intel – Techdirt (original) (raw)

American Companies Are Helping Power Russia’s Massive Facial Recognition System

from the inadvertently-aiding-in-the-war-(at-home)-effort dept

Russia’s fighting a war in Ukraine and a war at home. As residents express their displeasure with their government, the government’s cameras and facial recognition AI are going into overdrive to ensure Putin and his pals control the narrative.

Unfortunately, the Russian government is getting some help from the United States, albeit inadvertently.

Russia has been using cameras powered by facial recognition systems to crackdown on dissidents, according to reporting from Reuters. Several Russian companies are using algorithms trained and powered by chips made by U.S. firms Intel and Nvidia. Reuters said that one of the companies even received money from U.S. intelligence.

The full article from Reuters gives a more in-depth explanation of what’s going on here. For years, Russia has been expanding its domestic surveillance network. And it has always been used to track dissidents, opposition party members, and other government critics. Handling real-time facial recognition requires a lot of hardware power, and for that, the Russian government has turned to American tech companies.

The facial recognition system in Moscow is powered by algorithms produced by one Belarusian company and three Russian firms. At least three of the companies have used chips from U.S. firms Nvidia Corp or Intel Corp in conjunction with their algorithms, Reuters found. There is no suggestion that Nvidia or Intel have breached sanctions.

At this point, neither Nvidia and Intel are selling directly to Russia. Both companies ended all shipments to the country following the enactment of export restrictions last March. Whatever was purchased prior to the blacklisting was above-board, and what’s already in the Russian government’s hands is beyond the control of these companies.

More concerning is the US government’s slightly more direct participation in the development and expansion of Russia’s facial recognition programs.

Reuters also found that the Russian and Belarusian companies participated in a U.S. facial-recognition test program, aimed at evaluating emerging technologies and run by an offshoot of the Department of Commerce. One of the firms received $40,000 in prize money awarded by an arm of U.S. intelligence.

$40,000 is a drop in the surveillance budget bucket, but it’s still a bit disturbing to see the US government handing out money to companies most likely already providing surveillance tech to known human rights abusers. While it’s true that, as a spokesperson for the IARPA program stated, an award is not the same as providing direct assistance in oppressive surveillance programs, it’s still not a good look for the US Commerce Department or the National Institute of Standards and Technology — both of which are involved in awarding prizes to participants in IARPA (Intelligence Advanced Research Projects Activity) challenges.

While the US tech providers are doing what they can to prevent their products from heading to Russia, all they can really do is stop sending GPUs and other hardware there themselves. The Russian government has fans all over the world and it appears people who want to put these power graphics processors in the government’s hands are buying on behalf of the blacklisted nation.

Russian customs records show that at least 129 shipments of Nvidia products reached Russia via third parties between April 1 and Oct. 31, 2022, however. Records for at least 57 of these shipments stated that they contained GPUs. In response to these findings, the spokesperson said, “We comply with all applicable laws, and insist our customers do the same. If we learn that any Nvidia customer has violated U.S. export laws and shipped our products to Russia, we will cease doing business with them.”

Intel isn’t doing any better at preventing customers from making straw purchases for a nation that earned itself additional export controls following the Ukraine invasion.

Reuters has previously reported that at least $457 million worth of Intel products arrived in Russia between April 1 and Oct. 31, 2022, according to Russian customs records. “We take reports of continued availability of our products seriously and we are looking into the matter,” an Intel spokesperson said

The end result is the events detailed in the rest of the Reuters report, which is definitely worth checking out. The system — at least the facial recognition end of it — works. Reuters reviewed over 2,000 criminal cases, finding overwhelming evidence that most of the arrests and detainments were triggered by citizens — many of the anti-government protesters — passing by cameras deployed by the Russian government.

Through no fault of their own, American companies are now accomplices in oppression. While Nvidia and Intel appear to be doing what they can to comply with US regulations, there’s not much they can do to stop third parties from bypassing these restrictions. And there’s even less they can do about the products that are already in use, except take precautions in the future to limit their tech’s contribution to world’s many, many jackboots.

Filed Under: facial recognition, russia, surveillance
Companies: intel, nvidia

Intel, Who Coined The Term ‘Patent Troll’ Makes Deal With Patent Troll To ‘Monetize’ Unused Patents

from the making-a-mockery-of-what-intel-stood-for dept

The pattern is pretty common: young companies innovate, older companies litigate. When you can’t keep up and you can’t succeed by beating the market, you turn to the courts to try to squeeze cash out of those more successful than you. Often this involves patent lawsuits or handing your unused patents off to patent trolls to “monetize” them for you.

Over the years there have been a bunch of “patent aggregation” firms that showed up, offering to buy up patents and then promising to bring in free money by “licensing” them. The most well known of these was probably Intellectual Ventures, which scooped up tons upon tons of unused (perhaps because they were useless) patents, many from universities with loss-making tech transfer offices, which were eager to sell out in order to show they could make some money.

There were others, including one called IPValue, which has been around for a while now as well, and has a history of getting big, lumbering, no longer innovative companies to sell off useless patent portfolios for the sake of “licensing” them. Of course, when no one wants to license totally useless patents, the company has been known to engage in litigation. Like Intellectual Ventures, which came about with high minded talk of building these portfolios to enable new innovation, the reality is that they’re classic trolls, collecting tons of patents to then seek to shake down actual operating companies, not for copying ideas and infringing, but for doing something obvious with the technology that some vague, forgotten patent sorta kinda, maybe could describe if you squint and ignore the fact that patents are only supposed to cover “non-obvious” ideas.

Anyway, Intel recently announced that it was handing 5,000 patents off to IPValue, which (as it does in these kinds of deals) spun up a shell corporation called Tahoe Research Limited to go see who it can shake down over these patents. Usually, the way these deals work is that the company, Intel, gets some relatively modest amount of cash upfront, but also a piece of anything the troll can squeeze out of others. Considering these are basically zero value patents for Intel, the temptation must be great to at least get something out of them.

The Intel IP portfolio is likely to comprise older patents that have not yet expired but no longer represent cutting-edge technologies so the chipmaker likely sees this licensing arrangement as a way of maximizing its revenue from some of its patents by allowing IPValue to handle the legwork in licensing them to third parties.

