Criminals Gotta Crime

And by criminals, I mean the criminal enterprise formerly known as Facebook™, who just got caught surreptitiously installing facial recognition software on their smart glasses.

Seeing as how they did not announce this to their users, it would have been a selling point, it is not unreasonable to assume that this was for Meta to spy on its users.

Hell, everything in Meta is about spying on its users.

One day after WIRED revealed that Meta had quietly embedded an unreleased face-recognition system into an app installed on more than 50 million phones, the company removed it, according to a WIRED analysis of the latest version’s code.

The most recent version of Meta AI, a companion app for its line of smart glasses, strips out the unactivated software components that powered the system Meta internally called NameTag. The version published the day of WIRED’s report included several code libraries explicitly named for face recognition. Friday’s release includes none of them.

Andy Stone, Meta's vice president of communications, told WIRED on Monday that the feature is purely exploratory, adding: “No final decision has been made on what to do here, if anything.”

On Thursday, WIRED reported that Meta had quietly integrated substantial portions of the NameTag system into the Meta AI app. Though never publicly enabled, the feature was designed to convert faces captured by the glasses into unique biometric signatures, commonly known as faceprints, and compare them against a database of faceprints stored on the user's device. WIRED also found that faces the system failed to recognize were cropped, indexed, and stored locally for future processing.

Well, this seems completely aboveboard and not suspicious at all.

Not Enough Bullets

Go old school?DOGE and DHS tried to get the Social Security Administration to declare that 2.7 million people were dead in an attempt to harass immigrants.

Everyone involved in this should never see the sky as a free man.

The Trump administration had plans to classify 2.7 million living people — including some U.S. citizens and lawful permanent residents — as dead as part of its immigration enforcement efforts, according to a former senior Social Security executive. 

The previously unreported plan, which the Social Security Administration said was not carried out, would have used one of the government’s most consequential identity databases to effectively erase people from the financial system, potentially cutting them off from wages, banking, government benefits and other services.

Jeremiah Schofield, who worked at Social Security for 25 years and helped lead the agency’s IT modernization efforts before leaving in October, said he refused to help implement the plan after agency lawyers warned that falsely marking living people as dead could violate federal law. Schofield said he realized the plan’s possible intent — to intimidate and worsen the finances of immigrants — as well as its potential unlawfulness after taking a sample of people from the 2.7 million and discovering they were all alive. Some were U.S. citizens, lawful permanent residents, teenagers and senior citizens, including one widow who was a legal permanent resident receiving survivor benefits.


Schofield has provided details on the plan in a 49-page whistleblower disclosure to Sen. Elizabeth Warren (D-Mass.), who is on the Senate Finance Committee, and Sen. Richard Blumenthal (D-Conn.), the ranking member on the Senate Permanent Subcommittee on Investigations. The disclosure was reviewed by The Washington Post, and it offers the most detailed account yet of how officials from Elon Musk’s U.S. DOGE Service sought to use Social Security data in service of President Donald Trump’s immigration crackdown.

In an interview with The Post, Schofield said he is speaking publicly for the first time because he believes Americans need to understand how government data can be misused and, in some cases, already has been.

………

Schofield’s whistleblower complaint describes a tumultuous period inside Social Security, as career officials questioned the legality of such efforts and watched DOGE officials gain access to some of the government’s most sensitive databases. In one meeting, Schofield said, a DOGE official working with the Department of Homeland Security described the goal of declaring 2.7 million living people dead: making immigrants so miserable that they self-deported or went to Social Security offices for help, where they could be arrested.

I'm beginning to think that if the DOGE boy's parents had decided to drown them all at birth, (I'm looking at YOU, Mae Musk) the world would be a far better place. 

Well, Trash Belongs in Trash Bags

As the backlash against Flock AI driven license plate readers grow, more and more municipalities are canceling their contracts.

They find two problems, canceling the contract is difficult because of their Byzantine nature, and even when canceled, Flock does not remove them, so the cameras remain on their poles, where anyone with a Flock contract (think ICE) can continue to use them.

Some towns have an interesting way of dealing with this, they are sending out government workers to cover the cameras with trash bags. (alternate link)

The city of Dayton, Ohio has covered its Flock automated license plate reader cameras with black trash bags in part because police there are unsure whether the cameras are still active and the city also doesn’t seem to know whether it is allowed to take the cameras down. The move comes after months of resident outrage, a scandal in which the city was sharing Flock camera data for immigration enforcement apparently on accident, and a $30,000 audit into how the cameras are being used. 

