financially due to how they fell behind TSMC in fabrication technology?

If Intel instead had had the better process, how much more revenue would they have, what market cap would they have?

It has material impact. If someone unscrupulous could steal TSMCs tech, do you not think TSMC would suffer? Or how much better off the thief would be?

Preventing incremental material losses suffered by a U.S. mega-corporation isn't very high on my priority list and shouldn't be very high on yours, either, if you are a citizen of this planet. Many parts of the U.S. have third world conditions and the UN recently sent inspectors to parts of West Virginia because things have become so bad there. Millions are without healthcare, children are hungry, etc. Why waste even a breath on protecting proprietary semiconductor technology? The major semiconductor companies have significantly more motivation to protect their proprietary technology than the modicum of care we derive from our saber rattling and trying to maintain geopolitical dominance, and at the end of the day it is up to them to protect their own tech (their real enemy is AMD and ARM anyway). Imbuing certain technologies with vague and in practice largely imaginary export control laws does very little to prevent this kind of behavior, and it is in my mind a misappropriation of resources and mental energy to go out of our way to try to enforce them on a national scale. Shouldn't we, as a society, instead reserve this level of moral and legal outrage for CEOs who price-fix, bribe, and make 100x more than their average employee while people are starving in rural backwaters and dying in the streets and in hospital hallways thanks to COVID?

Oh I agree companies should have greater interest in the well being of their communities and we should effect change that makes them beholden to the communities that allowed them to prosper and not sell the communities out to the lowest bidder overseas.

But… tell China this same argument. That they should not protect their self interest and see how well it goes over.

> But… tell China this same argument. That they should not protect their self interest and see how well it goes over.

Note that my argument is one of prioritization -- to protect our self-interest, we have a lot more pressing problems at home than Intel et al losing a few bucks because they let the wrong people see the secret formula, and wasting precious human and financial resources on the wrong problems can be worse than doing nothing.

As you mention, China on the other hand can afford to worry about problems like this, given their 0.7% poverty rate as of 2015 (despite some glaring issues like ethnic genocide, free speech, etc). In the U.S. we have an 11% poverty rate and nearly a million dead from COVID. We don't have time or resources to play the secret formula game with walkie-talkies in tree forts. Gotta do our homework pretty soon or it will be bed with no supper.

The irony of all of this, of course, is that we already have a system in place for protecting secret formulae: you patent them.

But here we get on the issue of the question of international espionage and whether it matters if you patent something in the US if China will just refuse to enforce an international patent.

Perhaps there's a case to be made that in a world where superpowers don't obey each others' laws, this was never going to be Samsung's $1 billion to make...

It seems like when you sum it all up, this is a really good time to be a Chinese tech firm.

It just goes to show that patents are stupid at the end of the day. From a harm-reduction standpoint, they do way, way more harm in the form of astronomically high drug prices, core pieces of technology being off limits even though almost anyone could have invented them, etc., etc., than they help. Patents, like intellectual property, are an imaginary legal construct, and if you want to compete in a global economy, you have to assume others aren't going to buy into your imaginary concept, particularly when it comes to opposing global powers.

A lot of knock-off technology is actually _created_ by people reverse-engineering things _from_ the patent as well, which they then mass-produce in a jurisdiction that doesn't respect U.S. intellectual property (AKA China where most things are mass-produced).

The real system of protection in place for protecting secret formulae is simple and more effective -- don't show it to the wrong people, and don't outsource your factories.

I am a professor in the field of cryptography and there are export restrictions here as well. Under no circumstances would I allow the US government to block publication of my papers or talks based on the theory that my speech is “a technology export.” It’s not even clear that this would be allowed under the First Amendment. The government could obviously try to advance that theory, but I strongly suspect they would run into a hailstorm of First Amendment litigation and public blowback.

To me it feels like these prosecutions are an attempt to do exactly that — place restrictions on academic speech and collaboration — but to do it through the “backdoor” of funding agencies and fraud accusations. In other words, if the US said “we are prosecuting you for sharing your open research publications with Chinese scientists,” this would be politically and maybe constitutionally infeasible. So instead they’re trying to achieve effectively the same chilling effect by accusing these scientists of criminal fraud in their funding applications. It is also noteworthy that these cases keep falling apart, which makes me think that they’re exactly as weak as I'm making them out to be.

At the end of the day we need to be very careful about this. We are a country that values free speech, has a constitution that guarantees it, and also needs open speech to make science work. “The government passed some laws restricting technology export” does not actually mean the government can and should be able to restrict scientific speech, and it definitely shouldn’t let them do it through weird indirect means like this.

ETA: It is important to draw a distinction between proprietary or classified IP (and physical equipment) where espionage or theft of trade secret accusations might make sense vs. open scientific research where it sure doesn’t. These cases keep trying to blur the distinction. I suspect if they were clear examples of the former then the cases wouldn’t keep collapsing.

ETA2: I have NSF funding and I just want to add that NSF regulation is byzantine. There are too many regulations surrounding NSF applications and reporting for a human (with an actual job) to be 100% certain that they haven't misstated something or omitted a critical detail. I would guess every US scientist is vulnerable to this sort of (selective, politically-motivated) prosecution.

As you are of course aware, but perhaps some people reading this are not, the government did consistently advance the theory that publication of speech about cryptography was "a technology export" subject to export control --- specifically, cryptographic source code --- until Bernstein v. United States, which required a several-years-long court battle.