Of course, the real irony here — which long time patent observers have picked up on — is that the very term “patent troll” was initially coined by two Intel lawyers back in the 1990s. They were frustrated by all these patent owners (early on it was often ex-patent litigators who would set up a shingle, buy up some patents, and shake down firms, before the aggregators “professionalized” the business) demanding they pay up for obvious bullshit patents.

So, it seems notable that Intel is now embracing patent trolling.

Though, perhaps not that surprising. After all, when companies are still innovating and leading the way, they’re the target of patent trolls. It’s only after their best days appear to be behind them, and they’re desperate to keep Wall St. happy that suddenly they turn to trolling. And, Intel*, which spent decades as the top innovator in chips, has lately been looking like it’s completely lost its way. And that’s been reflected in the plummeting stock price. Basically, the company that spent decades doing no wrong… has had a lot go wrong lately.

And, so, the company is now desperate. Young companies innovate. Old companies litigate. Tragically, it looks like Intel (a company I actually worked for decades ago), has moved from one category to the other. It’s shame.

Filed Under: innovation, patent aggregators, patent troll, patent trolling, patents
Companies: intel, ipvalue, tahoe research limited

Intel Wants To Add Unproven ‘Emotion Detection’ AI To Distance Learning Tech

from the and-what-does-disabling-the-camera-denote? dept

Last week, Zoom announced its plans to add emotion detecting tech to its virtual meeting platform, something it apparently felt would facilitate the art of the deal. Here’s Kate Kaye, breaking the news for Protocol.

Virtual sales meetings have made it tougher than ever for salespeople to read the room. So, some well funded tech providers are stepping in with a bold sales pitch of their own: that AI can not only help sellers communicate better, but detect the “emotional state” of a deal — and the people they’re selling to.

In fact, while AI researchers have attempted to instill human emotion into otherwise cold and calculating robotic machines for decades, sales and customer service software companies including Uniphore and Sybill are building products that use AI in an attempt to help humans understand and respond to human emotion. Virtual meeting powerhouse Zoom also plans to provide similar features in the future.

Advocates for this AI, which includes customers like Zoom, claim this unproven tech could make it easier to “build rapport” during virtual meetings or, at the very least, give those performing pitches a heads up when they’re losing their audiences.

That’s all well and good when we’re talking about a bunch of consenting adults playing sales pitch poker while attempting to Voight-Kampff their way into a competitive edge. Any advantage should be exploited, even if it means subjecting potential customers to AI with no proven track record. It’s unclear how consent to be emotionally analyzed is obtained (or if it’s even sought), but, again, we’re dealing with adults in a sales situation where this sort of manipulation is considered normal behavior.

The problem with Zoom is it thinks this same tech should be inflicted on non-consenting minors. Again, it’s Kate Kaye with the news for Protocol.

Rather than simply allow instructors to draw inferences from student facial expressions and behavior, a couple of companies think they can make teachers better by throwing more tech (and surveillance) at their students.

Intel and Classroom Technologies, which sells virtual school software called Class, think there might be a better way. The companies have partnered to integrate an AI-based technology developed by Intel with Class, which runs on top of Zoom. Intel claims its system can detect whether students are bored, distracted or confused by assessing their facial expressions and how they’re interacting with educational content.

“We can give the teacher additional insights to allow them to better communicate,” said Michael Chasen, co-founder and CEO of Classroom Technologies, who said teachers have had trouble engaging with students in virtual classroom environments throughout the pandemic.

This means the cameras always need to be on, even though some instructors are capable of teaching classes without expecting students to open up a window to their home lives via laptop cameras. IM services, microphones, and texting seem to fill the face-to-face void quite capably. The ability to strip things back to text-only communication allows students without access to speedy internet connections to stay connected without exceeding the bandwidth they’re allotted or burning up their data if their operating under a cap.

The business version requires always-on cameras to record footage that can then be processed by the emotion detection AI to provide customers with insights on detected mood swings by their sales pitch recipients. Presumably, the school version will operate the same way until Intel and Classroom Technologies feel the AI has learned enough to go live.

The end goal is always-on surveillance of students, with the stated goal being better instruction and more student engagement.

“We are trying to enable one-on-one tutoring at scale,” said [Intel research scientist Sinem] Aslan, adding that the system is intended to help teachers recognize when students need help and to inform how they might alter educational materials based on how students interact with the educational content.

At this point, the product is still in the testing phase. To become a full-fledged product, it will need significant buy-in from educational institutions. That’s the sort of thing that often happens without consulting the stakeholders most affected by the addition of new in-home surveillance tech: the students who will be the testing ground for the product.

Even if the reps for both companies are to be believed — that the product is intended to help teachers better reach their students — the potential for misuse (or deliberate abuse) is omnipresent. On top of that, most humans are incapable of accurately reading the emotions of others and they’ve got a lifetime of experience and better innate learning systems. Add to that the fact that not all cultures utilize the same expressions or body language to signal mood shifts, and you’ve got a product with the potential to generate a ton of useless or counterproductive data.

Filed Under: ai, classrooms, emotion detection, surveillance, virtual learning
Companies: classroom technologies, intel, zoom

DRM Breaking Games Again, This Time Due To New Intel Chip Architecture

from the chips-and-blip dept

We were just discussing how Denuvo’s inability to renew one of its domains suddenly prevented lots of paying customers from playing several of their paid-for video games. While we can laugh at Denuvo’s ineptitude, the real point in all of that is once again how DRM in video games tends to prevent nothing when it comes to piracy, yet paying customers tend to get impacted for a variety of reasons. DRM, in other words, almost universally functions to punish paying customers, which is stupid.

And now here we are again, with DRM suddenly preventing paying customers from playing their games, albeit for a completely different reason. Intel released a list of something like 50 games where DRM breaks playability as a result of Intel’s new chip architecture. While the reason this occurs on these chips is somewhat technical, ArsTechnica has a writeup that includes a reasonable summary.

