In this week’s roundup of the latest news in online speech, content moderation and internet regulation, Mike is joined by guest host Kat Duffy, Senior Fellow for Digital and Cyberspace Policy at the Council on Foreign Relations and CEO of RightsDuff Strategies. They cover:
The 53-page report, titled “Delusion of Collusion: How the House Republican Majority Abused Oversight Powers to Protect Elon Musk and Silence His Critics,” exhaustively documents how Jordan launched a sham investigation in what appears to be a clear attempt to intimidate advertisers and bully them into subsidizing Musk’s ExTwitter, while falsely claiming it was about fighting “collusion.”
Because the Democrats tend to be inept and incompetent in explaining reality to people, Rep. Jerry Nadler released the report on New Year’s Eve where it basically got zero attention. As far as I can tell, the only news report to cover it was a small legal antitrust trade publication. By the time the ball dropped in Times Square hours after the report had been released, it had effectively disappeared.
However, it deserves way more attention for all of the nonsense it puts into the public record, specifically focusing on Jordan and Musk’s effort to attack GARM, a small non-profit that just worked with advertisers and social media platforms to encourage the platforms to protect the brand safety of advertisers. As we’ve covered, that attack was successful. Even though his ExTwitter had put out a press release talking about how excited they were to “rejoin” GARM just weeks earlier, Musk went on to sue GARM, which was almost immediately shut down by the World Federation of Advertisers.
The report breaks down how this was a clear case of Jordan and Musk weaponizing the government to silence critical speech.
By March 2023, Twitter’s value had fallen from $44 billion to $20 billion. The reason for this decline in value is no mystery, given the facts outlined above. Nevertheless, the Majority launched an investigation into the advertisers which have declined to spend money on the platform, accusing them of “colluding” to hurt the company’s profits. Since then, the Majority has spent countless dollars and hours of staff time trying to figure out why advertisers might be hesitant to risk their brands’ reputations on a platform whose owner told them, in November 2023, to “Go fuck yourself.”
Chairman Jordan’s so-called investigation culminated in a July 2024 “interim report” which used cherry picked documents and misleading transcript excerpts to suggest that the committee had uncovered evidence of “collusion” when in fact the very opposite is true. In fact, the complete and contextualized documents and testimony show that the Global Alliance for Responsible Media and its member companies were engaged in a pro-competitive effort to address the substantial brand risk that harmful online content poses to advertisers and to consumers.
Chairman Jordan’s report had an audience of one: Elon Musk. In fact, the entire report seems like pretext for a lawsuit Musk filed against various advertising entities and ultimately to silence the advertisers who expressed concern about content on his platform. The resources of this Committee should not be directed to further pad a billionaire’s bottom line.In contrast, this minority report is intended for the American public, who are entitled to the truth about this investigation and about Chairman Jordan’s true aims and abuse of congressional oversight power.
It’s hard to imagine a more blatant example of a powerful government official abusing his authority to carry water for a political ally and major GOP donor. The fact that Jordan is doing this while sanctimoniously claiming to be fighting the “weaponization” of government is beyond parody.
As the report calls out:
For the past 20 months, the Chairman of the House Judiciary Committee has abused his oversight power and the rule of law to push an agenda that would pervert the free market and undermine individual companies’ independent decisions as to where to place advertisements online. The spread of illegal, harmful, abusive, and false and misleading content online results in actual harm, both online and offline. We are left to conclude that the Majority’s ultimate goal was not to conduct antitrust oversight as they claim, but rather to silence criticism of harmful online content and those who promote it, deter content moderation, and protect the ability to use mis- and disinformation campaigns to achieve political ends.
Ya think? This was obvious from the beginning, but almost entirely ignored by the credulous media that uncritically amplified Jordan’s false claims.
The report thoroughly debunks Jordan’s flimsy antitrust pretext and exposes his true aim: strong-arming companies into boosting Musk and his political allies.
It also calls out the irony of the committee that claims to be fighting weaponization, actually being the chief party weaponizing the government against speech:
The Majority is engaging in a transparently political effort to use the antitrust laws to benefit their allies by conferring upon them outcomes that they could not otherwise achieve in the marketplace. This is not just a misuse of the antitrust laws, but fundamentally subverts the goals of those laws. The irony could not be greater. While spending most of this Congress attacking the Biden administration’s so-called weaponization of government, the Majority here is trying to weaponize the antitrust laws under a highly dubious theory to override legitimate market outcomes.
It also calls out the MAGA trend of falsely claiming that content moderation or boycotts could possibly violate the First Amendment:
Finally, the Majority bandies about words like censorship, in a misguided effort to evoke the First Amendment. But as the Majority well knows, the First Amendment only applies to government action. And in this case, the only governmental burdening of speech is the Majority’s onslaught against GARM and its members. It is an effort to bully the advertisers into subsidizing firms whose content moderation policies put brands and businesses at risk. It is an attempt to hijack free speech, as well as antitrust, for political purposes.
If reality mattered, this report would be a bombshell. But, again, everyone seems to be living in a fog of nonsense, where anything the MAGA world says it’s doing, no matter how obviously false, is treated as genuine. And any time anyone calls out the lack of clothes on the emperor, it’s dismissed as sour grapes or “derangement syndrome.”
The report is thorough and detailed. It explains why companies might not want to advertise on ExTwitter for totally legitimate business reasons, calling out examples of big brands having their ads show up next to “pro-Nazi” content, and noting that consumers (the marketplace again!) will often punish companies whose advertisements support such hatred:
Now consider the category of misinformation that the Majority alleges GARM’s members misapply to the detriment of conservative-voiced content. The GARM framework defines misinformation as “the presence of verifiably false or willfully misleading content that is directly connected to user or societal harm.” Consumer surveys suggested that inappropriate content, including misinformation, negatively affects brand trust and purchase behavior. These results explained, in part, why advertisers are concerned about the nexus between brand safety and misinformation. Additional studies examined this nexus in more detail. A 2024 article in NATURE reported the results of an experiment which demonstrated that consumers are likely to reduce purchases from firms that advertise on websites that publish misinformation compared to firms that do not. Unlike the surveys which measured intention to change purchase behavior, subjects in this experiment made actual economic choices. Additional research on consumer reaction to misinformation was provided by the IPG Mediabrands and Zefr MAGNA Media Trials Study which found that “advertising next to misinformation led to wasted dollars for brands, eroded brand perception, and negatively impacted KPIs [key performance indicators].”
The challenges of directing ad placement to trustworthy sites and away from misinformation sites continues to loom large. The 2024 NATURE study found that of the 100 most active advertisers, an astounding 79.8 percent that used digital advertising platforms had advertisements placed in online misinformation outlets in a given week. The authors attributed the problem to the use of such platform systems that allocate advertising to such websites. Another study, by the Pew Research Center, suggested that “for every $2.16 in digital ad revenue sent to legitimate newspapers, U.S. advertisers are sending $1 to misinformation websites.”