Joe Parlette, the deputy city manager of Dayton, said at a city commission meeting last week that the “Dayton Police Department agreed to work with Public Works to put bags over the cameras” as a stop-gap measure until Flock cameras could be removed entirely. I spoke to multiple people in Dayton who said they had seen bagged cameras in the last few days. The Dayton Daily News first reported on the baggings.

Dayton is not the first city to cover its Flock cameras with trash bags because they can’t figure out how to immediately terminate the use of the cameras. Late last year, the city of Evanston, Illinois also covered its cameras with trash bags while it was waiting for the company to remove them from the city. Cities around the country have been reconsidering their relationship with the surveillance company after reporting from 404 Media and local news outlets that showed data from the cameras was making its way to Immigration and Customs Enforcement through Flock’s national camera network.

Flock cameras are usually solar powered, so once installed, it costs more to take them down than it does to leave them there, particularly if the data can be sold to someone else.

Garbage bags are a decent solution to this dynamic. 

F%$# Flock

An amendment has been proposed to a major transportation bill that would ban the use of automated license plate readers for any purpose beyond toll booths.

It appears that the anti-Flock movement has friends in Congress. 

US lawmakers plan to introduce an amendment Thursday at a House committee markup hearing that would prohibit any recipient of federal highway funding from using automated license plate readers for any purpose other than tolling—a sweeping restriction that, if adopted, would bring an immediate end to state and local ALPR programs across the United States.

The amendment, obtained first by WIRED, is sponsored by Representative Scott Perry, a Pennsylvania Republican and Freedom Caucus member, and Representative Jesús “Chuy” García, an Illinois progressive whose state has become a flash point in the national fight over ALPR misuse.

The House Transportation and Infrastructure Committee will mark up the underlying bill—a $580 billion, five-year reauthorization of federal surface transportation programs—at 10 am ET on Thursday.

Neither Perry nor García's offices immediately responded to WIRED's request for comment.

The amendment runs a single sentence: “A recipient of assistance under Title 23, United States Code, may not use automated license plate readers for any purpose other than tolling.”

Here's hoping it passes. 

This is a Feature, Not a Bug

Is anyone surprised that police are using automated license plate readers to stalk their former romantic partners?

I'm not.

Police have routinely done this with other data sets going back decades.

In fact, I would argue that giving cops the ability to do this is a large part of why police like Flock cameras and their ilk so much. 

The cops never change. Only the tech toys do.

That’s the upshot of this report from the Institute for Justice, which has been tracking what cops have been tracking now that they have always-on access to massive networks of security cameras, including Flock Safety’s controversial offerings, which also include automatic license plate readers (ALPR).

The proliferation of police surveillance has led to repeated abuse. One shockingly common form: police officers using ALPR camera networks to keep tabs on their romantic interests, including current partners, exes, and even strangers who unwittingly caught their eye in public.

An Institute for Justice review of media reports has identified at least 14 cases nationwide of officers allegedly abusing ALPR data this way, with the bulk of those incidents happening since 2024.

This is the same stuff that cops have been doing for years. Access to criminal databases, drivers license info, and anything else swept up by government entities has resulted in numerous cases of abuse.

Maybe what cops need is not better spying tech, maybe what cops need is better people on the force.

There's an App for That

There is an app on the Android Play Store that detects whether or not there is a Meta glasshole in your immediate vicinity potentially recording you.

I approve.

Worried that someone wearing Meta's snooping spyware goggles could be creeping up on you? Android users now have access to an app that can warn them if someone is wearing such smart glasses in their vicinity by using Bluetooth.

Last week, Yves Jeanrenaud, a deputy professor at Darmstadt University of Applied Sciences in Germany, published Nearby Glasses, an Android app that scans Bluetooth Low Energy (BLE) advertising data for manufacturer identifiers associated with certain smart glasses, including Ray-Ban Meta AI Glasses.

"This app notifies you when smart glasses are nearby," Jeanrenaud explained in the project's GitHub repo. "It uses company [identifiers] in the Bluetooth data sent out by these [devices]."

In a LinkedIn post on Tuesday, he elaborated on how the software works.

"Bluetooth devices broadcast small advertising packets," he wrote. "Even though MAC addresses (identifying a particular device) and service UUIDs (identifying what they are doing) are randomized, manufacturer company IDs in BLE advertising frames are mandatory and immutable."

………

Pointing out other abuses associated with Meta's AI glasses, and Meta's reported plan to add facial recognition to its glasses, Jeanrenaud said, "This is not a perfect solution, but I hope it's useful for someone. Until consent and privacy are treated seriously in wearable tech, I hope this tool helps someone feel a little more safe."