We’ve already covered how Alder Lake’s hybrid “big.little” design splits the CPU’s workload into high-powered “performance” (P) cores and low-powered “efficiency” (E) cores. But after hinting at the potential issue in a developer FAQ last month, Intel is now confirming that some games contain DRM that Intel says “may incorrectly recognize 12th Generation Intel Core Processors efficient-cores (E-cores) as another system.” That issue can lead to games that “may crash during launch or gameplay or unexpectedly shut down,” Intel says.

PC Mag’s Chris Stobing explained that the issue arises from the DRM middleware treating the two different types of cores as two distinct systems. “Once it detects that some portion of the load has been split between the P- and E-cores, it sees the new cores as a new license holder (a separate system) and force-quits the game to prevent what it believes is two PCs trying to play one game on the same key,” he said.

That makes technical sense. What doesn’t make sense is why this DRM is used in the first place. Again, I can just about promise you that all or a majority of these games are being pirated anyway, despite the DRM. So here again we have the classic DRM scenario: pirates going to pirate, while the paying customer finds out the game they bought suddenly isn’t operable any longer.

Now, Intel has indicated that it will be patching this issue out and is working directly with game developers to do that. In addition, it has provided workarounds, such as:

Intel says users can get around the problem by upgrading to Windows 11 (in some cases) or launching into the BIOS setup and turning on Legacy Game Compatibility Mode, then activating it with the Scroll Lock key. But an Intel spokesperson told PC Mag that this mode isn’t yet available on at least some Alder Lake CPUs and “should be included in a future BIOS update.”

But when viewed through the lens of the average PC gamer, this is mostly absurd. Upgrade the entire OS to play a game you already bought? Making changes in the BIOS? Come on.

DRM needs to die. It’s almost entirely useless at everything other than screwing things up for game companies’ actual customers. How can all of this be worth it?

Filed Under: drm, video games
Companies: intel

Opening Up Information In A Pandemic, Rather Than Locking It Down: The Open COVID Pledge Is Important

from the information-sharing-is-the-way-forward dept

People who are actually engaged in real innovation know that the real breakthroughs and advancements don’t come from solitary geniuses having a eureka moment, but from the open sharing of information to bring in a variety of perspectives and to build upon the work of others. And yet, for years, some have drilled the myth of the lone inventor into people’s minds, along with the idea that we need to lock up ideas and knowledge to give those inventors “incentive.” Yet, if the true advancements come from people sharing and building on each other’s ideas, this is the exact wrong approach. Now in the midst of a massive global pandemic we’re seeing the ridiculous outcome of people trying to abuse or expand exclusivities — which most actual innovators know will hinder, rather than help.

So it’s exciting to see many pushing for a very different approach. A bunch of smart innovators, organizations, and academics have put together the Open COVID Pledge, in which they all agree to share whatever information and technologies they have that might otherwise be locked down, free of charge, for the purpose of ending the pandemic:

Immediate action is required to halt the COVID-19 Pandemic and treat those it has affected. It is a practical and moral imperative that every tool we have at our disposal be applied to develop and deploy technologies on a massive scale without impediment.

We therefore pledge to make our intellectual property available free of charge for use in ending the COVID-19 pandemic and minimizing the impact of the disease.

We will implement this pledge through a license that details the terms and conditions under which our intellectual property is made available.

I’m happy that this is being done, but honestly am a bit disappointed in how few organizations have signed up. Among companies the biggest is Intel, which is good to see, as that company can be a bit overly aggressive in patent enforcement at times. Mozilla and Creative Commons are there — which isn’t too surprising. There are a bunch of universities — but mainly the law school parts of the universities, rather than say, the science and engineering parts.

Hopefully more companies and universities (including science and engineering departments) will start to sign up for this. The middle of a pandemic is no time to be fighting over who gets to own what tiny piece of the solution.

Filed Under: covid-19, information sharing, innovation, open covid pledge, patents
Companies: creative commons, intel, mozilla

Former Refrigerator Manufacturer Says Companies Using Open Source, Royalty-Free Video Technology Must Pay To License 2,000 Patents

from the this-is-why-we-can't-have-nice-things dept

Video streaming is a key part of today’s Internet world. According to research from Sandvine, it represents 60.6% of total downstream volume worldwide. The centrality of video to the Internet experience makes video codecs one of the hottest technologies. The most popular format today is H.264, used by 91% of video developers. But H.264 is getting long in the tooth — its history goes back two decades. An upgrade is long overdue. There’s a successor, H.265, also known as High Efficiency Video Coding, or HEVC. However, the use of H.265 has been held back by patent licensing issues. As Wikipedia explains in painful detail, there are two main patent pools demanding payment from companies that use HEVC in their devices. For one of the pools, the patent list is 164 pages long. Partly in response to this licensing mess, and HEVC’s high per-device cost, the Alliance for Open Media was formed in September 2015:

Seven leading Internet companies today announced formation of the Alliance for Open Media — an open-source project that will develop next-generation media formats, codecs and technologies in the public interest. The Alliance’s founding members are Amazon, Cisco, Google, Intel Corporation, Microsoft, Mozilla and Netflix.

In contrast to the proprietary and expensive H.265, the new video standard, called AOMedia Video 1 (AV1), is open source and royalty-free. Those features, and the backing of many of the top Internet companies, would seem to make it an obvious choice for manufacturers to build into their devices, leading to better-quality video streaming for end users at no extra cost.

Life is never that simple. Back in March last year, Sisvel announced a “patent licensing program” for AV1. Sisvel is an Italian company that began as a manufacturer of white goods, particularly refrigerators, and has morphed into a group that “identifies, evaluates and maximizes the value of IP assets for its partners around the world”. The AOMedia group wrote in response:

AOMedia is aware of the recent third-party announcement attempting to launch a joint patent licensing program for AV1. AOMedia was founded to leave behind the very environment that the announcement endorses — one whose high patent royalty requirements and licensing uncertainty limit the potential of free and open online video technology. By settling patent licensing terms up front with the royalty-free AOMedia Patent License 1.0, AOMedia is confident that AV1 overcomes these challenges to help usher in the next generation of video-oriented experiences.