In sum, online advertising is very important for advertisers and for the websites that provide and host content, many of whose business models depend on it. But harmful content is challenging the business models of advertisers, content providers, and platforms alike. Consumers associate the online content with the brands that advertise there. When a brand is advertised near harmful content, its value is undermined because most consumers believe that the brand knowingly chose that content and site for its advertising.
In other words, there are completely and totally understandable business reasons for advertisers to stop advertising on ExTwitter.
And all GARM was trying to do was help advertisers make sure that they didn’t risk angering customers by having ads appear next to highly controversial content. And they did so in a way that everyone involved knew was just creating more information and allowing advertisers (and social media platforms) to make their own final decisions:
GARM’s voluntary frameworks, which the biggest social media platforms helped develop, provide structures for analysis and created a common lexicon. Much like the terms of art in marketing or expressions in mathematics, a shared terminology facilitates communication that is foundational for constructive working relationships across organizations. Such terminology enhances transparency, making market transactions more efficient. The buyer better understands what sellers are offering in terms of brand safety and the seller better understands what buyers want. Both advertisers and platforms benefit from this common approach and independent decision making is improved.
Crucially, the frameworks do not dictate advertising outcomes. Applying those frameworks is an inherently subjective exercise that includes tailoring to the specific requirements of the brands and leads to outcomes that vary across GARM’s members. Juhl described how GroupM customizes its work in ad placement to reflect the specific needs of their advertiser clients:
GroupM works to place our clients’ ads on media pursuant to their goals, preferences, and target audiences, and we continually engage with our clients to understand their particular risk tolerance levels. These risk tolerances shift due to our clients’ own business conditions and how they view the current political and social environments. Clients shift priorities very quickly and it is our job to execute their strategy with speed and precision. We always follow our client brand’s ad placement wishes.
It is also important to recognize that the application of the GARM frameworks usually operates within a firm’s set of marketing policies and hence was only one consideration among many. These marketing policies vary by firm. Most were created before the GARM frameworks and continue to shape online advertising choices.
But, Jim Jordan and Elon Musk bent over backwards to pretend that it was “illegal collusion” that violates antitrust law. And this report says that’s ridiculous to anyone who looked at all the facts.
The Majority’s July 2024 Interim Report offers no direct evidence of an agreement among GARM and its members. Mere status as a member of GARM would not, without more, support a finding of a conspiracy. Consistent with the key Supreme Court precedents Matsushita Elec. Industrial Co. v. Zenith Radio Corp.1 and Monsanto Co. v. Spray-Rite Service Corp., a plaintiff would have to “present evidence tending to show that association members, in their individual capacities, consciously committed themselves to a common scheme designed to achieve an unlawful objective.” In contrast, GARM and its members are absolutely clear that their advertising decisions are made independently. As Unilever USA President Patel testified during the hearing,
I want to be very clear on one crucially-important fact. Unilever and Unilever alone controls our advertising spending. No platform has the right to our advertising dollars. As we look across the available advertising inventory, recognizing we do not have unlimited money to spend on advertising, we choose the channels, the platforms, and the outlets that give us the greatest commercial benefit for our advertising investments.
During questioning Patel further confirmed that, “A hundred percent, Unilever makes its own decisions,” and does not follow any outside group’s direction to avoid any outlet. This sentiment is echoed by GARM’s Rakowitz during his transcribed interview:
Q: But just to nail down that point, GARM doesn’t tell individual members—
A: Absolutely not.
Q: —what to do?
A: No, we do not.
Q: Or where to place ads?
A: No, we do not.
Q: Or where to avoid placing ads?
A: We do not.
These comments are consistent with the advertiser decision making process discussed in Part IIB.
As the report highlights, nothing about this represents a serious antitrust inquiry.
A serious antitrust inquiry would need to address the ease of reaching and sustaining an agreement. Two major obstacles—large numbers of participants and participants with diverse interests—have long been recognized by antitrust law as making collusive schemes less likely. In the GARM setting, overcoming these obstacles would loom large.
The real reason companies stopped advertising on ExTwitter is no grand conspiracy to suppress free speech. It was a simple business calculation. Advertising there is bad for business:
The Majority focused on alleged harm caused by the demonetization of its favored conservative-voices. They assert that this loss of revenue is caused by a large conspiracy involving GARM and its 100 plus members to suppress conservative-voiced online platforms and outlets by stopping advertising support. But the most compelling explanation for this revenue decline is apolitical. Advertisers want to attract and retain customers. When their advertising is placed next to harmful content the advertisement instead repels customers. Not surprisingly, advertisers gravitate to outlets that pose less risk to their brands. Again, this isn’t rocket science.
Instead, the much more obvious conclusion is the one that we’ve been shouting from the rooftops for the past few years: that it’s Jordan who is weaponizing the government to silence speech:
As with other of this Committee’s recent investigations, we are left to conclude that its ultimate goal was not to “conduct[] oversight of the adequacy and enforcement of U.S. antitrust laws” as they claim, but rather to silence criticism of harmful online content and those who promote it, deter content moderation, and protect the ability to use mis- and disinformation campaigns to achieve political ends. The Majority’s desperate ploy to launder their failed censorship arguments through an antitrust framing itself fails. The Majority’s actions have intimidated organizations who call attention to the prevalence of hate, disinformation, and other harmful or unlawful content online. Fostering a more transparent, accountable, and responsible digital environment is not only lawful, it is good for businesses, consumers and the general public. Chairman Jordan’s investigation and others like it will undermine this work and lead to the further deterioration of our information ecosystem and will threaten free speech.
Antitrust is not about choosing winners and losers. It is about ensuring a fair fight. In this instance we see that the Majority is willing to condemn any outcome that they do not like as being unfair and the outcome appears to involve both a category of supposed victims as well as a particular victim—X. In fact, this investigation originated after the Speaker of the House Kevin McCarthy, Chairman Jordan, and Elon Musk were talking and Musk said, “‘by the way, there’s this organization GARM, because GARM is harm.’ [sic] I [Jordan] never forgot that sentence.” No he did not. Jordan embarked on an investigation whose outcome was a foregone conclusion and for which the resulting report’s title [GARM’s Harm] was effectively supplied by Musk himself. Despite all of the investigation’s shortcomings, it excelled in one regard—providing taxpayer funded discovery for the richest man in the world and one of Trump’s biggest donors. A lawsuit launched by X just days after the Majority’s interim report was released began by touting that the conduct was “the subject of an active investigation” by the House Judiciary Committee before reproducing the fruits of the subcommittee’s fishing expedition in the form of a document demand. Perhaps this assault on legitimate business activity seems worth it to the Majority.
It’s pretty scathing as Congressional reports go.
