The Metric Is Not The Mission: The Digital Climate
The Metric Is Not the Mission is a ten-part examination of how Big Tech moved from building and expanding the open internet to increasingly shaping it around its own metrics, incentives and assumptions. Across the series, the argument follows the evolution of the platform economy—from the optimism of the early internet to the growing tensions around power, prediction, geopolitics, accountability and the future of digital life.
The series will be published in two parts each week over five weeks, with each installment building on the one before it. At the end of the series, the complete essay will be brought together in a single PDF edition, providing the full argument in one place.
Part IV: The Digital Climate
Part III argued that the map can never fully capture the territory. This part looks beyond individual controversies to the changing environment itself and asks whether the assumptions that once made platforms successful still fit the digital world they helped create.
Every generation believes that it is living through an unprecedented period of change. Usually, this is little more than historical vanity. The technologies differ, the political actors change, and the crises acquire new names, but the underlying patterns often remain remarkably familiar. Societies innovate, institutions adapt, markets expand, and eventually each encounters the same fundamental challenge: success changes the environment in ways that success itself rarely anticipates.
Biologists have long understood this phenomenon. Evolution does not prepare organisms for the future; it prepares them for the past. Every adaptation reflects conditions that once existed, not necessarily those that are emerging. A species that becomes exquisitely suited to one environment may discover, often too late, that the environment itself has changed. Extinction is therefore not always the consequence of weakness. More often, it is the inability to recognize that the conditions which rewarded one form of intelligence have quietly given way to another.
Economists describe a similar phenomenon as path dependence. Institutions continue investing in strategies that have historically produced success because those strategies appear rational when viewed through the lens of accumulated experience. Military historians often note that generals tend to prepare for the previous war. Financial regulators strengthen oversight after the last crisis, rarely before the next one. Organizations, almost by definition, learn retrospectively. Their accumulated knowledge is also their greatest constraint. Technology companies are no exception.
If anything, the extraordinary success of the major platforms may have made them particularly vulnerable to this form of institutional lag. The models they developed during the internet’s age of expansion proved astonishingly effective at connecting people, organizing information and lowering the costs of participation. The metrics through which they evaluated success, such as growth, engagement, scale, and network effects, were not arbitrary inventions of venture capital. They reflected a period during which connecting more people genuinely created more value for everyone involved. The problem is not that these metrics were wrong. It is that they gradually became targets in themselves. The world changed while the metrics remained stable.
For much of the first decade of social media, the dominant challenge was abundance. The internet was expanding faster than anyone could meaningfully navigate it. There were too many websites, too many videos, too many blogs, too many voices competing for attention. Search engines, recommendation systems and social networks addressed this problem by filtering complexity. Their algorithms reduced uncertainty by pointing users toward content they were likely to find useful, interesting, or entertaining. In a world defined by informational overload, curation felt like liberation.
Today’s challenge is almost the inverse. Few people now struggle to find information. Instead, they struggle to determine which information deserves trust. The scarcity that once characterized knowledge has been replaced by a scarcity of confidence. We inhabit an environment saturated with images, opinions, analyses, commentary, and increasingly synthetic media. Artificial intelligence has accelerated this transformation still further, reducing the cost not only of distributing information but of producing it in almost limitless quantities. The bottleneck has shifted. Discovery matters less than discernment.
Yet the architecture of the dominant platforms remains largely optimized for the earlier age. Systems originally designed to maximize relevance increasingly operate within an environment where reliability has become the more precious commodity. Recommendation engines still reward attention because attention remains measurable. Trust, by contrast, is extraordinarily difficult to quantify. It develops slowly, depends on context and often requires forms of human judgment that resist computational simplification. The result is a subtle but increasingly consequential mismatch between what societies need and what platforms are designed to provide.
This helps explain why so many contemporary debates about social media seem oddly unsatisfying. Discussions about misinformation, disinformation, harmful content, or political polarization often assume that these are discrete policy failures awaiting technical solutions. Better moderation, greater transparency, more sophisticated artificial intelligence, or improved regulation are presented as though they might restore an earlier equilibrium. Each proposal addresses an important part of the problem, yet none fully confronts the possibility that the problem is structural rather than operational.
Climate science again offers a useful analogy. For decades, public debates about climate change focused overwhelmingly on individual weather events. Every hurricane, wildfire, or flood generated renewed discussion about global warming, only for the conversation to subside once the immediate crisis had passed. Scientists repeatedly pointed out that this way of thinking misunderstood the phenomenon itself. Climate change is not the accumulation of disasters but the gradual alteration of the conditions that make particular kinds of disasters more likely. Focusing exclusively on the storm obscures the atmosphere that produced it.
Something similar has happened in our understanding of digital platforms. We have spent years debating individual scandals as though each represented an isolated malfunction: election interference, extremist content, privacy violations, online harassment, conspiracy theories, mental health, child protection, deepfakes, artificial intelligence. Each deserves careful attention. Yet viewed collectively, they begin to resemble not separate failures but different manifestations of the same underlying transformation. They are symptoms of an ecosystem whose governing assumptions increasingly diverge from the societies that depend upon it.
This is not to suggest that the companies involved acted with malicious intent. Structural problems rarely emerge because individuals make consistently immoral choices. They emerge because institutions continue behaving rationally according to incentives that no longer correspond to external reality. A recommendation algorithm that prioritizes engagement is not malfunctioning when it amplifies emotionally charged content. It is doing precisely what it was designed to do. The more uncomfortable question is whether the objective itself remains appropriate for the world in which the system now operates.