But refrigerator companies don’t give up that easily. Sisvel has just announced that more companies have added patents to its pool. There are currently 1,050 patents that Sisvel says must be licensed, but in due course it expects that number will rise to around 2,000. The fact that people can claim that there are 2,000 separate patents involved in a video encoding format is an indication of how far the patenting madness has gone. The sheer number claimed for a single technology is an indication of how trivial most of them must be — and thus by definition undeserving of monopoly protection.

According to an article on c|net, Sisvel is “willing to pursue companies that don’t pay its AV1 licensing fees”. This probably means we are in for another few years of utterly pointless legal battles over who “owns” certain ideas. That’s bound to cast a chill over this whole area, and to negate some of the benefits that would otherwise flow from an open source, royalty-free video standard. Companies will waste money paying lawyers, and end users will miss out on exciting applications of the technology. And all “because patents”.

Follow me @glynmoody on Twitter, Diaspora, or Mastodon.

Filed Under: appliances, av1, h.265, patent trolls, patents, refrigerators, software patents, streaming video
Companies: alliance for open media, amazon, cisco, google, intel, microsoft, mozilla, netflix

Apple and Intel Sue SoftBank-Funded Patent Troll, Claiming Antitrust Violations For Patent Trolling

from the well-this-is-interesting dept

For years, I’ve argued that since patents and copyrights are government granted monopolies, it seems pretty straightforward to me that abusing those laws to stifle speech, innovation, or competition should be viewed as an antitrust violation. It’s taken a while, but earlier this decade, the Supreme Court actually agreed with regards to patents (it’s not there yet on copyright…).

So for all the talk about using antitrust laws against the tech giants, perhaps there’s a much better use of antitrust law in taking down abusive patent trolling operations. At least that’s the theory in a new lawsuit filed by Intel and Apple against patent troll Fortress Investment Group LLC. As the lawsuit notes, Fortress was a struggling company that was acquired by SoftBank in 2017 for $3.3 billion dollars, and then turned into a massive patent troll.

The complaint is a really good read on how patent trolls could violate antitrust laws. It even starts out by explaining how trolls violate the patent clause of the Constitution by not “promoting the progress.”

Rather than promote the progress of science and useful arts, patent assertion entities (?PAEs?), including Defendants, that aggressively pursue meritless litigation have long been recognized to harm and deter innovation. For example, one study estimated that patent litigation brought by PAEs in the United States resulted in expenditures of $29 billion in 2011 for licensing fees, legal fees, and other costs of responding to PAE litigation. Another study found, by looking at the impact on stock price, that lawsuits by PAEs from 1990 through 2010 were responsible for the defendants losing half a trillion dollars. And those losses are not offset by corresponding gains to patent holders that promote innovation. One study found that the profits received by PAEs from litigation amounted to less than 10% of the lost share value of companies targeted by the PAEs.

The complaint highlights, as we’ve detailed here over the years, how the Supreme Court has pushed back on excessive patenting and patent trolling, but also how patent trolls have “evolved” to deal with this:

In the face of these challenges, PAEs have evolved. PAEs have increasingly been partnering with investment firms to fuel their litigation. This trend is part of a larger trend in the growth of third-party investment in litigation generally. Although the precise scale of investment in litigation is unknown, estimates put it in the tens of billions of dollars.8 As one example, the largest litigation investor reported having investments of $2.8 billion in 2019.

Having deep-pocketed investment firms standing behind them has made PAEs only more aggressive. Indeed, to meet the expectations of their new investors for high returns, PAEs must act ever more aggressively. These new investors are content to incur loss after loss so long as they have the chance to hit a windfall reward that will justify their investment. Patent assertion thus becomes simply a numbers game disassociated from the merits of the underlying patents, with PAEs and their investors betting that serial assertions with aggressive demands will strike a jackpot eventually making up for many other losses. Consistent with this strategy, while the overall level of patent litigation may be declining, assertions by non-practicing entities are increasing.

And that brings us to Fortress Investment Group:

central player in this emerging investment strategy is Fortress Investment Group LLC (?Fortress?). Fortress is an investment firm that went public in 2007. Fortress?s shares traded at over 35pershareaftergoingpublic,butonedecadelater,Fortresswasstrugglingwithpoorreturnsanditssharepricehadplummetedtoaround35 per share after going public, but one decade later, Fortress was struggling with poor returns and its share price had plummeted to around 35pershareaftergoingpublic,butonedecadelater,Fortresswasstrugglingwithpoorreturnsanditssharepricehadplummetedtoaround5 per share in 2017. Fortress was acquired that year by SoftBank Group Corp. (?SoftBank?) for 3.3billion.Fortresscontendsitis?aleading,highlydiversifiedglobalinvestmentmanager?11andclaimstohaveapproximately3.3 billion. Fortress contends it is ?a leading, highly diversified global investment manager?11 and claims to have approximately 3.3billion.Fortresscontendsitis?aleading,highlydiversifiedglobalinvestmentmanager?11andclaimstohaveapproximately39.2 billion of assets under management as of March 31, 2019. One way in which Fortress has tried to turn around its performance and justify SoftBank?s investment in it is through increased speculation on patent assertions.

And the underlying issues in this case: Fortress has set up a ton of shell operations, and constantly shake down all sorts of companies actually making stuff:

Intel and Apple bring this complaint to end a campaign of anticompetitive patent aggregation by Fortress and a web of PAEs that Fortress owns or controls. Fortress has used its stable of PAEs to aggregate a massive portfolio of patents that purportedly read on high-tech consumer and enterprise electronic devices and components or software therein and processes used to manufacture them. By employing a network of PAEs that it either owns or controls, Fortress has created a web of entities that obscures Fortress?s puppeteering role in this scheme. Rather than enhancing efficiency, Fortress uses aggregation to undermine it by creating a structure in which Fortress and its PAEs benefit by asserting weak patents?i.e., those that never would have been asserted by their former owners?in order to stretch the resources of their targets and increase the possibility that those weak patents will improperly be found valid and infringed or the prospect that a target (like Intel or Apple) will agree to a license to resolve the threat posed by Fortress and its PAEs. Thus, rather than promoting the procompetitive benefits of the patent system by increasing innovation and output, Fortress?s scheme has the opposite effect. Fortress and its PAEs acquire and seek to monetize meritless patents that never would have been asserted by their original owners, imposing a tax on the electronics industry that increases prices, decreases output, and ultimately harms consumers. To the extent that Fortress and the other Defendants have patents that would actually be of value to potential licensees, the transfer of those patents to Fortress?s control limits access to them because those patents are now held by entities that have no incentive to license patents in a way that captures royalties that are commensurate with their actual value. Instead, those entities have incentives to obtain excessive monopoly rents by exploiting patent portfolios that aggregate any valuable patents with many meritless patents.