You were a country that values free speech. Now you are a country that values "content moderation" to suppress "misinformation".

This is exactly my point: the government advanced exactly this theory and the crucial point is they lost in court. They were forced to weaken the laws as a result. The new laws remain constitutionally untested and the government seems very eager to keep them that way.

Agreed. But public opinion has shifted; people in the US think "free speech" is a "right-wing issue" now, and as you can see every time China is mentioned on HN there are ten thousand xenophobic comments by assholes from the US who don't even know the difference between the Cultural Revolution and the Great Leap Forward. In theory public opinion shouldn't affect the outcome of a court case unless it results in statutory or constitutional changes, but of course it does. Given the current trend toward authoritarianism, the longer the government can delay the court battle, the better their chances.

With due respect, the right wing is perfectly happy to use the power of government to suppress speech they don’t like. See eg the new bills banning discussion of sexual identity in Florida schools. I usually regret responding to OT partisan comments, but statements of belief like “<my preferred political party> is the party of free speech” is exactly how you end up voting in a government that brutally suppresses free speech.

I didn't mean the right wing was actually in favor of free speech, and I should have been clearer about that. The right wing has never been in favor of free speech, nearly by definition. The consequence is that the US political scene is dominated by two illiberal parties, much like, for example, Egypt after Mubarak's flight, or the late Weimar Republic.

I'm surprised that you'd think right-wing US parties were my preferred political party; maybe you've forgotten who I am because I stopped using Twitter. Still, I'd think even my comments in this thread would provide abundant evidence of my commitments to liberalism and distaste for authoritarianism.

I don't think it is strictly a partisan issue but the potential precedence where a re-normalization event may occur. What struck me curious was a random quote quote from somewhere that reads: "conservatism is progressivism with speed limits". I mean, where have we seen instances where such precedences have been set in the grassroot moments all the way to our highest courts?

Crypto export control.

DeCSS.

SDMI.

Effects from twitter bots from the early years and now?

*jokes are hard

This comment does not contribute.

Concur.

The US can be extremely creative when it comes to mislabeling things. Wire Fraud, Munitions and Money can suddenly equate to selling something, cryptographic code and people. This makes no sense to uninitiated outsiders.

I don't understand how you think your point relates to it's parent. Are you just piggybacking off the point that the US does control export of semiconductor technology to make this separate point? Or are you suggesting the US shouldn't control export of semi-conductor technology in the same way they shouldn't restrict your publications on cryptography?

The point is that they aren't "controlling an export", they are trying to restrict free speech.

The parent comment says "any researcher's talk, paper, research, materials, etc. must undergo an export controls review first before it leaves the United States or is shown to a foreign national. This is SOP at every research lab in the United States." My purpose in the comment is to point out that such review is not legally required for public academic research, and there are good constitutional and policy reasons it should not be. Moreover we should be careful about letting the government implement such policies via indirect means.

While I'm not in the field of semiconductor research, the export control laws in my field are quite similar (in fact, they're literally the same Federal laws.) And if you read the US code, you'll see that it excludes "[p]ublic domain information which is published and is accessible or available to the public, unrestricted: publications & conferences, patents, fundamental research." These exceptions exist because earlier laws did not have such exceptions and were found (in a case involving cryptography) to be unconstitutional.

Thank you for the context

"[p]ublic domain information which is published and is accessible or available to the public, unrestricted: publications & conferences, patents, fundamental research."

That's the definition for what falls under public domain and isn't export controlled, but say you are the one first publishing and bringing it in to the public domain. Would that first publication of something that isn't currently in the public domain fall under the exception?

Yes, that's one exception that seems highly relevant to the current case. Software is another. I'd like to believe that "published research is protected by 1A, but the speech necessary to make new published research is not protected" would go over like a lead balloon in the courts. Notice that these laws haven't been tested: instead prosecutors are going after "grant fraud" (and then dropping all charges before a judge can rule.)

If you didn't take money from the federal government, I would be inclined to agree with you, that export controls are an abridgement of the freedom of speech, but if your work is taxpayer-funded, then the government absolutely has the authority to restrict how you use the fruits of those funds; if you cannot abide by them then perhaps consider stop taking federal funds?

I think you are 100% backwards in your thinking here. If the US taxpayer funds research then public funders actually have a very high bar in their ability to restrict the publication of that research. For the simple reason that the taxpayer funds it and government restrictions on speech face a very high Constitutional bar. By contrast if I take funding from Amazon or Apple it’s much easier for them to demand NDAs and speech restrictions. (Which I have been asked for and wouldn’t agree to.)

you are right in practice, but my argument is not about what is de facto the case, but what should be the case.

> government restrictions on speech

Well this obviously should not generally be the case. If you are, say a high ranking general at the pentagon -- which is a position directly paid by the government, so arguably way more "protected" by the constitution than a subcontractor awardee of a grant, which could be considered to "create a legal veil", then can you claim "first amendment" if you pass military secrets on to an adversarial government? Of course not.

There is a huge amount of constitutional jurisprudence around the notion of national security clearance and all of it rests on the idea that it is a strictly limited exception to the First Amendment, generally one that is used sparingly and only when needed. This is why private citizens must volunteer to obtain security clearances. The General you speak of is presumably a volunteer who also agreed to be bound by stricter rules around classified data.