That distinction matters because it shifts responsibility from individual decisions to institutional imagination. The challenge is no longer simply how to improve content moderation or reduce harmful behavior online. It is whether platforms can rethink the assumptions embedded within business models that were developed for a different internet, a different political environment, and, indeed, a different conception of human interaction.
Perhaps the most striking evidence that this shift has already occurred lies not in regulation or public opinion but in the behavior of users themselves. Increasingly, people describe social media less as a place they enjoy than as a place they feel obliged to inhabit. Creators remain because audiences are there. Businesses remain because customers are there. Politicians remain because voters are there. Journalists remain because news breaks there. Ordinary users remain because leaving often means losing access to relationships, professional networks, or cultural conversations that have become difficult to reproduce elsewhere. This is a curious form of success.
Historically, the most admired technologies inspired affection. They expanded human possibility in ways that people found genuinely exhilarating. The telephone shortened distances; the airplane compressed geography; and the web rewarded curiosity. By contrast, many of today’s dominant platforms increasingly resemble public utilities that happen to be privately owned: indispensable, deeply embedded in everyday life, and frequently resented by the very people who rely upon them.
That emotional transition, from enthusiasm to resignation, may prove more significant than any quarterly earnings report or competition case. Institutions rarely lose their influence all at once. More often, they lose something subtler first – the confidence that they understand the age they inhabit.
The great irony is that no organizations in history have possessed more information about human behavior than the leading technology companies of the twenty-first century. They know where we pause, what we share, whom we follow, when we lose interest, and what captures our attention again. They have become extraordinarily skilled at modelling behavior. But behavior is neither culture nor society. And behavior, however precisely measured, is certainly not the same thing as wisdom. That is the digital climate we have entered. The storms still occupy our attention. Meanwhile, the atmosphere has already changed.
Konstantinos Komaitis, PhD, is a veteran of developing and analysing Internet policy to ensure an open and global Internet.
Filed Under: behavior, big tech, metric not mission, open internet, optimization, platforms
Trump Breaks Law(s),Uses $20 Million In Taxpayer Money To Fund Midterm Propaganda
from the autocratic-agitprop dept
Trump FCC boss Brendan Carr endlessly pays empty lip service to the “public interest” and the “value of broadcasting integrity” as he engages in ham-fisted censorship of comedians and journalists the president doesn’t like. Or the illegal dismantling of media consolidation limits. Or helps the president and his friends steal TikTok.
But any time there’s an actual threat to broadcasting integrity or the public interest he’s unsurprisingly nowhere to be found. Like last week when his FCC rubber stamped heavy Saudi and Chinese investment in the Paramount merger despite obvious foreign influence concerns.
Or this week, when his FCC sat on its hands as the Trump administration used $20 million in taxpayer dollars to air rank propaganda ahead of the midterms across major broadcast networks.
A new set of rehashed ads feature Donald in black and white as he walks down a hallway promising to smite all of his enemies, including “the deep state,” “warmongers,” the “fake news media,” “Marxists,” and “fascists.” A notice at the bottom of the ad states it was “paid for by the US government.”
A previous version of the ad was funded by Trump’s campaign committee, not the US government. The ad violates laws restricting the use of taxpayer money for domestic propaganda (you can tell we’ve really been enforcing that one) — and the Hatch Act (ditto), correctly notes a complaint to the FTC, FCC, Government Accountability Office (GAO) and Office of Special Counsel (OSC):
“Under federal communications law, broadcast stations are strictly prohibited from censoring or rejecting ads sponsored by legally qualified political candidates. Even if a candidate’s ad
contains graphic imagery or disputed claims, stations generally must air it. But there are clear
exceptions to this rule.Broadcasters may be held liable for airing any advertisement that is in violation of federal law. Federal law explicitly prohibits using appropriated public funds for government propaganda, self-aggrandizement, or partisan political advertising. Yet, that is exactly the nature of Trump’s recent campaign ads that began airing on September 23, 2026. Furthermore, Donald Trump is not a candidate for any elective office. The FCC and broadcasters are not constrained by the special rule protecting candidate advertisements.”
According to Public Citizen, the ad ran during NBC’s Saturday Night Live, Fox News Sunday, and several college and NFL football games this past weekend. While the Trump administration is trying to dodge the law by falsely claiming the ad is a “public service announcement,” it doesn’t actively promote any real programs or services, making it a very clear violation of existing law.
In addition to Public Citizen’s complaints, Senate Democrats sent a letter to Homeland Security Secretary Markwayne Mullin this week stating that Trump is “illegally using Americans’ tax dollars to fund government propaganda and to put ads from his 2024 campaign back on the air.” It’s quite the waste of cash, given that associating yourself with a historically unpopular and corrupt president appears to be polling and political suicide this midterm season.
The Trump FCC will, of course, do nothing. Brendan Carr told the media that these are just “normal” “public service announcements” that “don’t raise any red flags.” He also insisted it’s fine because, he claimed falsely, the “left” does this all the time:
“Carr brushed off the concern, telling reporters that “there’s groups on the left that have, you know, long worked and pushed to weaponize the FCC’s license renewal process” but that the ads “don’t raise any red flags.”
Carr’s referring to a single event in 2023 when the volunteer-run Media and Democracy Project (MAD) petitioned the FCC to pull the license of a Fox Philly affiliate for repeatedly running lies about election fraud. What Carr leaves out is that while there was that one effort from “the left” to stop the renewal of a Fox license, at the same time there was an even bigger effort from a far right group to block the renewal of licenses for three CBS stations that aired an interview with Kamala Harris that the MAGA faithful deemed was edited to make Harris look good.