I wouldn’t normally quote so much directly from such a complaint, but it’s so clearly stated and easy to understand that it’s useful. The complaint makes it clear that this is an attempt to “extort” money from operating companies, and that the whole game is based on volume, rather than any validity to the claims:

Through its anticompetitive aggregation scheme, Fortress has engaged in anticompetitive conduct by creating a portfolio of patents that purportedly read on electronic devices and components or software therein and processes used to manufacture them that allows it to charge far more than the value of the inventive contributions (if any) of the patents and of competitive prices for licenses. Fortress and its PAEs seek to use that ill-gotten power to extract and extort exorbitant revenues unfairly and anticompetitively from Intel, Apple, and other suppliers of electronic devices or components or software for such devices and ultimately from consumers of those products. Fortress?s aggregation is thus intended for an anticompetitive purpose?to invest in patents at costs lower than the holdup value of the patents to ensnare as many potential licensees as possible and to allow it and the other Defendants to assert as many possible claims of infringement to tax the commercial use of existing technology at rates beyond the actual value (if any) of the aggregated patents.

In furtherance of the anticompetitive scheme, Fortress and the other Defendants have deployed patents in waves of lawsuits against their targets without regard for the merits of the claims. Rather than licensing and litigating based on the merits of the patents, Fortress and its PAEs operate based on volume and repetition, targeting the resolve of the targets instead of establishing the merits and value of the patents. Given the size of the portfolio, Fortress and its PAEs can deploy patent after patent in case after case against their targets with the threat of ever more patent assertions and ever more litigation. Faced with this threat, many victims have agreed to settle, rather than to challenge Fortress and its PAEs, for amounts that reflect not the merits of the underlying patents but the effectiveness of the Fortress model. Thus, Fortress and the other Defendants foreclose the possibility?which existed before aggregation?that litigation can be an economic alternative to licensing patents.

The complaint details some of Fortress’ practices in pushing for more and more patent trolling in the operations it supposedly “invests” in:

Fortress describes its investing approach as ?making control-oriented investments in cash flow generating assets.? When it comes to patent investments, Fortress has taken its ?control-oriented? approach to an extreme. Fortress?s model is to condition its investments in PAEs on terms so severe that the PAEs have no choice but to make aggressive and reckless patent assertions to attempt to generate the revenue required to meet their obligations to Fortress. When they fail to do so?as is often the case?Fortress steps in and assumes even more control and/or ownership of the patents, allowing it to ratchet up the aggressiveness of the assertions. In other instances, such as with VLSI, Fortress has skipped this intermediary step of finding a partner to do its bidding and acquired patents through a subsidiary outright from the start. The result is that Fortress has either acquired or controls a portfolio of well over a thousand U.S. patents for high-tech consumer and enterprise electronic devices and components or software therein and processes used to manufacture them to deploy against its targets.

It also highlights how Fortress’ own execs more or less admit that this is the game plan:

Fortress has targeted suppliers of high-tech consumer and enterprise electronic devices or components or software for such devices because they provide attractive targets for repeated and meritless assertions. An article co-authored by Eran Zur, Managing Director of the Intellectual Property Finance Group at Fortress, observes that courts can grant ?oversized awards? in the technology sector that ?stem from the sheer complexity of interoperable components and systems sold as part of functional units, if not integrated devices.? Further, the article notes that ?because technology invention tends to be incremental, to the extent an individual patent owner can be awarded damages on the price of the entire end product as opposed to their specific patent claim, a litigation incentive arises.? That litigation incentive is coupled with what the article notes are ?the substantial legal costs to defend a patent infringement suit,? creating a situation in which ?speculative behavior drives an ever-inflating price ceiling (given the possibility of oversized damages) [and] a price floor becomes set by the extreme expense of litigation defense, marked at just under nuisance value.?

It then highlights how this abuse of monopoly power is anti-competitive:

Aggregating patents in the way that Fortress has done harms competition. First, by aggregating patents covering technologies that are actual or potential alternatives for one another, Fortress injures competition in the same way as any merger or combination of competitors. Before aggregation, when multiple parties held such patents, those parties competed with one another to license the patents, and licensees benefited from that competition through more favorable licensing terms. Multiple holders of substitute patents were forced to compete with each other to offer better terms to secure licensees. Once the patents were aggregated and controlled by Fortress, however, that competition was eliminated.

Second, Fortress introduces a new cost to suppliers of electronic devices and the components and software for those devices that dampens incentives for product suppliers to invest in research and development to drive innovation, thereby further undermining competition and harming end consumers. Exposing the targeted suppliers to another cost benefits their competitors by making the targeted suppliers? products more expensive and/or less innovative. Those competitors might have previously owned some of the patents aggregated by Fortress but were unable to impose such high costs on suppliers using technologies claimed by the patents when the patents were not aggregated into a massive portfolio. Fortress?s aggregation thus undermines competition in the sales of electronic devices and components and software for those devices.

Third, the higher royalty payments that Fortress and its PAEs generate reward the creation of patents that are not actually inventive or are not actually used. Thus, the higher royalties that patent aggregation generates do not lead to welfare-enhancing additional innovation.

Fourth, Fortress?s hold-up power is amplified by the uncertainty it creates through the size of the portfolio it controls and obfuscation regarding the scope of that portfolio. After aggregation, potential licensees lose the ability to decipher the extent to which Fortress controls patents that they may actually have wanted to license ex ante or that would be substitutes to asserted patents. By way of example, Fortress employees are listed as managing members or directors of companies that otherwise have no publicly known ties to Fortress. Mysterious patterns emerge such as entities with names connoting an unspecified relationship with Fortress, by a prefix ?CF.? District court judges have gone so far as having to compel Defendants to reveal the ownership history of the asserted patents and the degree to which Fortress held rights in, and control over, those patents. The effect is that the hold-up power of asserted patents is imbued on other patents Fortress controls. Thus, rather than fostering pro-competitive patent licensing, Fortress?s aggregation scheme reduces potential licensees? ability to obtain licenses to any patents they might be interested in licensing while simultaneously elevating the value of weak patents.