The NSF does not (generally) fund classified research. They fund unclassified research at open academic Universities. If the researcher in TFA had joined a military lab and submitted to a clearance, this would be an open and shut case. He did not do anything of the sort, and so the prosecutors were forced to hunt down some workaround to achieve their desired result, then ultimately drop a groundless case.

>has the authority to restrict how you use the fruits of those funds; if you cannot abide by them then perhaps consider stop taking federal funds?

Not sure if you're familiar with this bit[0]:

   Congress shall make no law respecting an 
   establishment of religion, or prohibiting the
   free exercise thereof; *or abridging the 
   freedom of speech*,

Are you arguing that taking public money obviates the above?

I'd say that since the government is involved, the above (are you not familiar with that bit?) is absolutely relevant and needs to be respected -- given that it's the law of the land.

[0] https://www.law.cornell.edu/constitution/first_amendment

going to quote my sibling child comment here:

If you are, say a high ranking general at the pentagon -- which is a position directly paid by the government, then can you claim "first amendment" if you pass military secrets on to an adversarial government? Of course not.

I suggest you think about why a general gets treated differently than a private citizen. Suppose you are a private citizen and manage to snap a picture of a secret research plane coming out of area 51, while not in the bounds of the facility. You will very likely enjoy first amendment privileges.

>I suggest you think about why a general gets treated differently than a private citizen.

I didn't realize that being an academic at a private research institution was equivalent to being a member of the armed forces.

Thanks for clarifying that for me. I should have realized that they were exactly the same thing.

Please.

Listen, if you work at a private research institution and you raise funds through a nonprofit (like, say HHMI), then by all means, do what you want. You are not beholden to the taxpayer. Money from the government absolutely comes with strings attached. Do you think that an employee of Lockheed Martin would be able leak secret plane information on a government contract to an adversarial country with impunity, just because it is a "private company owned by private citizens", thus gaining "first amendment protections"?

There's a difference between classified information and scientific research. Yes, some scientific research could be classified, but unless the government can show "national security implications" (a pretty low bar, IMHO), they should not be restricting the flow of speech or ideas.

A general officer in the military is not only covered by the Constitution, but by the UCMJ[0].

As such, your comparison is flawed.

We're (or at least TFA and me) aren't talking about folks subject to NDAs or restrictions based on "national security."

Private companies and private citizens can (at least in the US), limit and censor speech and information as much as they want -- and for the most part be supported by the courts (vis a vis NDAs and the like).

The government on the other hand is explicitly barred from banning most speech (there are exceptions of course, including libel/slander and/or disclosing classified information to unauthorized folks -- note that publishing such information has been held to be "free speech" -- it's the folks who disclosed such information that could be civilly (in the case of an NDA) or criminally (in the case of classified information) liable.

However, the government is barred from restricting speech except in those narrow circumstances.

Please explain how such narrow circumstances apply in this case? Or, absent those circumstances, under what legal framework would government restrictions on speech apply even to government-funded research? Please do explain. I'd be fascinated to hear it.

I'll repeat, and I'll use the words of those who wrote the law of the land. Hopefully, you'll understand the second time around:

   Congress shall make no law respecting an 
   establishment of religion, or prohibiting the 
   free exercise thereof; or abridging the freedom 
   of speech, or of the press; or the right of the 
   people peaceably to assemble, and to petition 
   the government for a redress of grievances.

About which part of the above do you disagree?

If you feel the above is in error, I suggest taking the steps defined here[1]. Good luck with that.

[0] https://www.military.com/join-armed-forces/the-uniform-code-...

[1] https://www.archives.gov/federal-register/constitution

You act as if I haven't read the constitution. I've pored over the constitution so many times, I'm well versed in obscurities down to the twenty dollar clause and the post roads clause.

Can you explain to me how classifying content is not an abridgement of free speech in an absolute reading of the constitution?

"Congress shall make no law.."

It does not say "Congress shall make no law, unless it's classified"

The UCMJ is (of course) congressional law, so by your analysis, the UCMJ would be in abeyance of the first amendment.

You didn't answer my question. Rather, you moved the goalposts.

Do that and I'll reciprocate.

Have a great day!

contract law. done.

>contract law. done.

That doesn't answer my question. Which was:

>Please explain how such narrow circumstances apply in this case? Or, absent those circumstances, under what legal framework would government restrictions on speech apply even to government-funded research? Please do explain. I'd be fascinated to hear it.

Are you advocating that all government-funded research should (or already does) have NDAs associated with them?

That's news to me. Please provide some documentation for your claim.

Yes, Congress has implemented laws that are within the narrow exceptions to the the bar on restriction of speech. How does that apply in this specific case?

What you are claiming/proposing is not the law of the land. What changes to the law/constitution are you proposing?

I await your response with (hopefully) some actual exposition. Based on your previous writings in this thread, I won't hold my breath.

>You act as if I haven't read the constitution.

That's a result of your statements.

>I suggest you think about why a general gets treated differently than a private citizen.

It's called the Uniform Code of Military Justice[0], and only applies to military personnel. That's why a general gets treated differently to a private citizen.

[0] https://www.military.com/join-armed-forces/the-uniform-code-...

ETA2: every scientist should include a statement that they believe they have complied with necessary regulations but that to evaluate and comply with them completely and consistently is an onerous burden and that they are operating in good faith, and welcome an audit.