He also leaves out that his predecessor (the Democrat Jessica Rosenworcel) rejected all four of those requests, citing the First Amendment and pointed out that all such attempts to block license renewals for partisan reasons was an attack on free speech and a free press. Oh and he also leaves out that after he took over, he reinstated the efforts against the CBS affiliates, but not the Fox one.
In other words, only one side (Carr and his MAGA minions) appears to be engaged in partisan attempts to suppress speech.
Carr’s far more interested in dismantling all oversight of telecom monopolies, illegally dismantling media consolidation limits, trampling free speech, and forcing broadcasters to air discounted midterm ads. Broadcasters, meanwhile, are also unlikely to balk at running the ads, lest they want to face a fake Brendan Carr inquiry into “DEI” or some other, manufactured nonsense.
Another day in a corrupt autocracy run by the shittiest zealots imaginable.
Filed Under: advertisements, agitprop, brendan carr, donald trump, fcc, hatch act, illegal, midterms, propaganda, taxpayer money
The Metric Is Not the Mission is a ten-part examination of how Big Tech moved from building and expanding the open internet to increasingly shaping it around its own metrics, incentives and assumptions. Across the series, the argument follows the evolution of the platform economy—from the optimism of the early internet to the growing tensions around power, prediction, geopolitics, accountability and the future of digital life.
The series will be published in two parts each week over five weeks, with each installment building on the one before it. At the end of the series, the complete essay will be brought together in a single PDF edition, providing the full argument in one place.
Part IV: The Digital Climate
Part III argued that the map can never fully capture the territory. This part looks beyond individual controversies to the changing environment itself and asks whether the assumptions that once made platforms successful still fit the digital world they helped create.
Every generation believes that it is living through an unprecedented period of change. Usually, this is little more than historical vanity. The technologies differ, the political actors change, and the crises acquire new names, but the underlying patterns often remain remarkably familiar. Societies innovate, institutions adapt, markets expand, and eventually each encounters the same fundamental challenge: success changes the environment in ways that success itself rarely anticipates.
Biologists have long understood this phenomenon. Evolution does not prepare organisms for the future; it prepares them for the past. Every adaptation reflects conditions that once existed, not necessarily those that are emerging. A species that becomes exquisitely suited to one environment may discover, often too late, that the environment itself has changed. Extinction is therefore not always the consequence of weakness. More often, it is the inability to recognize that the conditions which rewarded one form of intelligence have quietly given way to another.
Economists describe a similar phenomenon as path dependence. Institutions continue investing in strategies that have historically produced success because those strategies appear rational when viewed through the lens of accumulated experience. Military historians often note that generals tend to prepare for the previous war. Financial regulators strengthen oversight after the last crisis, rarely before the next one. Organizations, almost by definition, learn retrospectively. Their accumulated knowledge is also their greatest constraint. Technology companies are no exception.
If anything, the extraordinary success of the major platforms may have made them particularly vulnerable to this form of institutional lag. The models they developed during the internet’s age of expansion proved astonishingly effective at connecting people, organizing information and lowering the costs of participation. The metrics through which they evaluated success, such as growth, engagement, scale, and network effects, were not arbitrary inventions of venture capital. They reflected a period during which connecting more people genuinely created more value for everyone involved. The problem is not that these metrics were wrong. It is that they gradually became targets in themselves. The world changed while the metrics remained stable.
For much of the first decade of social media, the dominant challenge was abundance. The internet was expanding faster than anyone could meaningfully navigate it. There were too many websites, too many videos, too many blogs, too many voices competing for attention. Search engines, recommendation systems and social networks addressed this problem by filtering complexity. Their algorithms reduced uncertainty by pointing users toward content they were likely to find useful, interesting, or entertaining. In a world defined by informational overload, curation felt like liberation.
Today’s challenge is almost the inverse. Few people now struggle to find information. Instead, they struggle to determine which information deserves trust. The scarcity that once characterized knowledge has been replaced by a scarcity of confidence. We inhabit an environment saturated with images, opinions, analyses, commentary, and increasingly synthetic media. Artificial intelligence has accelerated this transformation still further, reducing the cost not only of distributing information but of producing it in almost limitless quantities. The bottleneck has shifted. Discovery matters less than discernment.
Yet the architecture of the dominant platforms remains largely optimized for the earlier age. Systems originally designed to maximize relevance increasingly operate within an environment where reliability has become the more precious commodity. Recommendation engines still reward attention because attention remains measurable. Trust, by contrast, is extraordinarily difficult to quantify. It develops slowly, depends on context and often requires forms of human judgment that resist computational simplification. The result is a subtle but increasingly consequential mismatch between what societies need and what platforms are designed to provide.
This helps explain why so many contemporary debates about social media seem oddly unsatisfying. Discussions about misinformation, disinformation, harmful content, or political polarization often assume that these are discrete policy failures awaiting technical solutions. Better moderation, greater transparency, more sophisticated artificial intelligence, or improved regulation are presented as though they might restore an earlier equilibrium. Each proposal addresses an important part of the problem, yet none fully confronts the possibility that the problem is structural rather than operational.