In the end, this sordid saga illustrates the dangerous way that accusations of “censorship” and “collusion” are being cynically weaponized to bully companies into amplifying favored political content. Jordan and Musk’s campaign against GARM sets a troubling precedent.
By abusing the power of Congressional oversight to intimidate advertisers and platforms, they are effectively arguing that companies have an obligation to subsidize and support any speech, no matter how hateful or harmful, or else be accused of “censorship.” It’s an attempt to pervert the free market to serve their political agenda.
But as this report makes clear, advertisers’ decisions on where to place their ads are driven by legitimate business considerations about brand safety and consumer sentiment, not some nefarious plot to silence conservatives. The real threat to free speech is not content moderation or advertiser boycotts – it’s government officials like Jim Jordan trying to use their power to dictate what speech must be subsidized and supported.
Sadly, given the current media and political environment, it’s unlikely this report will get the attention it deserves. But for anyone who cares about the future of online speech, platform governance, and the abuse of government authority, it’s essential reading. It shines a harsh light on Jordan and Musk’s cynical, dishonest campaign and the damage it has done to free speech and the free market.
Elon Musk’s ExTwitter just set a new speed record: from enthusiastic joiner of an advertising coalition to potential plaintiff against the same organization in just over a week.
Sometimes, timing is everything.
This week has been a travel week for me, so on Tuesday evening, I wrote up a short article on last week’s news that ExTwitter had “rejoined GARM.” GARM is the Global Alliance for Responsible Media, which is a loose coalition focused on brand safety for advertisers, such that their ads are less likely to appear next to, say, neo-Nazi content.
The main focus of my post was that there was almost no way that anyone should believe that ExTwitter’s decision to rejoin GARM was a sincere statement that ExTwitter would now take brand safety and GARM’s recommendations seriously. Instead, I noted that whenever ExTwitter was desperate for advertisers to sign on, its advertising execs (Linda Yaccarino’s underlings) would tout its compliance with GARM guidelines. But then Elon would do something fucking crazy and drive away advertisers again.
I even predicted, “sooner or later (probably sooner) Elon will do something horrible…” I should have known that it would happen so soon that it was before I could even post my article.
Anyway, I wrote that Tuesday evening and scheduled it to go up on Techdirt on Thursday afternoon, since I’d be traveling and without internet access for large segments of time this week.
Little did I know that on Wednesday, before my post went up, Jim Jordan and the House Judiciary would release an astoundingly stupid “report” claiming that GARM was an antitrust-violating cartel that was pressuring websites into censoring conservatives.
And, in response to a tweet showing just a clip of some nonsense testimony at the House hearing about this report, Elon Musk announced on Thursday morning (before my post went up) that ExTwitter “has no choice but to file suit against the perpetrators and collaborators” (meaning GARM, its organizers, and its members) and also said that “hopefully, some states will consider criminal prosecution.”
Yes, that’s Elon Musk saying that he plans to file a civil lawsuit against GARM and its “collaborators” and hopes that state AGs will file criminal lawsuits against the very organization HIS COMPANY REJOINED JUST A WEEK EARLIER and celebrated with a hyped-up tweet:
So, last week ExTwitter was “excited to announce” that it’s rejoined GARM, and this week Elon says that GARM’s leaders should be criminally prosecuted, and he planned to sue them himself.
Cool, cool.
I can just imagine how Linda Yaccarino must feel about this. She clearly orchestrated the return to GARM as part of her desperate push to lure back advertisers.
But let’s be clear about this. Companies have their own First Amendment rights not to associate with anyone they want. And that includes not advertising on websites where your ads might show up next to controversial content, disinformation, or just general nonsense. Many companies recognize that it is bad for business to have advertisements showing up next to neo-Nazi content, or just plain old disinformation.
Private companies choosing not to advertise is not a violation of any law, civil or criminal. Private organizations setting up guidelines for brand safety is not an antitrust violation. Private organizations choosing not to advertise on the site formerly known as Twitter is an expression of their own First Amendment rights not to associate with whatever nonsense Elon is promoting these days.
Anyway, all that effort that Yaccarino put into “rejoining GARM” last week just went up in smoke. She was trying to convince advertisers that ExTwitter was a safe place for brand advertising, but now Elon is saying ExTwitter will be suing GARM and pushing for criminal prosecutions of everyone involved in GARM.
Which now includes Elon Musk’s ExTwitter as of last week. Can’t wait to see Elon sue himself.
What a clusterfuck of stupidity.
And I’m sure that it won’t be long before an Andrew Bailey of Missouri or a Ken Paxton of Texas opens an “investigation” into GARM (the group that Elon Musk’s company “excitedly” rejoined just last week).
Hilariously, this would be an actual First Amendment violation, in that it would be a government agency starting a criminal investigation for the pretty clear express purpose of intimidating companies out of expressing themselves.
Remember when Elon said he was against governments pressuring companies about their speech? Now he’s telling them to do that, but just to organizations he doesn’t like (even though his own company just joined the very same organization).
So, just to recap: last week, Elon’s company rejoined GARM, the advertising coalition to help make sure platforms are a safe place for brand advertisers to advertise. This week, the House Judiciary Committee falsely claimed that the First Amendment-protected rights of companies not to advertise on ExTwitter was an antitrust violation, leading to “First Amendment absolutist” Elon Musk saying he’s going to sue the very organization his company just “excitedly” joined. And, to top it all off, Elon hopes that states will open criminal investigations into this activity — an act that would actually violate the First Amendment rights of GARM and those involved with it. Which includes Elon Musk’s own company.
I should have stayed off the internet even longer.
The U.S. House Judiciary Committee held a hearing last week to discuss the spread of white nationalism, online and offline. The hearing tackled hard questions about how online platforms respond to extremism online and what role, if any, lawmakers should play. The desire for more aggressive moderation policies in the face of horrifying crimes is understandable, particularly in the wake of the recent massacre in New Zealand. But unfortunately, looking to Silicon Valley to be the speech police may do more harm than good.
When considering measures to discourage or filter out unwanted activity, platforms must consider how those mechanisms might be abused by bad actors. Similarly, when Congress considers regulating speech on online platforms, it must consider both the First Amendment implications and how its regulations might unintentionally encourage platforms to silence innocent people.
But there?s a lot platforms can do right now, starting with more transparency and visibility into platforms? moderation policies. Platforms ought to tell the public what types of unwanted content they are attempting to screen, how they do that screening, and what safeguards are in place to make sure that innocent people?especially those trying to document or respond to violence?aren?t also censored. Rep. Pramila Jayapal urged the witnesses from Google and Facebook to share not just better reports of content removals, but also internal policies and training materials for moderators.