Audits (if they aren't biased or motivated) can determine whether the failure to comply was ignorance, or intentional, and if ignorant, was it criminally so?

I mean, I've filed for NIH grants and there were parts that were self-contradictory and parts that my institution wouldn't let us do, so we always had to send it off to Contracts and Grants. Several proposal that I considered completely legit failed C&G review.

We (as in a company i have worked for) have caught employed chinese nationals try to install malware / provided internal account credentials to chinese ip addresses in simple retail big box industries. (I saw this happen personally and was shown what he was trying to install, and how a service account assigned to our group was trying to slurp up and probe the data from an IP address backtraced from China)

If that can happen in something relatively trivial you dont think a high stakes industry has it worse?

> If that can happen in something relatively trivial you dont think a high stakes industry has it worse?

I don't disclose my work for the federal government anymore on public profiles because I became tired of the constant hacking attempts directed at me. Once I scrubbed my profiles they stopped.

> Any researcher's talk, paper, research, materials, etc. must undergo an export controls review first before it leaves the United States or is shown to a foreign national.

Are you saying that if a professor in the US doing semiconductor research gave a talk in their university, that talk would have to undergo an "export controls review" before being given (since students could attend the talk, and many students are foreign nationals)? I'm having a very hard time picturing that. Do you have first hand experience, or is this just how you imagine this works?

I've personally witnessed Chinese national students attempting to acquire access to sensitive lab equipment and documentation their research wouldn't require (this is physics, here, not CS).

I've been on the defensive end of attacks by employees (all Chinese citizens, living in China) in our Chinese organization attempting to breach data and infrastructure firewalls.

Once upon a time my work touched some sensitive, classified areas. I was more or less constantly harassed by rather hilarious and obvious attempts by Chinese individuals to gain access to the classified bits for a time.

We don't live in a world where scientific and technological advances are all to the good for humanity. There are politically powerful people who would see these things used to ends that hurt others for their own gain. It happens all the time, and likely will as long as humans are what we are.

> There are politically powerful people who would see these things used to ends that hurt others for their own gain. It happens all the time, and likely will as long as humans are what we are.

Don't you think this group of politically powerful people might include the ones you research for?

Of course it would: but I'm struggling to understand the relevance, here. We do this stuff. They do this stuff. We each want to stop the other from doing it. There's nothing contradictory about it, as I'm not making an absolutist statements regarding ethics or morality.

"I've personally witnessed Chinese national students attempting to acquire access to sensitive lab equipment and documentation their research wouldn't require (this is physics, here, not CS)."

Well, I've personally witnessed american students attempting to acquire access to sensitive lab equipment and documentation their research wouldn't require, too. That included me! Thank goodness the attitude of my leaders was "smart people should have access to more tools than their job requires". that included chinese nationals who were in my program.

Certainly, my goal was the capture the hearts and minds of chinese people who were in my PhD program, so that they would move here, become US citizens, and contibute to our economy. It never even occurred to me that people could have a "dual scientific citizenship" and sort of hold positions in china and the us at the same time, and exploit that.

My goodness you read far too much into my comment: acting as though I thought every piece of sensitive information should be closely guarded to the death is...less than charitable.

is this level of industrial/scientific espionage even a critical factor in our current world

Yes. It completely bankrupted and destroyed the Canadian company Nortel [1]. Sure there were other problems that led to the downfall of Nortel, but this was a major contributing factor and has had a big impact on the Canadian tech ecosystem.

https://globalnews.ca/news/7275588/inside-the-chinese-milita...

So in the years before this hack, Nortel had already laid off more than 2/3rds of it's workforce, received a massive government bailout, and had top execs like their CEO fired for an "accounting mistake" that cost billions. "Other problems" is an understatement.

That story revolves around Brian Shields's account, which, while I don't doubt the details, I think lacks the business perspective and doesn't really explain Nortel's demise. Nortel was maximising growth during the the dot-com boom, spending heavily on acquisitions and not paying attention to its profitability. It couldn't have been worse placed to handle the dot-com bust when its main market shrunk hugely. Huawei and ZTE were not really major competitors internationally at the time.

It's dismaying to see history rewritten so brazenly.

Nortel collapsed because it made bad investments during the tech bubble, and then was hit by the nosedive in investment in network expansion when the bubble burst. On top of that, Nortel was terribly led, resorted to fudging its finances, and got caught.

That's what caused Nortel to collapse, not supposed Chinese spying. Huawei, the alleged beneficiary of the spying (which is just an unproven allegation at this point), was not even a major rival to Nortel when Nortel began to collapse. Huawei had barely even begun to expand out of China at that point. Nortel had much larger international competitors.

Yet somehow, almost two decades after Nortel began to implode, a rewritten history of Nortel's collapse, somehow premised on Chinese spying, has begun to become popular. I think this has much more to do with the recent increase in tensions between China and Canada than with any new information about how Nortel went under.

This shouldn't be a surprise. The Avro airplane was similarly infiltrated by soviet spies back in the 50's, thanks to lax immigration standards and policies.

The thing is that, while Nortel was pretty much an open bar for technology transfers (Canada didn't seem to care), what killed the company was accounting fraud.

What’s the primary export of the United States?

> Are you all really here in 2022, being like "we can't risk a Chinese national from even seeing our American semiconductor research, they will steal it and use it against us?" Are actual scientists concerned? Or just the state, and apparently most of the fellow commenters?