Climate science again offers a useful analogy. For decades, public debates about climate change focused overwhelmingly on individual weather events. Every hurricane, wildfire, or flood generated renewed discussion about global warming, only for the conversation to subside once the immediate crisis had passed. Scientists repeatedly pointed out that this way of thinking misunderstood the phenomenon itself. Climate change is not the accumulation of disasters but the gradual alteration of the conditions that make particular kinds of disasters more likely. Focusing exclusively on the storm obscures the atmosphere that produced it.
Something similar has happened in our understanding of digital platforms. We have spent years debating individual scandals as though each represented an isolated malfunction: election interference, extremist content, privacy violations, online harassment, conspiracy theories, mental health, child protection, deepfakes, artificial intelligence. Each deserves careful attention. Yet viewed collectively, they begin to resemble not separate failures but different manifestations of the same underlying transformation. They are symptoms of an ecosystem whose governing assumptions increasingly diverge from the societies that depend upon it.
This is not to suggest that the companies involved acted with malicious intent. Structural problems rarely emerge because individuals make consistently immoral choices. They emerge because institutions continue behaving rationally according to incentives that no longer correspond to external reality. A recommendation algorithm that prioritizes engagement is not malfunctioning when it amplifies emotionally charged content. It is doing precisely what it was designed to do. The more uncomfortable question is whether the objective itself remains appropriate for the world in which the system now operates.
That distinction matters because it shifts responsibility from individual decisions to institutional imagination. The challenge is no longer simply how to improve content moderation or reduce harmful behavior online. It is whether platforms can rethink the assumptions embedded within business models that were developed for a different internet, a different political environment, and, indeed, a different conception of human interaction.
Perhaps the most striking evidence that this shift has already occurred lies not in regulation or public opinion but in the behavior of users themselves. Increasingly, people describe social media less as a place they enjoy than as a place they feel obliged to inhabit. Creators remain because audiences are there. Businesses remain because customers are there. Politicians remain because voters are there. Journalists remain because news breaks there. Ordinary users remain because leaving often means losing access to relationships, professional networks, or cultural conversations that have become difficult to reproduce elsewhere. This is a curious form of success.
Historically, the most admired technologies inspired affection. They expanded human possibility in ways that people found genuinely exhilarating. The telephone shortened distances; the airplane compressed geography; and the web rewarded curiosity. By contrast, many of today’s dominant platforms increasingly resemble public utilities that happen to be privately owned: indispensable, deeply embedded in everyday life, and frequently resented by the very people who rely upon them.
That emotional transition, from enthusiasm to resignation, may prove more significant than any quarterly earnings report or competition case. Institutions rarely lose their influence all at once. More often, they lose something subtler first – the confidence that they understand the age they inhabit.
The great irony is that no organizations in history have possessed more information about human behavior than the leading technology companies of the twenty-first century. They know where we pause, what we share, whom we follow, when we lose interest, and what captures our attention again. They have become extraordinarily skilled at modelling behavior. But behavior is neither culture nor society. And behavior, however precisely measured, is certainly not the same thing as wisdom. That is the digital climate we have entered. The storms still occupy our attention. Meanwhile, the atmosphere has already changed.
Konstantinos Komaitis, PhD, is a veteran of developing and analysing Internet policy to ensure an open and global Internet.
Filed Under: behavior, big tech, metric not mission, open internet, optimization, platforms
DHS’s Nevada ‘Illegal Voter’ Claims Have Dropped From 16,000 To 200 To ZERO In Less Than A Month
from the bullshit-goes-unbuttressed dept
Donald Trump has been claiming for years that all elections he hasn’t (or won’t) win are rigged. This also goes for mid-terms — especially this one — where Republicans seem likely to lose seats to Democratic Party members. We won’t know if Trump will try to become President For Life until 2028, but it seems immediately clear he won’t leave anything to chance. And by “chance,” I mean a “fair democratic process unimpeded or derailed by Donald J. Trump.”
The administration continues to make wild, baseless claims about “illegal” votes being cast by non-citizens. This dovetails nicely with the administration’s “remove all the (non-white) foreigners” operations, which shifted from merely overenthusiastic to the defining feature of the second Trump administration.
This form of voter fraud is almost nonexistent. And there’s good reason to believe most of the extremely few “illegal” votes being cast are a mistake, rather than acts of people hoping to sway an election with their ill-gotten democratic processes. States have combined processes for voter registration and drivers licenses/identification cards in hopes of increasing participation in the democratic process. Unfortunately, this may lead a few non-citizens to believe obtaining an ID card also gives them the right to vote.
Instances of actual voter fraud by non-citizens are so rare as to be less than a rounding error when it comes to total votes. Nevertheless, the Department of Homeland Security (DHS) (for some fucking reason) is taking the lead in Trump’s voter suppression efforts and its principal officials continue to make outlandish claims about voter fraud while simultaneously making outlandish (and likely illegal) demands for states’ voter data.
Nevada — a state that swings towards Trump except where it matters most, the cities with the largest populations — is one of Trump’s primary targets as he advances his lies about rigged elections and widespread voter fraud.
DHS Secretary Markwayne Mullin kissed the ring and delivered the message, claiming without any facts in evidence that nearly 16,000 non-citizens were registered to vote in Nevada. Note that this is not the same thing as claiming these 16,000 people had actually ever voted. But the DHS encouraged the MAGA faithful to believe it was the same thing because Trump himself appeared to believe it was the same thing.
The state of Nevada pushed back against this claim (as well as the DHS’s demand for state voters’ identifying info). That pushback forced the DHS to walk back its previous, social media-broadcasted claims. In a telephonic meeting a few days later, DHS officials actually admitted the number of voters they suspected to be registered illegally was less than 200.