Better transparency is not only crucial for helping to minimize the number of people silenced unintentionally; it?s also essential for those working to study and fight hate groups. As the Anti-Defamation League?s Eileen Hershenov noted:
To the tech companies, I would say that there is no definition of methodologies and measures and the impact. [?] We don?t have enough information and they don?t share the data [we need] to go against this radicalization and to counter it.
Along with the American Civil Liberties Union, the Center for Democracy and Technology, and several other organizations and experts, EFF endorses the Santa Clara Principles, a simple set of guidelines to help align platform moderation practices to human rights and civil liberties principles. The Principles ask platforms
to be honest with the public about how many posts and accounts they remove,
to give notice to users who?ve had something removed about what was removed, and under what rule, and
to give those users a meaningful opportunity to appeal the decision.
Hershenov also cautioned lawmakers about the dangers of heavy-handed platform moderation, pointing out that social media offers a useful view for civil society and the public into how and where hate groups organize: ?We do have to be careful about whether in taking stuff off of the web where we can find it, we push things underground where neither law enforcement nor civil society can prevent and deradicalize.?
Before they try to pass laws to remove hate speech from the Internet, members of Congress should tread carefully. Such laws risk pushing platforms toward a more highly filtered Internet, silencing far more people than was intended. As Supreme Court Justice Anthony Kennedy wrote in Matel v. Tam (PDF) in 2017, ?A law that can be directed against speech found offensive to some portion of the public can be turned against minority and dissenting views to the detriment of all.?
Back in July we were flabbergasted to see a stunningly misleading and dishonest video put out by the the House Judiciary Committee trying to claim that FOSTA had been a huge success in stopping sex trafficking. There is literally no evidence to suggest this, while there’s plenty of evidence to show the harm that has been created by FOSTA. One of the claims in the video came from Rep. Ann Wagner, who was the original sponsor of FOSTA and has been a leading voice in stoking the exaggerated and misleading moral panic around sex trafficking (which is a real problem, but very, very limited compared to what many — including Wagner — have said about it). Wagner’s latest trick has been to try to massively expand the PATRIOT Act for spying on Americans by again freaking everyone out about sex trafficking.
As we noted back in July, in the video, Wagner tries to imply that FOSTA helped kill off 90% of sex trafficking. She worded it awkwardly so that it clearly implies 90% of sex trafficking went away due to FOSTA, but it could also be read to just say that 90% of sex trafficking ads went away. As we pointed out at the time, this was clearly not true either way. While Backpage contained many ads, it stopped with those ads a year and a half before FOSTA was law, and was taken down by the feds before FOSTA was signed. So there was literally no way that FOSTA could be in any way credited for a drop in ads coming from Backpage.
I missed it, but a few weeks later, the Washington Post set its fact checker on these specific claims, and did an even more thorough analysis, even asking Wagner’s office for details. And those details make Wagner look even worse, leading the Washington Post to give her the full three Pinocchios in their final ruling on the accuracy of her claim. Specifically, Wagner’s office argued that a DARPA analysis saw a “weekly global ad volume dropped 87 percent from January to April.” But, as the WaPo article notes (and as we did as well) the vast majority of that was from the takedown of Backpage, which was not due to FOSTA.
But, from there things get even worse. The Washington Post asked DARPA for what happened after April and found… things are not at all what the House Judiciary Committee and Wagner were claiming. Indeed, while there was an initial decline due to Backpage shutting down (again, not due to FOSTA), it quickly went back up after April — conveniently ignored by Wagner and the HJC. Why contaminate the narrative with facts:
Worldwide ads had a daily average of about 105,000 when FOSTA-SESTA passed on March 21 and had dropped 28 percent by the time Backpage was closed on April 5. It then plunged another 75 percent and reached a low of 19,456 on April 17, for a total decline of about 82 percent.
But on the day the Judiciary Committee posted the video, sex-trade ads were back at about 50 percent of the daily volume before the law had passed; as of Aug. 11, they were at almost 75 percent….
?The volume of ads dropped dramatically after the shutdown of Backpage but has been climbing since,? said Chris Dickson, director of research engineering at Uncharted. ?There is now a volume approaching what we observed before.?
So, once again it appears that Wagner and the HJC like to completely make up stories for grandstanding political purposes. They over-hyped the size of sex trafficking to pass this bad law, and then they massively over-hyped the impact of the law, ignoring (1) that the impact they took credit for had nothing to do with the law, and (2) ignoring that the data didn’t actually support what they claimed.
So why the hell is anyway still listening to Rep. Ann Wagner and her use of “sex trafficking” to try to pass new laws?
So, yesterday the House Judiciary Committee did what the House Judiciary Committee seems to do best: hold a stupid, nonsensical, nearly fact-free “hearing” that serves as nothing more than an opportunity for elected members of Congress to demonstrate their ignorance of an important topic, while attempting to play to their base. This time, the topic was on the content filtering practices of Facebook, Twitter and Google. Back in May there was actually a whole one day conference in Washington DC on this topic. The Judiciary Committee would have been a lot better served attending that than holding this hearing. I’d recommend not wasting three hours of your life watching this thing, but if you must:
The shortest summary would be that some Republican members of Congress think that these websites censor too much conservative speech, and some Democratic members of Congress think that they don’t censor enough other speech (including hoaxes and conspiracy theories)… and almost no one wants to admit that this is not even remotely an issue that Congress should be concerned about. There’s a narrative that has been picked up by many that insist that social media platforms are unfairly censoring “conservatives.” There is basically zero evidence to support this. Indeed, a thorough analysis of the data back in March by Nieman Labs and Newswhip found that conservative-leaning sites get much, much, much more engagement on Facebook than liberal-leaning sites.
But, never let facts get in the way of a narrative. Since that seems to be the way many hyperpartisan sites (at either end of the spectrum) deal with these things, Congress is helping out. The only bit of sanity, perhaps bizarrely, came from Rep. Ted Lieu, who reminded everyone of the importance of free markets, free speech and the fact that private platforms get to decide how they manage their own services. Considering that Republicans often like to claim the mantle of being the “small, limited government” party who wants the government’s hands out of business regulation, the fact that most of the hearing involved Republicans screaming for regulating internet platforms and a Democrat reminding everyone about the importance of a free market, capitalism and free speech, it really was quite a hearing. Lieu’s remarks were some of the rare moments of sanity during the hearing — including defending Facebook leaving Alex Jones’ conspiracy theories on its site. Let’s start with that high point before we dive into the awfulness. His comments come at about 2 hours and 10 minutes into the video:
… we’re having this ridiculous hearing on the content of speech of private sector companies. It’s stupid because there’s this thing called the First Amendment. We can’t regulate content! The only thing worse than an Alex Jones video is the government trying to tell Google… to prevent people from watching the Alex Jones video. We can’t even do it if we tried. We can’t even do any legislation out of this committee. And we’re having this ridiculous second installment hearing after the first hearing about Diamond and Silk not getting enough likes on Facebook.