It’s intellectual property. The majority of the worlds cutting edge research are designed, developed, tested, and eventually marketed in the United States. That’s the advantage of free markets and gives the US first mover advantage.

Yes, it’s exceedingly important that is maintained (to the United States). If that stops occurring... manufacturing is gone, mining and energy sectors are regulated to oblivion. There’s not much left. Food and services? Healthcare? None of those produce enough wealth to maintain the nation, only information and advanced / advancing technology can support US growth.

>That’s the advantage of free markets

The large amounts of research done for and funded by the US government helps a lot too as do the various methods to give R&D tax advantages.

The majority, really? TSMC's fabs are in Taiwan, using ASML machines from the Netherlands. Much of their chips run ARM, from Cambridge. As I look at my tabs, I see Google Maps (invented by an Australian in Sydney), and Spotify (from Sweden).

The world is far, far, bigger than the USA; and as an Australian, I'd say it's extremely pretentious or naive to think the US has a 'majority' of cutting edge tech.

Google Maps and Spotify are not cutting edge tech.

ASML tech leveraged investments made over the decades by US national labs and consortiums like EUV-LLC as well as by companies like Cymer (DUV) which they acquired. IIRC, Part of the reason the US can forbid ASML selling EUV machines to China is they are covered by some patents held by US DOE/NIST.

For sure, ASML did the “final mile” in perfecting EUV and making it work in a production environment vs the prototypes in the labs 20 years ago, but I don’t think the funding that the US government and consortium members dumped on decades of research should be swept under the rug.

The parent does have a point, a common argument for pharmaceuticals is that governments and universities fund, or participate in funding, everything except the ‘last mile’ of drug development, i.e. human clinical trials. So a naive reader may assume they were primarily responsible but in fact it is the ‘last mile’ that consumes the vast majority of resources. Although the breakdown may not be clear in the case of EUV, it is possible that the vast majority of effort was getting across that ‘last mile’ of development, and thus a primarily non-US effort.

Okay, but that's just ignoring that ASML's largest design center is in NY just a few tens of miles from NYC and White Plains. Why is it there? Global Foundries (former IBM and AMD) is nearby and that was where the research was happening. Their second largest design center is in Norway, so not even in the EU. But sure, they're a Dutch company that does most of its design work outside of the Netherlands.

In terms of research, well the USA is still leading in total number of dollars, researchers, labs, and light sources dedicated to semiconductor research. Heck, we have three of the world's most advanced accelerators (Fermi Lab, Argonne National Labs, and SLAC) that are critical to semiconductor device research and were where most of the EUV breakthroughs were made.

In terms of total effort for EUV, yeah most probably wasn't in the USA. But if I had to hazard a guess, probably 50-60% of the research and development of EUV was done in the USA, Canada, Taiwan, Japan, and South Korea. So essentially countries firmly within our sphere of influence with the rest occurring in Europe (not the EU, the continent), the People's Republic of China, India, etc. And as we're moving forward, TSMC's stated reason for putting their most advanced fabs in the USA going forward is because they want faster and easier access to where the research is occurring at the bleeding edge of the research without the problem of the time gap that delays and impedes communication due to sleep schedules.

The integrated circuit was invented in the U.S., and the semiconductor industry was pioneered in the U.S.

Op said that the U.S. pioneers and develops technology, giving a first mover advantage. Yes, eventually other countries catch up and start innovating on their own, but the point that the U.S. is generally the pioneer stands.

The IC was invented over 60 years ago. That's human 2-3 generations.

The IP of how to do things (patents) is made up and detrimental to society. Other IP such as brand names is reasonable, we don't want people saying they're selling us Coca-Cola if it isn't from that particular manufacturer.

If there are ways to make stuff, everyone should be allowed to try, because it's in setting up the complex web of inputs and bringing the product to market that people benefit.

The advantage of free markets isn't that it allows the US to make IP, it's that it lets various people have a go at supplying something that people want.

A friend of mine is a scientist in a relevant field, and has caught Chinese PhD students photographing documents in one case, and taking a video recording of a sensitive part of a laboratory in another case. In both cases, what they were recording was not related to their own work.

Not everything is published openly. Some research is kept quiet until the economic value has been realized, and sometimes the "what" is published but not so much the "how".

This happens so often and so frequently that it makes me wonder if there is some official part of the Chinese government set up to consume and distribute this information.

I mean if an American went to a prestigious university or worked a job where they got access to sensitive information for a random company like lets say Ferrari. They wouldnt immediately know who to sell this info too or have a buyer. But it seems like we hear constant stories in such a way that makes me think they were already briefed on who to tell and get this info too.

Idk, maybe im being naive, maybe the CIA sends emails or reaches out to American nationals who work on sensitive projects.

Is that illegal? If not, their actions are not exactly nefarious. Sharing unpublished knowledge is a boon to humanity. (Not that the Chinese are publishing their “findings” openly.)

>Surely, everyone is working on different things here and there, but this idea that there is such critical intellectual property the US that needs to be guarded so intensely in this like James Bond scenario, just feels a little wacky

lots of technology that took billions of dollars and decades to develop? Why would you give that away for free?

Protecting IP isn't some new thing, humans have done it for thousands of years

>As Constantine Porphyrogennetos' warnings show, the ingredients and the processes of manufacture and deployment of Greek fire were carefully guarded military secrets.