Even with this reduction, the DHS refused to turn over the information it possessed that it claimed supported its accusations against 185 Nevada voters. The state continued to press the DHS for more info, pointing out that it couldn’t do anything on its own end if it was only given “voter case” numbers generated by the DHS.
The DHS still continued to claim more than 6,000 residents might be registered illegally, but said its internal vetting had only “confirmed” the 185 “voter cases” it was willing to share with Nevada officials.
That number is even lower now. In fact, it’s one of the lowest numbers you can use when you’re counting entire human beings, as the New York Times reports:
When the Department of Homeland Security sent Nevada a list of 185 names last month that it claimed identified noncitizens who had been improperly allowed to register to vote, state election officials got to work.
On Tuesday, the state sent a strongly worded rebuke to the federal government: Every single person on the list was a lawful citizen.
“As for the 185 individuals, our records indicate that they were citizens,” Greg D. Ott, a deputy attorney general in Nevada, wrote, according to a copy of a letter sent to Department of Homeland Security officials obtained through open records requests. “You have provided nothing of substance to suggest otherwise.”
Trump and his DHS hyped the alleged discovery of nearly 300,000 “illegal” voters across four states where Trump did less business than he had hoped during the past two elections. What’s been exposed in Nevada is sure to be echoed in other states the administration claims are letting non-citizens vote.
It took less than a month to whittle down the DHS’s bullshit claims from nearly-16,000 to zero. This won’t be an anomaly. Voter fraud by non-citizens happens roughly 0% of the time anywhere during any election. Just like every other criminal act, immigrants commit voter fraud far less often than actual citizens do.
The administration is throwing people, time, and money at this attempt to suppress voter participation and it still can’t come up with anything that demonstrates it’s worth the resources being expended. But proving this isn’t the point. Generating a culture of distrust in local elections is the real game, which will allow Trump to pretend any election result he doesn’t like was the result of criminal activity.
Filed Under: dhs, ice, liars, nevada, rigging elections, trump administration, uscis, voter fraud, voter suppression
US Human Rights Record Plummets From An Already Dismal Starting Point
from the land-of-the-not-so-free-after-all dept
This article is republished from The Conversation under a Creative Commons license. Read the original article.
In the lead-up to the 2026 midterm elections, how is the U.S. government performing in terms of human rights? Listening to politicians, it can be difficult to know.
On one hand, some have pointed to a number of potential human rights abuses committed by the U.S. government inside and outside its borders in the past two years.
On the other hand, while marking the United States’ 250th anniversary this year, President Donald Trump called the country “the home of freedom,” “the land of liberty” and “the most extraordinary, most exceptional, most incredible nation ever to exist on the face of the earth.”
“We’re doing better now than we’ve ever done before,” he said.
So which is it?
According to many of the human rights data sources available, the U.S. lags behind most other high-income democracies – countries that are both members of the Organisation for Economic Co-operation and Development and classified as high income by the World Bank. The U.S.’s poor performance spans civil and political rights, such as the right to be free from torture and the right to political participation, as well as economic and social rights, including the rights to health and food.
In fact, the U.S. has been a relatively poor performer across many of these rights for quite some time, regardless of the political party in power.
Even accounting for the U.S.’s prior poor practices, however, new data from the Human Rights Measurement Initiative shows that in 2025 the country’s civil and political rights performance deteriorated even further.
As a human rights scholar who has been measuring human rights practices for 20 years – and as one of HRMI’s co-founders – I believe this data raises serious questions about the United States’ commitment to human rights.
How are rights practices measured?
Governments rarely publish accurate information on their own human rights abuses, so the task of measuring them typically falls to advocates and scholars.
Historically, most international measures of civil and political rights have depended on public reports from the U.S. State Department, Amnesty International, Human Rights Watch or news organizations.
However, these reports cannot cover every human rights violation that experts know about, especially in countries with low government transparency or high levels of repression.
Thus, data produced using those sources will show fewer abuses than those that have actually occurred and will be more accurate about some places than others. Further, the Trump administration has changed one of the most important sources of human rights information, the U.S. State Department’s Country Reports on Human Rights Practices, to contain more bias in favor of the administration’s foreign allies and less information on human rights abuse overall.
Since 2017, HRMI has used a different approach. Recognizing the problems with public reports, we collect information directly from researchers for human rights organizations, human rights lawyers and others, asking them about governments’ practices across several rights and using a statistical model to assess their information and ensure that the resulting metrics are comparable across countries. The result is a robust dataset that includes information previously missing from public reports.
Today, HRMI’s civil and political rights dataset covers 62 countries, with more added every year.
How does the US compare?
According to the Human Rights Scores dataset, which uses publicly available information to score more than 190 countries on physical integrity rights – such as the rights to be free from torture, killing, forced disappearance and political imprisonment – the U.S. never ranked higher than 81st from 2001 to 2021 or higher than 109th from 2010 to 2021.
Indeed, from 2003 to 2021, the United States’ score was either the worst or the second-worst every year among high-income OECD member states, which include the United Kingdom, Japan and New Zealand.
The Human Rights Scores dataset has not been updated past 2021, but HRMI’s data collected since that time suggests that the pattern has continued.
Among the six high-income OECD countries on which HRMI currently collects data, the U.S. has been the worst performer for physical integrity rights since 2017.
It has also frequently been rated as the worst performer on “empowerment rights,” or the rights to assembly, association, opinion, expression, political participation, religion and belief.
What changed in 2025?