There’s a lot more in the complaint. As you can hopefully see from the above quotes, it’s written in an extremely clear and readable fashion, unlike many lawsuits. It goes into details of all of the Fortress-connected defendants and the various shakedown attempts they’ve made, as well as some of Fortress’s “lending” practices that effectively force patent trolling on recipients.

There’s a lot in the 57-page complaint (even if parts are redacted). I have no idea what chance it actually has in court. Indeed, I’d guess that courts are unlikely to accept the idea that asserting a patent, or even the overall process of patent trolling in this manner is an antitrust violation, but at the very least, it makes a compelling case as to how government-granted monopolies can be used in an abusive manner that appears to violate antitrust laws, and that clearly go against the stated purpose of patents in the Constitution. This should be a case worth following.

Filed Under: anti-trust, monopolies, patent investment firms, patent trolls, patents
Companies: apple, fortress investment group, intel, softbank

Qualcomm Used Patent Monopolies To Shake Down The Entire Mobile Phone Industry For Decades

from the and-we-all-suffered dept

Just a few weeks ago, Qualcomm and Apple settled a massive patent dispute on the eve of a trial. In the run-up to the settlement, Apple had made a really compelling case that Qualcomm’s practices involve blatant abuse of its patents to jack up prices to insane levels and to limit any real competition. Just recently we wrote about how media-tracking giant Nielsen was abusing patents for anticompetitive purposes, but they looked like blatant amateurs compared to Qualcomm. As we noted in that post, our founding fathers worried quite a bit about the impact of patent monopolies and how they would stifle innovation and competition. James Madison said:

“But grants of this sort can be justified in very peculiar cases only, if at all; the danger being very great that the good resulting from the operation of the monopoly, will be overbalanced by the evil effect of the precedent; and it being not impossible that the monopoly itself, in its original operation, may produce more evil than good.”

And Thomas Jefferson, who grudgingly ran the patent system for a while (and where he tried to institute rules to prevent abuse) seemed to regret the entire concept of patents:

… generally speaking, other nations have thought that these monopolies produce more embarrassment than advantage to society; and it may be observed that the nations which refuse monopolies of invention, are as fruitful as England in new and useful devices.

But one quote regarding such monopolies that sticks most with me is that of UK Parliament member Thomas Macauley, who gave a famous speech in 1841, in which he declared:

Sir, that I may safely take it for granted that the effect of monopoly generally is to make articles scarce, to make them dear, and to make them bad…

He further noted that if there is a case that the only way to remunerate people is through such a monopoly they may grudgingly be granted, but warns of the evil this will enable if allowed to do more than incentivize the original creation:

It is good that authors should be remunerated; and the least exceptionable way of remunerating them is by a monopoly. Yet monopoly is an evil. For the sake of the good we must submit to the evil; but the evil ought not to last a day longer than is necessary for the purpose of securing the good.

I’m thinking about these quotes, yet again, in relation to Qualcomm. Just a month after settling the lawsuit with Apple, a federal judge in an ongoing antitrust lawsuit filed by the FTC, laid out in massive detail just how evil Qualcomm has been in using its monopoly powers to stifle innovation and competition. The 283-page ruling is incredibly damning — and if you don’t feel like reading the whole thing, Tim Lee over at Ars Technica has an only slightly shorter summary of just how Qualcomm’s shake down scam worked.

I read every word of Judge Koh’s book-length opinion, which portrays Qualcomm as a ruthless monopolist. The legal document outlines a nearly 20-year history of overcharging smartphone makers for cellular chips. Qualcomm structured its contracts with smartphone makers in ways that made it almost impossible for other chipmakers to challenge Qualcomm’s dominance. Customers who didn’t go along with Qualcomm’s one-sided terms were threatened with an abrupt and crippling loss of access to modem chips.

“Qualcomm has monopoly power over certain cell phone chips, and they use that monopoly power to charge people too much money,” says Charles Duan, a patent expert at the free-market R Street Institute. “Instead of just charging more for the chips themselves, they required people to buy a patent license and overcharged for the patent license.”

Qualcomm — which it should be noted has been one of the most aggressive agitators against patent reform (and sometimes in favor of making patent law much, much worse) — is basically the world’s largest patent troll, and it loved every minute of it.

Qualcomm’s patent licensing fees were calculated based on the value of the entire phone, not just the value of chips that embodied Qualcomm’s patented technology. This effectively meant that Qualcomm got a cut of every component of a smartphone?most of which had nothing to do with Qualcomm’s cellular patents.

“Qualcomm charges us more than everybody else put together,” Apple executive Jeff Williams said. “We’ve never seen such a significant licensing fee tied to any other IP we license,” said Motorola’s Todd Madderom.

And, of course, Qualcomm leveraged its position to hold everyone over a barrel:

These high royalties reflected an unusual negotiating tactic called “no license, no chips.” No one could buy Qualcomm’s cellular chips unless they first signed a license to Qualcomm’s patent portfolio. And the terms of these patent deals were heavily tilted in Qualcomm’s favor.

Once a phone maker had signed its first deal with Qualcomm, Qualcomm gained even more leverage. Qualcomm had the right to unilaterally terminate a smartphone maker’s chip supply once the patent licensing deal expired.

“If we are unable to source the modem, we are unable to ship the handset,” said Motorola executive Todd Madderom in a deposition. “It takes many months of engineering work to design a replacement solution, if there is even a viable one on the market that supports the need.”

That made Qualcomm’s customers extremely vulnerable as they neared the expiration of a patent licensing deal. If a customer tried to negotiate more favorable terms?to say nothing of formally challenging Qualcomm’s patent claims in court?Qualcomm could abruptly cut off the company’s chip supply.