He then went on to ask questions “so the American public understands what a dumb hearing this is.” And those questions — again — seemed like the kinds more expected from supposedly “free market” conservatives. Specifically he asked the companies if they were private companies aiming to maximize profits for shareholders. And he wasn’t doing that to show that companies were evil, he was doing that to show that that’s how the free market works. He followed up with this:
I noticed all of you talked about your own internal rules. Because that’s what this should be about. You all get to come up with your own rules. But not because government tells you what to do. Or because government says you have to rule this way or that way. And the whole notion that somehow we should be interfering with these platforms from a legislative, governmental point of view is an anathema to the First Amendment. And really it’s about the marketplace of ideas.
Kudos to Rep. Lieu. This is the kind of speech that you’d normally expect to hear from a “small government” conservative who talks about respecting the Constitution. But, in this case, it’s a Democrat. And it’s shameful that others (on both sides of the aisle) weren’t making the same point. Instead, there was a ton of pure nonsense spewed from the Republicans at the hearing. It’s hard to fathom that the following statements were made by people we’ve actually elected to our legislative body. There were so many dumb statements made that it’s difficult to pick out just a few.
Let’s start with Rep. Steve King, who has made quite a name for himself saying and repeating bigoted nonsense. Starting at about an hour and five minutes in the video, King seemed particularly concerned about traffic to Gateway Pundit, a site famous for trafficking in utter nonsense.
It’s a matter of Congressional record that Gateway Pundit, Mr. Jim Hoft, has introduced information into the record that in the span of time between 2016 and 2018, he saw his Facebook traffic cut by 54%. Could you render an explanation to that?
Um… what? How the hell is it of any concern to Congress whatsoever the traffic a single site gets? And, as we were just discussing recently, traffic to lots of news sites from Facebook has dropped massively as Facebook has de-prioritized news. In that post, we pointed out that Slate was self-reporting a drop in Facebook traffic over that same period of time of 87%. Based on that, why isn’t King asking about Slate’s traffic dropping? Perhaps because Gateway Pundit publishes the kind of nonsense King supports and Slate points out that King is a bigot?
And… isn’t that, again, kind of the point of the First Amendment? To protect news sites from having Congress play favorites?
Incredibly, King then concludes his time by first claiming he’s all for free speech and free enterprise, but wonders about turning social media sites into regulated utilities.
I’m all for freedom of speech and free enterprise and for competition and finding a way that we can have competition itself that does its own regulation, so government doesn’t have to, but if this gets further out of hand, it appears to me that Section 230 needs to be reviewed, and one of the discussions that I’m hearing is ‘what about converting the large behemoth organizations that we’re talking about here into public utilities.’
Are we living in an upside down world? A Democrat is praising the free market, profits and free speech, and a Republican is advocating for limiting free speech and in favor of turning some of the most successful US companies into public utilities? What is even going on here?
Around an hour and 18 minutes, we get our old friend Rep. Louis Gohmert, who has a fairly long and extensive history of making the dumbest statements possible concerning technology issues. And he lived down to his usual reputation in this hearing as well. It starts off by him trying to play down the issue of Russian interference in elections, by claiming (?!?) that the Russians helped Truman get elected, and then claiming that Russians had helped basically every Democratic President get elected in the past 70 years. And then spent a long time trying to complain that the platforms wouldn’t tell him if Chinese or North Korean intelligence services had also used their platforms. Remember, these companies were asked to come and testify specifically about Russian use of their platforms to interfere with the election and Gohmert stepped in with this insane “what about other countries, huh?” argument:
Gohmert: I need to ask each of you. You’ve been asked specifically about Russian use of your platforms. But did you ever find any indication of use of your platform, utilized by the Chinese, North Korea, or any other foreign country intelligence or agency of that country. First, Ms. Bickert?
Bickert/Facebook: I would note, Congressman, that we’re not in North Korea or China. In terms of whether we’ve seen attacks on our services, we do have — we are, of course, a big target — we do have a robust security team that works…
Gohmert: Well, but that’s not my question. It’s just a very direct question. Have you found… You don’t have to be in North Korea to be North Korean Intelligence and use… We have foreign government intelligence agencies IN THIS COUNTRY. So have… It seems to me you were each a little bit vague about “oh yes, we found hundreds” or whatever. I’m asking specifically, were any of those other countries besides Russia that were using your platform inappropriately? It should be a yes or no.
Actually, no, it shouldn’t be a yes or no. That’s a dumb and misleading question for a whole long list of reasons. Of course, lots of other intelligence agencies are using Facebook, because of course they are. But, the entire point of this line of questioning seems to be Gohmert trying to play down Russian use of the platform, which is… odd. Especially after he started out by praising the fact that maybe the Russians might help “our side” get elected going forward.
Bickert: I don’t have the details. I know we work to detect and repel attacks…
Gohmert: I know that. But were any of them foreign entities other than Russia?
Bickert: I can certainly follow up with you on that.
Gohmert: SO YOU DON’T KNOW?!? You sure seemed anxious to answer the Democrats questions about RUSSIA’s influence. And you don’t really know of all the groups that inappropriately used your platform? You don’t know which were Russians and which were other foreign entities?
No, that’s not what she’s saying at all. She’s pretty clearly saying that this hearing was specifically about Russian influence and that’s what she was prepared to testify on. She didn’t say that Facebook can’t tell Russians from other entities, just that the other entities aren’t the ones accused of messing with the election and thus there isn’t that much relevant right now. But that’s quite a deflection attempt by Gohmert.
Let’s move on to Rep. Tom Marino at about an hour and a half into the video. Marino seems to have a fairly bizarre understanding of the law as it concerns defamation. He focuses on the guy from Twitter, Nick Pickles, and starts out by reading a definition of “libel.” Then he asks
Have any of you considered libel? Or do you think you are immune from it?
This is an incredibly stupid question. Twitter is clearly not immune from libel. Marino’s line of questioning is an attempt to attack CDA 230, which provides immunity to Twitter from liability for defamatory statements made by its users. This is an important distinction that Marino conveniently ignores as he continues to bug Pickles.
Pickles: We have clear rules that governs what happens on Twitter. Some of those behaviors are deplorable and we want to remove them immediately… So, terrorist content is one example, where we now detect 95% of the terrorist accounts we remove…
Marino: Okay, I understand that sir. But how about… we in Congress, we put up with it all the time. I know we’re public officials, same with people in the movies… but do you specifically look for and address… republication can be used in a defamation case. Do you look at libel and defamation content?