Even starting from this poor position, the U.S.’s civil and political rights practices dramatically worsened in 2025. HRMI measures nine civil and political rights, and the U.S. significantly worsened on eight of them.
These were not small changes. Measured against every significant change HRMI has recorded since 2017, the U.S. showed the largest decline ever observed for the right to be free from forced disappearance and for the right to freedom of religion and belief.
The U.S. showed the second-largest decline ever observed for the right to be free from arbitrary or political arrest and for the right to opinion and expression.
The U.S. also showed major declines in the rights to be free from torture and extrajudicial killing, and the right to political participation – that is, the right to participate in one’s government via free and fair elections, to take part in public affairs and to have access to public service. U.S. scores likewise fell in the right to assembly and association, which are, respectively, the rights to peacefully gather in public or private and to form or join groups and organizations.
What caused the scores to fall?
Human rights experts raised several issues that led to these declines. For instance, the U.S.’s worsened scores for the rights to be free from arbitrary detention, disappearance and torture were tied to the expanded use of practices such as incommunicado detention and warrantless arrest by ICE and other immigration-related agencies.
While several killings occurred during ICE operations in 2025, the worsened extrajudicial killing score was also tied to the country’s use of airstrikes against boats in the Pacific and Caribbean, which caused the deaths of at least 123 people in 2025 alone.
Other changes enacted in 2025 contributed to large declines in the U.S.’s rights scores, including crackdowns on anti-ICE demonstrations and government pressure on universities, including limits on free speech and class subjects.
Last year also saw threats to television networks over political expression, restrictions on mail-in voting, state measures targeting American Islamic organizations by labeling them “foreign terrorist organizations” and an executive order focused on eradicating “anti-Christian bias.”
While current HRMI data does not yet cover 2026, early indicators are not promising, with continued abuses by ICE, further killings in international waters and allegations of attempts to use government powers to punish critics.
K. Chad Clay is the Director of the Center for the Study of Global Issues (GLOBIS) and an Associate Professor of International Affairs at the University of Georgia.
Filed Under: human rights
Companies: oecd
Ctrl-Alt-Speech: Going To The Ballot Bot
from the ctrl-alt-speech dept
Ctrl-Alt-Speech is a weekly podcast about the latest news in online speech, from Mike Masnick and Everything in Moderation‘s Ben Whitelaw.
Subscribe now on Apple Podcasts, Overcast, Spotify, Pocket Casts, YouTube, or your podcast app of choice — or go straight to the RSS feed. To get extended episodes with additional coverage, support us on Patreon.
In this week’s episode, Ben is joined by Katie Harbath, Meta’s former director of public policy for global elections and now founder of technology policy firm Anchor Change. She is also the author of Disrupting Politics, out on Tuesday. Together they discuss:
- TSE (Brazil Election Commission) clarifies rules on deepfakes in the 2026 elections (O Diário do Paraná)
- AI tools keep recommending candidates, monitoring finds (Tribuna Livre Brasil)
- On Eve of Brazil Elections, Meta Content Moderation Decisions Raise Questions (Tech Policy Press)
- Deepfakes, foreign-run accounts and unauthorised ads, new research exposes Meta’s Brazil election ad machine (Ekō)
- Social media ban for under-16s to come into effect next March, Culture Secretary confirms (LBC)
- Trump says top tech firms have signed accord to ‘self-police’ AI development (NPR)
And in the extended episode for Patreon supporters, they cover:
- Meta-Led Anti-Terrorism Group Faces Mass Resignation of Expert Advisers (WIRED)
- Meta to report child abuse material directly to Indian authorities (BBC)
- Some NPR podcasts started getting mysterious comments on Spotify (Instagram)
- Content Moderation Knowledge Sharing Shouldn’t Be A Backdoor To Cross-Platform Censorship (Techdirt)
- J.M. Berger on resigning from GIFCT’s Independent Advisory Committee (Bluesky)
- Mark: A journey to the heart of history’s largest nonterritorial empire (Colossus)
Our fun links this week include legal-walls.net for all your legal graffiti questions and, courtesy of Katmai National Park in Alaska, Fat Bear Week.
Follow us on Instagram, YouTube, and Bluesky for video clips from this week’s episode!
If you’re already a Patreon supporter, you can get the extended episode on Patreon.
Filed Under: ai, artificial intelligence, brazil, content moderation, gifct, mark zuckerberg, ncmec, trust and safety
Companies: meta
Bricks & Minifigs Really Doesn’t Want You To Know About The BAM Map, Sues To Unmask Whoever Runs It
from the hit-BAM-hollers dept
A few days ago I wrote about the anti-SLAPP motion filed by “Reckless” Ben Schneider against Bricks & Minifigs (BAM Franchising) over the ridiculous RICO suit they filed against him for his efforts to get BAM to pay back the money owed to Bryan Mansell. The original suit was based on the laughable claim that Ben and Mansell were involved in a conspiracy to extort money from BAM — a claim that is undermined by the fact that BAM then turned around and settled with Mansell, returning some of his Lego sets and cutting him a check (while still insisting it did nothing wrong, of course. For what it’s worth, even after the settlement, Mansell says BAM returned only about 35 of the 387 sets its own records showed as unsold, and seven of those weren’t even his, which makes me wonder what exactly the settlement said and how that didn’t get worked out ahead of time).
In researching that previous article, I came across the site used for that last link, The BAM Map, which is an invaluable journalistic goldmine, in that whoever is behind it is cataloging in great detail basically every sketchy or nefarious thing that BAM is doing.