>The technological advantage it provided was responsible for many key Byzantine military victories, most notably the salvation of Constantinople from the first and second Arab sieges, thus securing the Empire's survival

https://en.wikipedia.org/wiki/Greek_fire

> lots of technology that took billions of dollars and decades to develop? Why would you give that away for free?

I don't know, ask Alexander Fleming maybe. To save lives? Make the life better for scores of other people, no matter their nationality or class? Why do you even do research, just for the bilions?

>Alexander Fleming

not a great example, Penicillin was only made possible due to government funding to refine the manufacturing process for war

>As to the chemical isolation and purification, Howard Florey and Ernst Boris Chain at the Radcliffe Infirmary in Oxford took up the research to mass-produce it, which they achieved with support from World War II military projects under the U.S. and British governments

and you could make the case that if the Allies just handed that research over to their enemies it would prolong the war

Scientists do not necessarily see the global picture here -- if anything, they'll be inclined to downplay geopolitical and nat-security issues (or at look the other way) if it produces more and better science. This is excluding any consideration of funding -- just what a scientist is supposed to care about.

Some people need to take science more seriously, but there's a kind of crowd that needs to stop worshipping scientists.

>>> if it produces more and better science

Its not even about science anymore, its all about money and prestige [0]. I am not saying there are no principled scientists, but money and prestige makes the world go around.

[0]: https://www.nytimes.com/2021/12/21/science/charles-lieber.ht...

Two sides of the sword: The reason China got nuclear weapon so fast is that USA pushed the top nuclear scientist back to China. This could happen to semiconductor or other fields.

I don't see any solution though, if you don't control, you will lose valuables faster, if you do, you will overdo it sometimes.

Nothing is perfect, if things went too far in the process, let's trust the legal system to sort it out.

Industrial espionage can be an existential matter for companies. They rarely make the headlines but there's plenty of court cases like AMSC vs Sinovel, where Sinovel (China) stole the control software AMSC had developed and pretty much ruined them as a company.

This is a paranoia induced reaction that is being fostered by the media and the war industry, who are interested in creating a new cold war scenario.

Incorrect. It is a MORAL issue. Technology stolen from the US companies is advancing the national security agendas of countries with no regard for human rights and the rule of law. It is immoral to turn a blind eye to this very real threat.

It is kind of curious fact that the US is not a member of ICC. Could you guess why?

There are good reasons why geopolitics cannot be held to strict interpretation of laws, not the least of which is the enforceability paradox, whereby countries more likely to break the norms are less susceptible to enforcement action. This, and attaching the name 'court' on something does not make it any less arbitrary or politically motivated

I'm pretty sure US companies aren't thinking about the morality of it at all. Only $$$ lost and shareholders' emotions.

We know US companies aren't thinking about the morality, nor are US universities, which is why the government is stepping in.

If this was happening during the cold war, CEOs and University Presidents would be being indicted for treason.

Well, many military technologies developed by the US companies are being used to manufacture weaponries used to kill foreign civilians... what's the moral of that story?

Also, by your logic what if totalitarian regimes are to steal medical technologies from US companies to develop medicines for their citizens? Is that moral or immoral?

false equivalence. If you actually cared you would acknowledge the vast difference between China's human rights record and that of the United States.

Interesting. So you think morality is quantifiable? Like it's fine to violate X amount of human rights but not X + 1. I'd like to hear your criteria.

If you think genocide, ethnic cleansing, natural resource exploitation on a global scale, modern day slavery & deadly global pathogen release are equivalent to US misconduct I wouldn't know where to begin in engaging with you.

I can spin this too if you allow me:

"genocide" of Muslims via constant war-waging; "ethnic cleansing" of native Americans; natural resource exploitation on a global scale; "modern day slavery" (i.e. extreme capitalism) both domestically by trapping workers into lifetime debts, and overseas by outsourcing to sweat shops; "deadly global pathogen release" by politicising pandemic responses while failing at it.

Rivalry states smear each others all the time. Also no contender for hegemony is morally superior than another.

> no contender for hegemony is morally superior than another.

Contemplate Nazi Germany winning WWII, or Soviet Russia winning the Cold War. Your contention doesn't stand up to history's lessons.

That’s exactly the point. Had the Nazis won WWII, the narrative would have been the allies were the one that commit genocides, the morally corrupt, and the war-mongers; the Nazis will boast themselves as the most righteous, and most people would have no problem believing that. Hard to imagine for sure, but in a world where the winners write the history, it would be naive to think only the winners must be just because “justice always prevails”.

> It is a MORAL issue. Technology stolen from the US companies is advancing the national security agendas of countries with no regard for human rights and the rule of law. It is immoral to turn a blind eye to this very real threat.

This 100%.

With all due respect, this is completely irrelevant. This same technology and better is being sold out to countries like Saudi Arabia[1] and Israel[2], which are not exactly shining beacons of human rights.

What seems to be the argument here ? "If China wants to use US tech to oppress and kill people with no regard for human rights and the rule of law... they have to PAY for it first, like all respectable totalitarians do"?

[1]https://en.wikipedia.org/wiki/Human_rights_in_Saudi_Arabia [2]https://en.wikipedia.org/wiki/Human_rights_in_Israel

Your point being: since the US provides aid to countries whose human rights records are controversial China should be free to steal and misuse US intellectual property. Great argument. Have fun in a world where China's views on human rights prevail. I'm sure you & your family and their families will be very happy.