I don’t even know what that means. Do you look at libel content? What? How does Twitter know if something is libelous? Especially against public officials? How is Twitter supposed to make that judgment when that’s what courts are there to figure out? And, for what it’s worth, Twitter has been known to abide by court rulings on defamatory speech in deciding to take down that content, but Marino seems to be asking if they make an independent judgment outside of the courts of what’s libelous. Which is both crazy and impossible. Pickles makes a valiant effort in response, noting how Twitter focuses on its rules — which is all that it’s required to do — but Marino clearly seems to want to attack CDA 230 and magically make Twitter liable for libelous content on its platform. After Pickles again explains that it focuses on its rules, rather than making judicial rulings that it cannot make, Marino puts on a dumb smirk and makes another dumb statement:
With all due respect, I’ve heard you focus on your rules about 32 times. DO. YOU. LOOK. FOR. LIBEL. OR. DEFAMATION. IN. YOUR. COMPANY’S. OPINION?
You can’t “look for libel or defamation” like that. That’s not how it works. Marino is a lawyer. He should know this. The Facebook and YouTube representatives neatly sidestep Marino’s silly line of questioning by pointing out that when informed of legal rulings determining “illegal” speech, they take it down. Marino doesn’t even seem to notice this very specific distinction and asks “where do you draw the line?”
At an hour and forty minutes, we have everyone’s favorite, Rep. Lamar Smith, author of SOPA back in the day. He spews more utter nonsense claiming conservatives have been more negatively impacted by the moves of these social media companies, and then (bizarrely) argues that Google employees forcing the company not to help surveillance activity is somehow an attack on conservatives. Excuse me? Conservatives don’t support the 4th Amendment any more? Say what? But the real craziness is this line:
Google has also deleted or blocked references to Jesus, Chick-Fil-A and the Catholic religion.
I’m going to call time out here and note [citation needed] on that one, Smith. Google pretty clearly shows me results on all three of those things. I’ve been trying to figure out what the hell he’s referring to, and I’m guessing that Smith — in his usual Smithian nonsensical way — is confusing Google for Facebook, and Facebook’s bad filter that initially blocked a page about “Chick-fil-Appreciation Day,” and some Catholic church pages. The “Jesus” blocking is also Facebook and was in reference to an ad for a Catholic university.
All of these examples were not, as Smith implies, evidence of “liberal bias” on behalf of Facebook, but rather evidence of why it’s so problematic that governments are putting so much pressure on Facebook to magically filter out all of the bad stuff. That’s not possible without making mistakes. And what happens is that you set up guidelines and those guidelines are then handed to people who don’t have nearly enough time to understand the context, and sometimes they make mistakes. It’s not bias. It’s the nature of trying to moderate millions of pieces of content every damn day, because if they don’t, these same idiots in Congress would be screaming at them about how they’re letting the bad content live on. I mean, it’s doubly ridiculous for Smith to use the Jesus example as even the guy who bought the ad, the university’s web communications director, specifically said that he didn’t believe it had anything to do with bias, but was just a bad decision by an algorithm or a low level staffer.
Finally (and there are more, but damn, this post is getting way too long) we get to Rep. Matt Gaetz. At around an hour and 55 minutes into the hearing, he suddenly decides to weigh in that the First Amendment and CDA 230 are somehow in conflict, in another bizarre exchange between Gaetz and Twitter’s Pickles.
Gaetz: Is it your testimony or is it your viewpoint today that Twitter is an interactive computer service pursuant to Section 230 sub c(1).
Pickles: I’m not a lawyer, so I won’t want to speak to that. But as I understand, under Section 230, we are protected by that, yes.
Gaetz: So Section 230 covers you, and that section says “no provider of an interactive computer service shall be treated as the publisher or speaker of any information provided by another”… is it your contention that Twitter enjoys a First Amendment right under speech, while at the same time enjoying Section 230 rights?
Pickles: Well, I think we’ve discussed the way the First Amendment interacts with our companies. As private companies we enforce our rules, and our rules prohibit a range of activities.
Gaetz: I’m not asking about your rules. I’m asking about whether or not you believe you have First Amendment rights. You either do or you do not.
Pickles: I’d like to follow up on that, as someone who is not a lawyer… I think it’s very important…
Gaetz: Well, you’re the senior public policy official for Twitter before us and you will not answer the question whether or not you believe your company enjoys rights under the First Amendment?
Pickles: Well, I believe we do, but I would like to confirm with colleagues…
Gaetz: So what I want to understand is, if you say “I enjoy rights under the First Amendment” and “I’m covered by Section 230” and Section 230 itself says “no provider shall be considered the speaker” do you see the tension that creates?
There is no tension there. The only tension is between the molecules in Gaetz’s brain that seemed to think this line of nonsensical argument makes any sense at all. There is no conflict. First, yes, it’s obvious that Twitter is clearly protected by both the First Amendment and CDA 230. That’s been established by dozens of court rulings with not a single ruling ever holding otherwise. Second, the “tension” that Gaetz sees is purely a figment of his own misreading of the law. The “no provider shall be considered a speaker” part, read in actual context (as Gaetz did earlier) does not say that platforms are not speakers. It says that they are not considered a speaker of other people’s speech. In fact, this helps protect free speech by enabling internet platforms the ability to host any speech without facing liability for that speech.
That helps protect the First Amendment by ensuring that any liability is on the speaker and not on the tool they use to distribute that speech. But Twitter has its own First Amendment rights to determine what speech it decides to keep on its site — and which speech it decides not to allow. Gaetz then, ridiculous, tries to claim that Pickle’s response to that nonsensical response is somehow in conflict with what Twitter’s lawyers have said in the silly Jared Taylor lawsuit. Gaetz asks Pickles if Twitter could kick someone off the platform “for being a woman or being gay.” Pickles points out that that is not against Twitter’s rules… and Gaetz points out that in the Taylor case, when asked the same question, Twitter’s lawyers stated (1) that Twitter has the right to do so but (2) never would.
Again, both Pickles and Twitter’s lawyers are correct. They do have that right (assuming it’s not a violation of discrimination laws) but of course they wouldn’t do that. Pickles wasn’t denying that. He was pointing out that the hypothetical is silly because that’s not something Twitter would do. Twitter’s lawyers in the case were, correctly, pointing out that it would have the right to do such a nonsensical thing if it chose to do so, while also making it clear it would never do that. Again, that’s not in conflict, but Gaetz acts as if he’s “caught” Twitter in some big admission.
Gaetz falsely then claims that Pickles is misrepresenting Twitter’s position:
Right but it is not in service of transparency if Twitter sends executives to Congress to say one thing — that you would not have the right to engage in that conduct — and then your lawyers in litigation say precisely the opposite.
Except that’s not what happened at all. Pickles and the lawyers agreed. At no point did Pickles say that Twitter did not have “the right” to kick people off its platform for any reason. He just noted that it was not a part of their policy to do so, nor would it ever be. That’s entirely consistent with what Twitter’s lawyers said in the Taylor case. This is Gaetz making a complete ass out of himself in completely misrepresenting the law, the constitution and what Twitter said both in the hearing and in the courthouse.