Seeing as how the app is open source, it seems to me that you could add some features, things like projecting Goatse onto the lenses.  (If you do not know what Goatse is, don't click through)

Headline of the Day

Palantir Workers Are Finally Noticing The Skulls On Their Caps

—Techdirt, referencing Mitchell & Webb's finest comedic bits.

Are We the Baddies?

I do not know what took Palantir employees so long to realize that they are evil minions.

There’s a famous Mitchell & Webb sketch where two SS officers, mid-conversation on the Eastern Front, suddenly notice something troubling about their uniforms. “Hans,” one asks, peering at his cap, “are we the baddies?” The skulls had been there the whole time. The skulls are kind of a giveaway. But it took a while for the question to surface. You’ve probably seen it:

………

I thought about that sketch reading Wired’s reporting on the internal turmoil at Palantir, where both current and former employees are starting to ask that question of their own work:

Around that time, two former employees reconnected by phone. Right as they picked up the call, one of them asked, “Are you tracking Palantir’s descent into fascism?”

“That was their greeting,” the other former employee says. “There’s this feeling not of ‘Oh, this is unpopular and hard,’ but ‘This feels wrong.’”

Two weeks ago, we wrote about Palantir going mask-off for fascism, specifically about CEO Alex Karp’s company posting a 22-point manifesto that included some genuinely ugly stuff about how “certain cultures” are “regressive and harmful” and how pluralism is a “shallow temptation.” I argued that this kind of public ideological positioning was both morally bankrupt and strategically suicidal. The moral bankruptcy part should be obvious (if it’s not, go do some soul-searching). But doing so at a time when American-style fascism is historically unpopular basically everywhere, including within the US, just seems like you’ve bet on the losing team at a time when it’s clear they have no chance of coming back to win.

To Quote Upton Sinclair, "It is difficult to get a man to understand something, when his salary depends on his not understanding it." 

Just as an FYI

I opted out of facial recognition at the TSA check in.

The entire exchange took perhaps 5 seconds. 

Just to be clear, this has nothing to do with the Libyan hush money allegations.

Eric Arthur Blair ⃰ Is Spinning in His Grave Fast Enough to Power All of Dubuque, Iowa

At the latest reporting of (obscene) profits for the cyber-stalking as a service company Palantir, its CEO declared the company a, "Guardian of Americans' rights."

Mad as a Panrovian monk, he is.

Palantir had a whopper of a Q4, showing accelerating revenue growth, beating Wall Street's profit estimates, and enjoying a share price jump of as much as 11% during pre-market trading on Tuesday before coming back down to earth.

At such a triumphal moment, it was striking that the company was forced on the defensive, not least for its involvement with the controversial US agency Immigration and Customs Enforcement (ICE), which began in 2011. ICE has drawn criticism for its activities in Minnesota over the last month after its agents shot and killed two American citizens while bystanders captured the events on video.

As it announced its booming financial results, Palantir was prepared. CEO Alex Karp told CNBC: "If you are critical of ICE, you should be out there protesting for more Palantir. Our product, actually, in its core, requires people to conform with Fourth Amendment data protection."

The technical term for the above term is bullsh%$.  (I miss January so much right now)

………

On an earnings call, Karp said the company's software inherently instills the protection of individual rights.

"From the beginning, we have stuck to our very strong values of expanding what we believe is the noble side of the West ... meaning domestic institutions, intelligence institutions [are] essentially taking an incatenation of the Fourth Amendment, which is completely represented by our pipelining, Foundry, and impregnating institutions with it so that every institution that uses our product is doing it within conformity of the law and the ethics of America," he said.

………

"The construction of such a platform, one that reflects our ethical commitments, should, of course, be a rallying cry for progressives and critical thinkers across the political spectrum who profess to be interested in advancing the values of the Fourth Amendment," he said.

To correct the late, great, Douglas Adams, the marketing division of the Sirius Cybernetics Corporation as “a bunch of mindless jerks will not be the first against the wall when the revolution comes, they have to wait in line until this guy is done.

 

^*Better known by his pen name, George Orwell.

Of Course They Did

It has been revealed that Elon Musk's DOGE was illegally transferring personal data from the Social Security (SSA) database to right wing vote suppression groups. 

After months of denials, the Trump administration has admitted that staffers affiliated with the Department of Government Efficiency (DOGE) misused Social Security Administration (SSA) data. In an extraordinary court filing, “NOTICE OF CORRECTIONS TO THE RECORD,” government lawyers representing the SSA revealed that in March 2025, a DOGE staffer signed an agreement to share the private data of Americans with a “political advocacy group” seeking to “overturn election results in certain States.”