>Your point being: since the US provides aid to countries whose human rights records are controversial China should be free to steal and misuse US intellectual property.

That's not my point at all. My point is that it's hypocritical to worry about China using stolen US tech to augment and cement it's oppressive regime when there are countries whose entire dark existence is carved out of US tech and support.

Quantitatively speaking, the marginal utility that China gets from each stolen piece of US tech is probably lower than what they get from each million dollar the US government borrows from them, or each 100000 teenager who downloads tiktok, or from each 100 customer who buy their phones. What I'm trying to say here is that China is not dependent on stolen R&D to opress and brutalize people, they can opress and brutalize people perfectly fine without it, it's just an added bonus on what they already can do.

On the other hand US support for the 2 countries that I mentioned is literally keeping them together against astounding resistence, I would gladly bet my life they would fall apart in <= 10 years if the US was taken out of all geopolitical equations in front of everybody's eyes. If you knew the sheer amount of tyranny and evil the US is single-handedly pumping into the Middle East through those 2 countries alone, you wouldn't be able to sleep at night or day.

So to sum up : I don't think your offense at China's use of US tech is proportional to the amount of damage that tech actually ends up doing. There are other countries where US tech and arms are at least 2 orders of magnitude more destructive and suffering-inducing.

>human rights records are controversial

What a creative way to understate the fact that one of the countries mentioned cut a journalist in half with a chainsaw, while the other bombs civilians on a yearly basis and kills literal babies. "Controversial" is how you describe an things where there are credible differences of opinion on; not raw, pure evil.

Surveillance tech, AI, civil engineering & biotech are all examples of appropriated technologies used for heinous crimes against Tibetans, Uyghurs, Falun Gong & general dissent in China; and to hold SE Asia hostage to vast water sources now controlled by China. You're mistaken because you're poorly informed.

You're similarly misinformed about Israel. The IDF Ruach Tzahal sets forth an ethical code of conduct that has resulted in numerous prosecutions when violated. Yes, immoral actors (eg. Ariel Sharon) can effect terrible suffering as individuals on civilian populations. That's war and everything about it sucks. But, you won't find serious ethical codes of conduct in the Saudi or Chinese doctrines.

And this is my point... mistakes, imperfections, atrocities... it is inseparable from sustained conflict regardless of who is fighting. Some countries try to do better inside this reality, some pay lip service, some don't care. You lump it all together out of some personal and poorly informed bias.

The point is that if it were a MORAL issue as you suggested, then the US would not have sold their technologies to other countries whose human rights records are controversial. The fact that they had not problem doing that entails that it was not a moral issue, at least for the US companies.

If your problem is with human rights and the rule of law, you should stop US violations first, because the US the most powerful country. The American military continues its aggression on human rights around the world.

One of my favorite college professors got nabbed for secretly collaborating with the Japanese government: https://voiceofoc.org/2014/02/former-uci-professor-pleads-gu...

It does happen, even at university level.

I worked in cyber security at a major industrial company for 5 years. Yes “insider threat” is a very real concern and people get busted exfiltrating company data all the time.

I'd like to point out that the majority of replies to this comment are about industrial espionage, and not academic research.

Academic research tends to run some time ahead of industrial applications.

But the discussion we're having right now is about academic research, which gets published openly. Professors move between institutions all the time. Has Harvard sued MIT because some professor didn't get tenure and took their research and students with them?

I think that there's a difference if the professor in question were consulting for US companies, or specifically in projects within the DoD(versus receiving a grant, which again is a different situation), and shared those with Chinese entities. But as far as I can tell, nobody in this comments section is looking into that and we're just having arguments about other situations instead of what has actually happened. That obfuscates the matter at hand.

You can't be serious about opsec and not take seriously the risk of an internal bad actor with access to your systems.

Yeah, that was my thought - they treat everybody like a spy, because that's literally their job description.

The majority of the cases I'm familiar with had an internal bad actor.

If you don't take that into account when formulating your security policy and threat model then you have not done your homework.

The only critical approach is to begin with the beginning and the trigger event, refusal to provide the passwords requested to his devices. In the UK it's a crime to refuse law enforcement request for your passwords. If the similar is true in the USA, or there's critical details involved - which frankly only CCTV footage could possibly reveal - about this incident, then there's your cause and effect from whence forward everything looks like semantic disassembly to me on the academics part and a desperate need to show substantive accusations on the governments part.

> ... is this level of industrial/scientific espionage even a critical factor in our current world?

Yes

A history of espionage and theft of trade secrets demonstrates the necessity of concern.

"Xiaoqing Zheng who holds dual-citizenship in the United States and China, used elaborate and sophisticated methods to steal countless digital files containing trade secrets from General Electric regarding their wind turbine technology" https://blog.twinstate.com/news/ge-trade-secrets-theft

Jizhong Chen and Xiaolang Zhang indicated in theft of trade secrets https://www.cnbc.com/2019/12/09/apple-has-deep-concerns-that... https://www.digpu.com/business-and-finance/chinese-man-jizho...

"According to today’s conviction, Xu attempted to steal technology related to GE Aviation’s exclusive composite aircraft engine fan – which no other company in the world has been able to duplicate – to benefit the Chinese state." https://www.justice.gov/usao-sdoh/pr/jury-convicts-chinese-o...