Seriously, people, we need to elect better Representatives to Congress. This is embarrassing.
As you may have heard, Rep. John Conyers recently stepped down from his role as Ranking Member (basically top member of the minority party) on the powerful House Judiciary Committee, and this week has announced his retirement, in response to multiple accusations of sexual harassment. That has kicked off something of an interesting and important debate over who should replace him as ranking member on the Judiciary Committee.
The next in line by seniority is Rep. Jerry Nadler. But right behind him is Rep. Zoe Lofgren. By way of disclosure, I’ll note that I’ve gotten to know Lofgren over the years, and have donated to her election campaign. But even before I’d ever spoken to her, I’ve noted how she remains one of the few people in Congress who seems to consistently do the right thing on basically all of the issues that we care about at Techdirt. You can see our past coverage of stories involving Lofgren. Most specifically on copyright and surveillance, she hasn’t just been on the right side, she’s been leading the way. She is, almost single-handedly, the person who stopped SOPA from passing. She has consistently raised important issues and introduced important bills and amendments concerning copyright, NSA surveillance and the CFAA among other things.
Obviously, I think she’d make a great ranking member for the Judiciary Committee (or the chair should the House flip sides in the future). So I was happy to see her recently announce her intention to run for the Ranking Member position against Nadler. Who knows if she’ll actually get the position, but I found it odd that upon announcing it, she was immediately attacked by, of all places, The Intercept, which put forth a really strange article accusing Lofgren of being a Google shill. This was strange on multiple levels — though, I get it. Lofgren gets called a “Google shill” for the same reasons that we do here at Techdirt. Because, even though we frequentlydisagree with Google on a variety of issues, on the whole we support many of the same policies that protect free speech and open innovation online.
That’s also true of Lofgren. While she’s supported key policies on copyright, online speech, innovation and surveillance, she’s similarly pushed back against Google quite frequently as well. She’s publicly criticized the company for its lack of diversity. She’s voted against a bill to expand H1-B visas that Google supported. She voted against Trade Promotion Authority (which Google stupidly supported — as noted in one of my links above) that paved the way to moving forward on TPP. On top of that, the tech industry has mostly pushed back on CFAA reform, such as Lofgren’s Aaron’s Law, because companies want to have it as a tool to use against employees at times. Just recently, Lofgren has started digging into competition inssues in Silicon Valley, warning about the lack of competition and how it’s a problem — a position that, more than likely, Google finds worrisome.
That’s just part of why it’s so odd that the Intercept, of all publications, would post this article suggesting that Lofgren doesn’t belong as the ranking member on the Judiciary Committee just because she’s “close” to Google. Even odder, is the fact that the authors of the piece — two reporters whose work I’ve long respected, Ryan Grim and Lee Fang — focus entirely on claiming that Lofgren is a product of Google, while ignoring anything about Nadler. Not only has Nadler been on the wrong side of many of these same key issues, if you consider Lofgren somehow tied to Google (again, incorrectly) then you would similarly have to conclude that Nadler is in the pocket of the legacy entertainment industry, and their ongoing quest to destroy the internet as we know it. If you start looking at Nadler’s campaign finance situation, it sure looks like he’s the MPAA and the RIAA’s favorite Congressman.
In the last campaign cycle, the RIAA gave significantly more to Nadler than any other Democrat. Same with Disney. Same with Sony. Same with Time Warner. Same with Universal Music. Same with the Association of American Publishers. Same with ASCAP. While Viacom gave a bit more to three other members, Nadler was the 4th highest support on the Democratic side. Comcast gave a little more to Conyers, but again, Nadler is near the top of the list. The Grammys have given more to Nadler than any other Democrat, and he repays them by holding events with them all the time.
There’s a pretty clear pattern here. If the legacy copyright players want something on the Democratic side, Nadler’s their guy. And, maybe that doesn’t matter to the Intercept. Maybe it doesn’t matter that bad copyright policies that he promotes would have serious downsides to the way the internet works, to free speech and to privacy. Maybe, the Intercept has decided that any possible “connection” to Google is worse than everything else. But considering that the whole creation of The Intercept came about because of the Snowden revelations, and a key focus of The Intercept is to report on the evils of government surveillance, it’s kind of surprising that it would publish an article promoting Nadler over Lofgren while ignoring that Nadler has not always been a close friend of surveillance reform. It’s true that he’s sponsored some reform efforts, including the USA Freedom Act, but just last month he was seen voting against an important amendment brought forth by Lofgren, to end backdoor searches in the ongoing effort to reform Section 702.
So it seems odd that the Intercept is effectively arguing that Nadler would make a better ranking member on Judiciary, even as Lofgren has a stronger record on stopping government surveillance, just because some (falsely) believe that Lofgren is “tied” to Google. And, at the very least, if they’re going to tar Lofgren because her views sometimes align with Google’s, it seems that it could at least treat Nadler equally by looking into his close connections with the legacy entertainment business.
As we’ve discussed there’s this stupid big fight going on these days, in which some in Congress — mainly at the urging of the legacy entertainment industry — are looking to move the Copyright Office out of its historical home in the Library of Congress. The first proposal to sort of (but not completely) do that, involved just making the head of the Copyright Office a Presidential appointment position, rather than (as now) appointed by the Librarian of Congress. The main reason that various members of Congress put forth in support of this change was that this would magically give the Copyright Office the freedom to modernize. Of course, there are few facts to support this argument. We broke the story about serious incompetence at the Copyright Office in managing its own modernization efforts, and there was also plenty of evidence that the current Librarian of Congress was successfully moving forward with a thorough modernization plan.
And, yet, the House Judiciary Committee still voted overwhelmingly to move the bill out of committee. Thankfully, it appears the bill is pretty much dead in the water for now, apparently in part because some people noticed that it’s not really the Judiciary Committee’s jurisdiction. Judiciary has power over issues related to copyright, but this isn’t a bill about copyright, but about administration. That belongs elsewhere and apparently some folks are none too pleased that the Judiciary Committee went behind their backs on this effort.
And then there’s this: last week in the Appropriations Committee’s latest appropriations bill for the legislative branch, it pointed out that the Library was doing a good job in modernizing the Copyright Office. Here’s the relevant section:
Copyright Modernization: The Committee is encouraged by the
collaborative work between the United States Copyright Office
(USCO) and the Library of Congress?s Office of the Chief Information
Officer?s Office (OCIO) and is looking forward to the USCO?s
revised provisional IT plan which is expected in early August. The
Committee continues to support a shared-services approach with
regards to commodity IT services. Copyright modernization is
something the Committee fully supports and will continue to provide
appropriate resources. As we go forward OCIO is encouraged
to engage with stakeholders both in the Congressional-community
and beyond to outline clear benchmarks for progress.