SSA determined in its recent review that in March 2025, a political advocacy group contacted two members of SSA’s DOGE Team with a request to analyze state voter rolls that the advocacy group had acquired. The advocacy group’s stated aim was to find evidence of voter fraud and to overturn election results in certain States. In connection with these communications, one of the DOGE team members signed a “Voter Data Agreement,” in his capacity as an SSA employee, with the advocacy group. He sent the executed agreement to the advocacy group on March 24, 2025.

The filing says that emails “suggest that DOGE Team members could have been asked to assist the advocacy group by accessing SSA data to match to the voter rolls.”

The use of government data for political purposes is unlawful. The Hatch Act prohibits a federal employee from using “his official authority or influence for the purpose of interfering with or affecting the result of an election.” In the filing, government lawyers say the DOGE employees involved were referred to the U.S. Office of Special Counsel for possible Hatch Act violations.

………

While the court filing does not name the “political advocacy group” coordinating with DOGE, there is evidence pointing to True the Vote, a right-wing group with a history of pushing false claims of election fraud. True the Vote published “An Appeal to DOGE” in March 2025, the same month that the DOGE staffer signed an agreement with the political group. In the open letter, published on the group’s website on March 5, True the Vote encouraged DOGE to investigate the country’s voter registration system. At the top of the letter, Democracy Docket notes, True the Vote Founder Catherine Engelbrecht wrote, “We’ve received word that this message is being carried forward.”

At the earliest opportunity, Elon and his DOGE thugs need to be arrested put in the dock.

The Hero We Need

As you may be aware, Glassholes are back, but this time, it's the criminal enterprise formerly known as Facebook™ that is selling spy spectacles masquerading as a VR headset.

Some influencer wannabee was making funny noises on the New York City Subway and filming reactions for a TikTok, and a woman snatched the Ray-Bans off of his face and broke them in two.

Good on her, and bad on the skeevy cretin who was filming people without their knowledge:

A New York subway rider is going viral after a TikToker accused her of breaking his Meta AI glasses, a moment that instantly made her a folk hero among privacy-conscious internet users.

The eyewear, which can discreetly record video, has been criticized as a creeping surveillance threat. Many online viewers argued that the woman simply did what others have only joked about doing.

While the TikToker insists the incident was unprovoked and filed a police report, the internet has already taken her side, celebrating her as the anti-AI vigilante of their dreams.

………

In the description on another video, he claims he did nothing to provoke this woman.

“She was like 12 feet away from me and I wasn’t addressing her, I was just making sounds that I and others thought was hilarious for a video to post,” he wrote. “There were so many different routes one could take instead of breaking someone else’s $300 glasses such as asking nicely to not post it or blur my face.”

Ummm....No.

It's your job to get affirmative permission before filming people in an attempt to create a funny video.

Alan Funt you ain't.  (Also, he got waivers signed for anything that he showed) 

Here’s Hoping This Takes Them Down

A judge has ruled that the output from Flock spy cameras are public records which must be made available to the public.

Hopefully this will turn over rocks and to see the corruption beneath. 

A judge in Washington has ruled that police images taken by Flock’s AI license plate-scanning cameras are public records that can be requested as part of normal public records requests. The decision highlights the sheer volume of the technology-fueled surveillance state in the United States, and shows that at least in some cases, police cannot withhold the data collected by its surveillance systems.

In a ruling last week, Judge Elizabeth Neidzwski ruled that “the Flock images generated by the Flock cameras located in Stanwood and Sedro-Wooley [Washington] are public records under the Washington State Public Records Act,” that they are “not exempt from disclosure,” and that “an agency does not have to possess a record for that record to be subject to the Public Records Act.”

She further found that “Flock camera images are created and used to further a governmental purpose” and that the images on them are public records because they were paid for by taxpayers. Despite this, the records that were requested as part of the case will not be released because the city automatically deleted them after 30 days. Local media in Washington first reported on the case; 404 Media bought Washington State court records to report the specifics of the case in more detail.

Flock’s automated license plate reader (ALPR) cameras are used in thousands of communities around the United States. They passively take between six and 12 timestamped images of each car that passes by, allowing the company to make a detailed database of where certain cars (and by extension, people) are driving in those communities. 404 Media has reported extensively on Flock, and has highlighted that its cameras have been accessed by the Department of Homeland Security and by local police working with DHS on immigration cases. Last month, cops in Colorado used data from Flock cameras to incorrectly accuse an innocent woman of theft based on her car’s movements.