" A notable example is the 2014 lawsuit by T-Mobile, in which Huawei was accused of, among other conduct, sending an engineer to a T-Mobile facility to see "Tappy," the company's computer driven robot with a mechanical arm used to test smartphone screens to improve the reliability of its handsets. The engineer slipped one of the robot's fingertips into his laptop bag but was caught on camera" https://www.mondaq.com/unitedstates/trade-secrets/1052104/th...

"A former associate scientist was sentenced to 24 months in federal prison in federal court today for stealing proprietary information worth more than $1 billion from his employer, a U.S. petroleum company." https://www.justice.gov/opa/pr/chinese-national-sentenced-st...

There are dozens of such visible cases. The vast majority of trade-secret theft is uncaught.

The opposite of 'wacky' - key R&D and especially know-how is critical to the development of specific industries and in some cases, it should be guarded.

It can take decades of effort, R&D, product development to develop the key understanding towards making for example, jet turbines, high speed trains, chip fabs etc..

Economic espionage is real and part of China's industrial strategy.

It should be noted that 'everyone else' will also steal R&D to the extent that it's related to weapons, but generally not industrial information.

That said - probably most research isn't that important, and this kind of thing can easily turn into 'Witch Hunt'. Paradoxically, the CPP will also use an overbearing posture by the US on this issue as part of propaganda efforts.

The CCP right now is fighting a 'narrative war' with all the touchy issues: they are circulating information indicating that COVID started in Europe, and in response to stern words by Canada PM Trudeau, China is telling it's citizens the a recent outbreak is due to 'frozen food from China'. Controlling information means being able to tell their citizens whatever they want and having them believe it for the most part. That a few independent thinkers want to 'VPN out' and form their own opinions isn't that important if you can tell 99% of people what the truth is.

CCP is playing 'full court press competition' - they are engaging and comprising students on campus [1] , seeking information from well placed ex-patriots [2], planting spies everywhere [3], grabbing people and re-patriating Chinese citizens abroad by force [4], running disinformation campaigns [5], compromising state officials [6] and more.

So it's important that our 'values' of true justice, equality, transparency etc. are not compromised, and we don't want to regress to early 20th century racism, but at the same time, it's a 'war' of sorts and it's incredibly naive to operate on the basis that this is not happening and affecting everyone.

Probably it will take more than just a simple plan, but a major new way of thinking about China, and a whole series of changes in that regard.

Even things such as governments new 'Trustless' IT policy published here yesterday, are a step in the right direction.

[1] https://www.propublica.org/article/even-on-us-campuses-china...

[2] https://www.csis.org/programs/technology-policy-program/surv...

[3] https://www.washingtonpost.com/opinions/explain-the-chinese-...

[4] https://thethaiger.com/news/world/china-forced-nearly-10000-...

[5] https://thediplomat.com/2020/06/chinas-disinformation-campai...

[6] https://www.sfgate.com/politics/article/Barbara-Boxer-China-...

> So it's important that our 'values' of true justice, equality, transparency etc. are not compromised, and we don't want to regress to early 20th century racism

This is a challenging problem for a democracy being exploited by an authoritarian government.

The challenge is 1) Not causing xenophobia (as the article implies) while still 2) recognizing and preventing the practices and influence of the CCP on our population, corporations, and economy which do not align with our moral values and also preventing the export of externalities onto democracies whose very structure allows it.

Most of the books I read have something along the lines of first recognizing the author's appreciation for and love of the Chinese people then discussing what the CCP is doing and what a democracy can do to protect itself without compromising its values.

If anyone looking into this dichotomy has not seriously studied the subject and just thinks the US government is trying to start a Cold War, is paranoid, or that open markets and will magically fix this problem I recommend a book published in 2022. "The World According to China" - Elizabeth C. Economy.

Yes, that's exactly it.

The problem is that the 'infiltration' is pervasive, as designed by the CCP and it therefore gets really nuanced in terms of trying to uphold justice and anti-discrimination.

CCP is pursuing '360 Thought Control' within their borders ... but using every avenue outside their borders as well.

There are also subtle gradient problems as well: some Chinese students on campuses which have had CCP intimidation have state 'yes it's a thing, but really most of us consider it to be kind of ridiculous'. In that case, it may be a lot of 'intent' but little real influence.

But the indirect, knock-on effects can be as dramatic. I once saw a presentation by an American early state investor for the Shenzhen/HK area, keen to give an idea of how the area was expanding. His presentation talked about strategic initiatives of the local governors there, and it was right out of the governments talking points. In short, he was acting as shill for the CCP. I think the only reason he was doing it was to stay in the 'good books' of local governors and to develop relations. The leverage consistently utilized by Chinese government on all such matters is palpable. Hollywood which is an array of people within an industry, not a 'single entity' exemplifies this. The Dalai Lama no longer gets many meetings with high profile dignitaries, and Hollywood has basically 'shut up about the issue' for the most part due entirely to economic coercion.

Certainly, one of the measures Western Nations should implement is a transparency policy whereby their measures are given public airing, and possibly 'tit for tat' sanctions, which is to say all measures taken by them imply some kind of in-kind response. It would take a lot of coordination.

Edit: China is indicating the recent Omicron outbreak there is from Canadian frozen food (not Chinese, as I miswrote). A blatantly terrible bit of propaganda.

Reply on news.ycombinator.com