But digging through that site, I discovered that BAM (and the McNeff family behind it) appear to have launched a new censorial bit of SLAPP litigation and general lawfare against The BAM Map as well.
Sites that catalog corporate shenanigans aren’t always the most trustworthy, but The BAM Map goes out of its way to detail and post their evidence, and flag whether particular claims are merely asserted as opposed to proven as fact. It’s actually quite impressive.
Anyway, it appears that the lawfare against the BAM Map started with some highly questionable DMCA takedown claims, aka the coward’s censorship tool. BAM first issued a DMCA notice to The BAM Map’s host because the site had posted an internal BAM document that analyzed the impact on their stores from Reckless Ben’s reporting on how BAM refused to abide by the consignment agreement they had taken over. While it is an internal document, so long as The BAM Map itself broke no laws in obtaining the document, they had every legal right as journalists to publish it. The idea that it’s copyright infringement is laughable on its face. That’s not what copyright is for, and also there’s no indication that BAM ever even registered the copyright, which it would need to do before it could even sue for copyright infringement. Indeed, since any registration would then come after the alleged infringement, statutory damages are also completely off the table, leaving only “actual damages” from… being unable to license their internal documents?
In other words, the DMCA takedown is bullshit.
And then they did it again. Just last week they filed a second DMCA notice, once again demanding that BAM’s internal documents that The BAM Map published be removed (even though, as far as I can tell, the first notice had already gotten them taken down).
Let’s be clear: this is abuse of copyright law by BAM. DMCA notices are supposed to be for stopping actual infringement: someone distributing your work in a way the law doesn’t allow. But publishing documents as part of news reporting is a strong fair use case (especially for factual material like sales data and PR talking points). Not to mention, if you could just DMCA any internal corporate files that journalists received from whistleblowers, you would basically end whistleblowing completely. Using copyright this way isn’t just abuse of the DMCA, it’s an attack on First Amendment-protected reporting.
But BAM didn’t stop there. It then filed a John Doe suit against the site’s anonymous operators, once again using Dentons, the massive global law firm that it used to file the SLAPP suit against Schneider and Mansell. Notably, while the lawsuit mentions the DMCA notice and the existence of “copyrighted materials,” it does not include any copyright claims. Instead, it claims that Google Drive audit logs show someone with “franchise-level access” downloaded the files. So, according to the lawsuit, whoever runs The BAM Map is likely a current or former franchisee (or some other insider) who is breaching their contract with BAM and misappropriating trade secrets under both Utah’s trade secrets law and the federal Defend Trade Secrets Act. For what it’s worth, the operators of the site claim that the knowledge base that is central to the claims was not only accessible to those who had specific permissions but rather “was open to anyone on September 26th” and that it was only earlier this week that BAM locked it down.
This case is marginally stronger than the lawsuit against Reckless Ben, but only barely. Even if the documents are coming from an insider, it’s hard to see how BAM can credibly claim that its crisis PR talking points about Reckless Ben qualify as “trade secrets” that “derive economic value” from being kept secret. The entire point of this lawsuit just seems focused on creating massive chilling effects for whoever is operating the site to silence them. To me, that makes it a pretty clear SLAPP suit.
The key point of the lawsuit is BAM trying to unmask who is behind The BAM Map, and to that end, after filing the lawsuit, BAM got a magistrate judge to approve expedited discovery, authorizing subpoenas to Vercel and GoDaddy for subscriber records and IP logs to identify who is behind the BAM Map, in order to properly serve them and include them in the lawsuits. While magistrates granting subpoenas like this isn’t uncommon, it’s a bit worrisome that it was done with no supporting declaration, and the order has no notice provision or period to object.
Under the First Amendment, you have a right to speak anonymously, but it’s not absolute. The way this generally works is that a court will allow the subpoena to issue, and then (one hopes) the companies on the receiving end (GoDaddy and Vercel) alert the site’s operators so they can move to quash. There are a couple of different standards that courts use to determine when it’s okay to unmask an anonymous speaker, with Dendrite and Cahill being the most common. In the past, the federal courts in Utah have recognized the Dendrite test in quashing a subpoena when Koch Industries tried to unmask an anonymous critic.
Hopefully the anonymous operators of The BAM Map have access to lawyers who understand this, and can step up in response to the subpoenas, in order to protect their anonymity. At the very least, they’ve started a GoFundMe for legal support.
Bricks & Minifigs’ continued efforts to abuse the legal system to suppress reporting by critics says an awful lot about the McNeff family, both in how they run their business and in how they treat the people who call out their practices. It’s an attack on First Amendment protected speech and an attempt to create a chilling effect to scare off future reports or whistleblowers.
And, finally, shame on Dentons. A giant global law firm putting its name on this is just embarrassing.
Filed Under: 1st amendment, ammon mcneff, copyright, dmca, free speech, journalism, reporting, slapp, slapp suit, takedowns, talking points, utah, whistleblowing
Companies: bam franchising, bricks & minifigs, dentons, the bam map
Daily Deal: The Complete Raspberry Pi And Alexa A-Z Bundle
from the good-deals-on-cool-stuff dept
Learn Raspberry Pi and start building Amazon Alexa projects with The Complete Raspberry Pi and Alexa A-Z Bundle. Catered for all levels, these project-based courses will get you up and running with the basics of Pi, before escalating to full projects. Before you know it, you’ll be building a gaming system to play old Nintendo, Sega, and PlayStation games and a personal digital assistant using the Google Assistant API. You will also learn how to build Alexa Skills that will run on any Amazon Echo device to voice control anything in your home, and how to build your own Echo clone. The bundle is on sale for $30.