And, of course, Qualcomm used this power to crush the competition and stamp out any kind of non-Qualcomm innovation.

Qualcomm’s first weapon against competitors: patent licensing terms requiring customers to pay a royalty on every phone sold?not just phones that contained Qualcomm’s wireless chips. This gave Qualcomm an inherent advantage in competition with other chipmakers. If another chipmaker tried to undercut Qualcomm’s chips on price, Qualcomm could easily afford to cut the price of its own chips, knowing that the customer would still be paying Qualcomm a hefty patent licensing fee on every phone.

Also notable: right after Apple and Qualcomm settled their legal dispute in April, Intel announced it was exiting the mobile 5G chip space, and admitted it was entirely because of that settlement:

?In light of the announcement of Apple and Qualcomm, we assessed the prospects for us to make money while delivering this technology for smartphones and concluded at the time that we just didn?t see a path,? [Intel CEO Bob] Swan said.

And the ruling a month later made clear, Qualcomm’s patent attack against Apple was very much driven by fear of competition from Intel:

Freed of Qualcomm’s chip supply threat, Apple began to challenge Qualcomm’s high patent royalty rates. Qualcomm responded by cutting Apple off from access to Qualcomm’s chips for new iPhone models, forcing Apple to rely entirely on Intel for the cellular chips in its 2018 models. Qualcomm sued Apple for patent infringement in courts around the world, while Apple pressed the Federal Trade Commission to investigate Qualcomm’s business practices.

And, of course, Qualcomm has long used its patents to directly stifle any competition as opposed to the indirect versions described above:

Chipmakers are ordinarily expected to acquire patents related to their chips and indemnify their customers for patent problems. But Qualcomm refused to license its patents to competitors, putting them in a difficult position.

“The prevailing message from all of the customers I engaged with was that they expected us to have a license agreement with Qualcomm before they would consider purchasing 3G chipsets from MediaTek,” said Finbarr Moynihan, an executive at chipmaker MediaTek.

If a chipmaker asked to license Qualcomm’s patents, Qualcomm would only offer a promise not to sue the chipmaker itself?not the chipmaker’s customers.

As Tim Lee highlights, Qualcomm was so comically evil in executing this plan, it literally laid out the details of how to starve MediaTek (MTK) in a PowerPoint presentation:

This is summarized in the judge’s opinion:

The slide includes the strategy ?make sure MTK can only go after customers with WCDMA SULA,? with an arrow leading to ?Reduce # of MTK?s 3G customers to ~50.?… The next strategy is ?Formulate and execute a GSM/GPRS strategy to destroy MTK?s 2G margin & profit,? with an arrow to ?Take away the that MTK can invest in 3G.? …. Thus, Qualcomm?s refusal to license MediaTek was designed to (and in fact did) limit MediaTek?s customer pool and reduce MediaTek?s revenue base to invest in future cellular generations.

There’s a LOT more in the ruling by Judge Lucy Koh and it basically lays out a road map of pure evil by Qualcomm, allowing it to skim a ton of (literal) monopoly rents off the entire mobile phone market, even when its own innovations were a minimal part of that market — while at the same time abusing its patent position to deliberately stifle competition and innovation. The court order says that Qualcomm needs to renegotiate its licenses, and can no longer use its “no license, no chip” deals that block customers from buying Qualcomm chips if they don’t have a license.

Of course, Qualcomm won’t go down without a fight, and has announced that it’s appealing the ruling. So this will go on for some time. But if you ever want a pure example of the “evil” created by excessive patent monopolies, Qualcomm is about as pure an example as you can find.

Filed Under: antitrust, extortion, licensing, lucy koh, mobile phones, monopolies, no license no chips, patents, shake down
Companies: apple, intel, mediatek, qualcomm

Qualcomm's Patent Nuclear War Turning Into Nuclear Winter

from the none-of-this-is-good dept

We haven’t written much about Qualcomm and Apple’s all out nuclear war over patents, but a few recent developments suggest it’s worth digging in and discussing. In some ways it sweeps in other companies (mainly Intel) and also involves the FTC and the ITC. I won’t go through the entire history here because I’d still be writing this post into next year. Qualcomm is a pretty massive company and while it does produce some actual stuff, it has long acted quite similar to a patent troll. It has also vigorously opposed basically all patent reform efforts, while at the same time quietly funding a bunch of “think tanks” that go after anyone advocating for patent reform (I expect some fun comments to show up below).

The reason Qualcomm acts this way is that it has long abused the patent system to jack up prices to ridiculous rates. And it’s finally facing something of a reckoning on that. In early 2017, the FTC went after Qualcomm for abusing its patents — notably: “using anticompetitive tactics to maintain its monopoly in the supply of a key semiconductor device used in cell phones and other consumer products.” Specifically, the FTC alleged that Qualcomm, despite promises to the contrary to get its patents into important standards, was not following the FRAND (Fair, Reasonable and Non-Discriminatory) licensing of its patents, as required to have its inventions be a part of the standard. Just days later, Apple sued Qualcomm, also regarding Qualcomm’s patent shakedown, claiming that Qualcomm had been massively overcharging Apple for the use of its patents, rather than licensing them on a FRAND basis.

A few months later, Qualcomm sued Apple, claiming that it had been sharing Qualcomm’s proprietary code with Intel. Apple had been using chips from both Qualcomm and Intel, but was in the process of dropping Qualcomm entirely. Qualcomm also launched multiple parallel proceedings at the ITC. As we’ve discussed for over a decade now, patent holders ridiculously get two shots at anyone they accuse of patent infringement (so long as the accused manufacturers its goods outside the US). The International Trade Commission (for reasons that make no sense) feels that it can judge on its own if patents have been infringed, and if so, it can block the further importation of the “infringing” good. That’s the only remedy at the ITC, but it can have quite an impact, obviously, in blocking a product out of the US market. Incredibly, the ITC need not follow the same rules as a regular court and it can do its own analysis while a case is in federal court (which might rule entirely differently).