In other words, the very reason given by the Judiciary Committee for why we need a separate Copyright Office has been totally undermined by the Appropriations Committee, who actually took the time to figure out what was going on. Now, some of this might just be fighting over domains, but it raises even more questions about why some in Congress are so eager to yank the Copyright Office away from the Library of Congress at a time when the modernization program seems to be moving forward successfully.
For many, many years, Senator Ron Wyden has been directly asking the US intelligence community a fairly straightforward question (in his role as a member of the Senate Intelligence Committee): just how many Americans are having their communications swept up in surveillance activities supposedly being conducted on foreigners under the FISA Amendments Act (FISA being Foreign Intelligence Surveillance Act). Wyden started asking way back in 2011 and got no answers. His continued questioning in 2013 resulted in Director of National Intelligence James Clapper lying to Congress in a public hearing, which Ed Snowden later claimed was a big part of the inspiration to make him leak documents to the press.
Just last month, we noted that Wyden had renewed his request for an accurate depiction of how many Americans have had their communications swept up, this time asked to new Director of National Intelligence, Dan Coats. Unfortunately, for all these years, it’s basically felt like Senator Wyden tilting at a seeming windmill, with many others in Congress basically rolling their eyes every time the issue is raised. I’ve never understood why people in Congress think that these kinds of things can be ignored. There have been a few attempts by others — notably on the House Judiciary Committee — to ask similar questions. Almost exactly a year ago, there was a letter from many members of the HJC, and there was a followup in December. But, notably, while there were a number of members from both parties on that letter, the chair of the House Judiciary Committee, Bob Goodlatte, did not sign the letter, meaning that it was unlikely to be taken as seriously.
Suddenly, though, it seems that the ins-and-outs of Section 702, and how the “incidental” information it collects on Americans is used has taken on a much wider interest, following President Trump’s misleading suggestion that President Obama tapped his phone lines, and some Trump supporters trying to twist typical 702 surveillance to justify those remarks. Either way, if that leads people to actually look at 702, that may be a good result out of a stupid situation. And, thus, we get to this surprising moment, in which Goodlatte has actually sent a similar letter to Coats (along with ranking member John Conyers) asking about the impact of 702 surveillance on Americans. And since (for reasons that are beyond me) Reuters refuses to link to the actual source materials, you can read the full letter here or embedded below.
The letter demands an answer by April 24th. And, yes, it’s notable that Goodlatte has signed on, because Section 702 is up for reauthorization at the end of the year, and if Goodlatte is not on board with reauthorization, then the NSA is going to have some difficulty in getting it through.
You have described reauthorization of Section 702 as your “top legislative priority.” Although Congress designed this authority to target non-U.S. persons located outside of the United States, it is clear that Section 702 surveillance programs can and do collect information about U.S. persons, on subjects unrelated to counterterrorism. It is imperative that we understand the size of this impact on U.S. persons as our Committee proceeds with the debate on reauthorization.
The letter then even points to Coats’ response to Wyden during Coats’ confirmation hearing that he was “going to do everything I can to work with Admiral Rogers in NSA to get you that number.” Of course, back in December, it was said that the intelligence community might finally deliver that number… in January. And it’s now April. Still, with Goodlatte finally taking an interest in this, it’s a sign that the NSA can’t just coast by and continue to completely ignore this.
Two bipartisan Congressional committees are the latest to express their opposition to government-mandated encryption backdoors. The House Judiciary Committee and the House Energy and Commerce Committee have arrived at the same conclusion as the experts FBI director James Comey insists on ignoring: encryption backdoors are a net loss for everyone, no matter what gains might be experienced by law enforcement and intelligence agencies.
This is stated plainly in the first bullet point of its encryption report [PDF]:
Any measure that weakens encryption works against the national interest
While the committees acknowledge encryption can impede investigative efforts, the downsides of backdoors cannot be offset by making things easier for certain government agencies.
[S]takeholders from all perspectives acknowledged the importance of encryption to our personal, economic, and national security. Representatives of the national security community told the EWG that strong encryption is vital to the national defense and to securing vital assets, such as critical infrastructure. Civil society organizations highlighted the importance of encryption for individual privacy, freedom of speech, human rights, and protection against government intrusion at home and abroad. Private sector stakeholders—in particular, their information security officers—and members of the academic community approached the question from an engineering perspective—against a wide array of threats, foreign and domestic, encryption is one of the strongest cybersecurity tools available.
However, the committees still believe there might be a way to reconcile competing interests, even though it has more questions than answers at this point. The report suggests more “collaboration” between tech companies and law enforcement agencies — a term that generally means most of the compromises will be made by the private sector. Whether this means companies collecting more data and communications and storing them where law enforcement can access them or creating “one time” backdoors in response to court orders remains to be seen.
More encouragingly, the report suggests the “smart guys” in law enforcement haven’t fully taken advantage of the tools and data available to them.
It also remains unclear whether the law enforcement community is positioned to fully leverage the unencrypted information still held by many companies. A number of stakeholders acknowledged the potential benefit of improving law enforcement’s understanding of what data or information is available, who controls it, and how it could be useful to investigators. In particular, companies are often able to provide volumes of unencrypted metadata associated with their products or services. In some cases, this source of information could be useful to investigators. In others, one representative of a law enforcement agency told the EWG, access to a stream of metadata might be more like “looking for a particular grain of sand on the beach.”
This is probably the result of the law enforcement mindset. It often seems agencies are more interested in what is quickest and easiest, rather than what might be more productive, if just a bit more difficult. (A number of cases where warrants were never obtained, despite officers having both the time and probable cause to do so, is evidence of this mindset.) The report suggests this is one area where things could be improved by collaboration with private companies. It’s not a terrible suggestion but it’s one that requires agencies to move on from their defeatist attitudes and to stop pretending advances in technology are always far more beneficial to criminals than to law enforcement.
The report also inadvertently points out just how disingenuous it is to shrug off mass surveillance concerns by saying, “It’s just metadata.”
Metadata may not completely replace the loss of encrypted content, but metadata analysis could play a role in filling in the gap. The technology community leverages this information every day to improve services and target advertisements. There appears to be an opportunity for law enforcement to better leverage this information in criminal investigations.
The report also touches on “legal hacking” as a potential solution — albeit one with very limited practical application. If this is the route the government chooses to go more frequently in response to encrypted devices, it will signal the end of the already mostly-worthless Vulnerabilities Equity Process. It would also — as the report acknowledges — only further the “us vs. them” conflict between tech companies and law enforcement, as the government’s interest in keeping vulnerabilities secret would tend to outweigh its obligation to divulge security holes to affected companies.
While the report breaks very little new ground in terms of issues raised, it does at least signal that legislative efforts to undermine encryption aren’t likely to find much bipartisan support. So, for the time being, device encryption is still safe. It’s the other issues raised — legal hacking, compelled disclosure, etc. — that will need to be watched closely in the future.