Note: The Techdirt Deals Store is powered and curated by StackSocial. A portion of all sales from Techdirt Deals helps support Techdirt. The products featured do not reflect endorsements by our editorial team.
Filed Under: daily deal
Flock Promises Even More Ineffectual Features In Response To Congressional Demands
from the will-this-do dept
Flock is in the middle of a long losing streak. It’s one thing to lose hearts and minds via PR work. It’s quite another to lose actual dollars because you’ve lost a ton of hearts and minds by being on the losing end of mainstream media coverage for most the past two years.
If it bleeds, it leads, as every cop PR team knows. The problem for Flock is that it leads so often it’s bleeding customers, which has forced the company to finally start pitching some fixes rather than just pitching fits when confronted with (legitimate) criticism.
No longer able to ignore the bylined writing on the collective wall of US journalism, Flock stepped up to say it would do the bare minimum to stave off even more reporting on cops using its systems to stalk ex-wives, ex-significant others, the current partners of their ex-wives/significant others, protesters, critics, and anyone else officers wanted to take a peek at without having any legitimate law enforcement-related reason to do so.
In response to nationwide criticism, Flock Safety promised to make some optional features on-by-default. This announcement — paired with the outrageous claim that only Flock cared about police accountability — only raised more questions. Dropping the retention period to seven days made us wonder why that hadn’t always been on by default — especially when the company is collecting 20 billion plate reads per month.
Requiring case codes for searches should always have been the default, but only now seemed to be a thing Flock wanted to implement. The same thing goes for the newly announced “Audit Assistance,” which would allow interested agencies more control and oversight of searches performed by officers. Once again, we were forced to ask “why the fuck hasn’t this always been in place?”
The problems are no longer local. I mean, it’s still local agencies divesting themselves of Flock cameras and contracts, but even the those on Capitol Hill are starting to make noise about Flock, its pervasive surveillance network, and the apparently continual abuse by people in law enforcement who have access to Flock databases.
This congressional noise even includes Josh Hawley, who is perhaps best known for encouraging violent insurrectionists to continue with their insurrection — one that targeted his place of business but presumably wouldn’t target him personally.
So, it’s with that lump of presidentially pardoned salt that we take in this recent news: Hawley is all heated up about Flock and constant abuse of its systems by law enforcement. And Flock, for its part, is willing to make soothing noise in the general direction of Washington DC. Here’s the Wall Street Journal with the underwhelming details:
The Atlanta-based company has made changes to the information users provide when explaining their searches, Flock’s chief legal officer said in a letter to Sen. Josh Hawley, a Republican from Missouri. Flock has said its cameras scan about 20 billion plates a month. While users have always been required to justify their searches of the company’s database, the information was entered “in a free-text field,” yielding answers that weren’t always specific, the company said in the letter viewed by The Wall Street Journal.
Flock has now added a drop-down menu with standardized categories modeled after the Federal Bureau of Investigation’s data-reporting system. Law-enforcement searches will require a case number in order to conduct a search by the end of the year, although the system can be bypassed in an emergency, Flock said.
The only people who actually think a lack of accountability can be addressed — much less fixed — by a “drop-down menu” either work for Flock or the cop shops employing stalkers that just haven’t been caught yet.
Adding a case number field doesn’t really solve anything either, not if cops can bypass this supposed roadblock by typing anything from “lmao” to “jkjkjkjkjkjkj” in the field before proceeding with their illicit searches.
And it’s at this point that I start to side with Flock. A case number field only works if Flock knows which case numbers are valid (or, at the very least, follow case number nomenclature). Flock can’t because it’s something stapled on top of police stuff, rather than being intertwined from its genesis.
A drop-down menu may help classify searches should anyone decide to audit them, but all this really does is apply a veneer of legitimacy to illegitimate searches.
Flock’s failures to place location restrictions on searches (rather than just allow any local cop to search the entirety of Flock’s network) and limit data retention periods are on the company. That much it can definitely control. The other stuff it’s doing now means about as much as doing nothing, since the default settings can easily be overridden and most of the restraints are meaningless if law enforcement supervisors aren’t regularly checking search logs for anomalies like thousands of non-justified searches targeting certain plates or cops justifying their searches by hammering out some random letters until the “case number” field turns green.
Josh Hawley’s personal interest may be to ensure future pro-Trump insurrectionists arrive in DC without leaving a permanent Flock record behind. Then again, he may actually care about the issue he currently cares about, but I generally don’t trust anyone who thinks the civil liberties you should have access to depend largely on who you’ve voted for.
As for Flock, the best way out of this mess is probably the least profitable. It should be proactively terminating contracts any place officers have been criminally charged for abusing the system. It should also apply geofences to searches and require law enforcement agencies to enter agreements with agencies outside of their geofence to perform expanded searches. Finally, it should create contractual language that requires case management software cross-referencing. If cops aren’t willing to give Flock the information its needs to block unjustified searches, they shouldn’t be allowed to access Flock’s network and databases.
Tough times call for tough measures. Unfortunately, the last people willing to be tough are tech companies with shareholders to satiate with year-over-year growth and the law enforcement agencies which have steadfastly shown they’re unwilling to police themselves.
Filed Under: acab, alpr, josh hawley, location tracking, police abuse, police misconduct, privacy, stalking, surveillance, surveillance abuse
Companies: flock, flock safety
No comments:
Post a Comment