So that’s the history. Basically, Apple and Qualcomm are in an all out patent nuclear war, with the FTC and ITC involved around the edges. In the last few weeks, however, pretty much everything has been looking pretty bad for Qualcomm. While an administrative law judge at the ITC did find some infringement, he (somewhat surprisingly) announced that he would not recommend an import ban (again, this is the only remedy the ITC can offer). The full ITC needs to review this recommendation and make a final call. Tons of patent maximalists are screaming their heads off about how the ITC must start blocking iPhones, but as Judge Thomas Pender recognized, banning an entire product because it may have infringed on a single patent is ridiculous. In the language of the judge “the statutory public interest factors weigh against issuing a limited exclusion order as to products found to infringe patents asserted in this investigation.” In other words, “grow up Qualcomm, this isn’t such a big deal that you get to completely ban the product.”

And, now, the latest is that the FTC’s case against Qualcomm went in the FTC’s direction, with Judge Lucy Koh granting the FTC’s motion for partial summary judgment and saying that Qualcomm was violating its FRAND promises. This isn’t everything to do with the case, but does involve questions around whether or not Qualcomm can limit its licensing to just device makers, or if it also has to license its patents to other chipmakers, like Intel. And Koh points out that basically everyone recognizes that the FRAND agreement it made applies to everyone — not just a limited subset of companies. Koh repeatedly highlights Qualcomm’s own previous statements that support this.

Furthermore, Koh points out that allowing Qualcomm to discriminate against chipmakers would hand the company a total monopoly, and that clearly goes against the concept behind the FRAND agreement to put the technology into the standard:

If a SEP holder could discriminate against modem chip suppliers, a SEP holder could embed its technology into a cellular standard and then prevent other modem chip suppliers from selling modem chips to cellular handset producers. See Lemley, Intellectual Property Rights, 90 Calif. L. Rev. at 1902 (stating that a company with a SEP ?will effectively control the standard; its patent gives it the right to enjoin anyone else from using the standard?). Such discrimination would enable the SEP holder to achieve a monopoly in the modem chip market and limit competing implementations of those components, which directly contradicts the TIA IPR policy?s stated purpose to ?enable competing implementations that benefit manufacturers and ultimately consumers.? TIA IPR at 6. See Borg v. Transamerica Ins. Co., 47 Cal. App. 4th 448, 456 (1996) (holding that a court may not interpret a contract in a way that contradicts the contract?s plain meaning). Qualcomm never attempts to explain how discrimination against modem chip suppliers is consistent with the stated purposes of the IPR policies.

I know there’s a lot of jargon in there, but it’s basically saying that the whole point of the standardization process, as everyone agreed, was to create standardization across multiple competitors (while still allowing a reasonable license for patent holders). But if Qualcomm can reinterpret this agreement to say that FRAND only applies to downstream users, then that completely overturns the entire intention of the standards making process and just gives Qualcomm a total monopoly on the chips (going well beyond its patents). Qualcomm will undoubtedly appeal, but it’s not a good start for the company.

Oh, and speaking of not a good start, the post-Apple world for Qualcomm isn’t looking great either. Its latest earnings projections going forward were below what Wall St. was expecting and some are noticing a $5-billion-ish Apple-shaped hole in the books.

While I’ve been equally critical of Apple in the past when it’s abused the patent system, in this situation it seems pretty clear that Qualcomm completely overplayed its hand with its patents in ways that were abusive, and that drove up costs in an unfair manner, against its own agreements. And so far, the various courts and administrative bodies are not buying into Qualcomm’s desperate attempts to keep up its monopolizing.

It all goes back to the point we’ve been making for decades: if you have a good product, compete in the marketplace. Don’t abuse the patent system to try to block competitors or to artificially inflate the price. That just telegraphs that you’re bad at innovating and you know that competitors can do a better job than you. Qualcomm is now learning how that plays out in the long run.

Filed Under: antitrust, ftc, itc, monopoly, patents, phones, sep, standards, standards essential patents
Companies: apple, intel, qualcomm

NSA Denies Prior Knowledge Of Meltdown, Spectre Exploits; Claims It Would 'Never' Harm Companies By Withholding Vulns

from the lol-ok-then dept

News surfaced late last week indicating everything about computing is fucked. Two critical flaws with zero perfect fixes — affecting millions of processors — were exposed by security researchers. Patches have been deployed and more are on their way, but even the best fixes seem to guarantee a noticeable slowdown in processing speed.

The government has stepped up to say that, for once, it’s not involved in making computing less safe.

Current and former U.S. officials… said the NSA did not know about or use Meltdown or Spectre to enable electronic surveillance on targets overseas. The agency often uses computer flaws to break into targeted machines, but it also has a mandate to warn companies about particularly dangerous or widespread flaws so that they can be fixed.

Rob Joyce, White House cybersecurity coordinator, said, “NSA did not know about the flaw, has not exploited it and certainly the U.S. government would never put a major company like Intel in a position of risk like this to try to hold open a vulnerability.”

The veracity of this statement is largely dependent on the credibility attributed to the person making it. While it is conceivable the NSA did not know about the flaw (leading to it being unable to exploit it), it’s laughable to assert the NSA wouldn’t “put a major company in a position of risk” by withholding details on an exploit. We only have the entire history of the NSA’s use of exploits/vulnerabilities and its hesitant compliance with the Vulnerability Equities Process to serve as a counterargument.

The NSA has left major companies in vulnerable positions, often for years — something exposed in the very recent past when an employee/contractor left the NSA in a vulnerable position by leaving TAO tools out in the open. The Shadow Brokers have been flogging NSA exploits for months and recent worldwide malware/ransomware attacks are tied to exploits the agency never informed major players like Microsoft about until the code was already out in the open.

These recently-discovered exploits may be the ones that got away — ones the NSA never uncovered and never used. But this statement portrays the NSA as an honest broker, which it isn’t. If the NSA had access to these exploits, it most certainly would have used them before informing affected companies. That’s just how this works. As long as exploits are returning intel otherwise inaccessible, the NSA will use the exploits for as long as possible before disclosing this info to US companies. The agency has historically shown little concern about collateral damage and I don’t believe putting someone new in charge of the VEP is going to make that much of a difference in the future.

Filed Under: meltdown, nsa, rob joyce, spectre, vep, vulnerabilities, vulnerabilities equities process
Companies: intel