………

The case highlights the lengths that police departments and cities are willing to go to in order to prevent the release of what they incorrectly perceive to be private information owned by their surveillance vendors (in this case, Flock). Stanwood’s attorneys first argued that the records were Flock’s, not the city’s, which is clearly contradicted in the contract, which states “customer [Stanwood] shall retain whatever legally cognizable right, title, and interest in Customer Generated Data … Flock does not own and shall not sell Customer Generated Data.” The attorneys then argued that images taken by Flock cameras do not become requestable data until it is directly accessed and downloaded by the police on Flock’s customer portal: “the data existing in the cloud system … does not exist anywhere in the City’s files as a record.” The city’s lawyers also argued that Flock footage is police “intelligence information” that should be exempt from public records requests, and that “there are privacy concerns with making ALPR data accessible to the public.”

The degree to which the state security apparatus is using private vendors to conceal theoir violations of our privacy is truly appalling.

Mao Was Right About Landlords

Guess what?  Landlords now demanding their tenants' logins to their payroll systems.  (Alternate link here)

Whey want you to log in through an app called Argyle, which scrapes an enormous amount of data from your employer/payroll website.

I'm feeling charitable today, so I'll just suggest that landlords and executives at ApproveShield be arrested tried and jailed, and not, as Mao Zedong did, that the legal niceties be ignored and a bullet be put in their head.

Landlords are using a service that logs into a potential renter’s employer systems and scrapes their paystubs and other information en masse, potentially in violation of U.S. hacking laws, according to screenshots of the tool shared with 404 Media.

The screenshots highlight the intrusive methods some landlords use when screening potential tenants, taking information they may not need, or legally be entitled to, to assess a renter.

“This is a statewide consumer-finance abuse that forces renters to surrender payroll and bank logins or face homelessness,” one renter who was forced to use the tool and who saw it taking more data than was necessary for their apartment application told 404 Media. 404 Media granted the person anonymity to protect them from retaliation from their landlord or the services used.

……… 

The person said earlier this year they were verifying their income in order to start a lease at an apartment complex in Atlanta. The apartment complex used a tenant screening service called ApproveShield, the person said. The landlord required 60 days of pay history, or four pay stubs, the person said.

ApproveShield is in-part powered by a tool called Argyle, which verifies peoples’ income. It does this by having people log into their corporate employer HR services, such as Workday, and scraping information stored within. I’ve covered Argyle before, when I found it was linked to a wave of suspicious emails that offered people cash for their workplace login credentials. 

The renter said ApproveShield’s Argyle-powered widget asked them to log into their employer’s Workday. That's when they noticed something unusual.

“Argyle hijacked my live Workday session, stayed hidden from view, and downloaded every pay stub plus all W-4s back to 2024, each PDF seconds apart,” they said. “Workday audit logs show dozens of ‘Print’ events from two IPs from a MAC which I do not use,” they added, referring to a MAC address, a unique identifier assigned to each device on a network. 

Yeah, this is hacking.  Even if the accesses were approved by the renter, they would violate the terms of of service of literally every payroll system out there.

I know that jails are overcrowded, but we can let out some shoplifters and drug addicts to lock these people up forever.'

Or we can take off and nuke the site from orbit.  It's the only way to be sure. 

The Term is "Glasshole"

The folks at Gizmodo think that there may be a backlash coming against the current crop of "smart" glasses.

Gee, I wonder why people would object to folks wearing hardware which surreptitiously records everything around them.

People hated Google Glass for this reason, and they will hate these for the same reason. 

We've already had security warnings of cree[ps using Ray-Ban Meta AI glasses on campus to film students without their knowledge or consent.

I'm sure mark Zuckerberg thinks that these are neat, but Zuckerberg is a creep's creep himself.

No, just no. 

Predictable Fail

The UK has a law, the Online Safety Act of 2023, which requires all online services to age verify all of their users. 

This has created a market niche for firms which collect user data, images of their identification, and other personally identifiable information (PII) to do this.

Well, one of these verifiers has leaked over 70,000 Discord users PII.

I when this legislation was first mooted, this one of the problems that opponents cited:

Communication platform provider Discord has admitted that around 70,000 users had their government IDs stolen as part of its recent data breach.

The breach, which Discord insists occurred at an unidentified third-party customer service provider, involved government ID scans that users upload to verify their age.

Some countries have introduced legislation requiring platforms to vet users and ensure only those that meet a certain age threshold are allowed access. The most recent example is the UK with its Online Safety Act.

The UK says in-scope platforms, like Discord, must implement mechanisms to verify their users' ages "without collecting or storing personal data, unless absolutely necessary." 

Discord's help article on how users can verify their age details two methods.

The first sees users take a photo of themselves holding a photo ID and a piece of paper with their username on it. This process is completed through Discord itself.

This was foreseeable, and foreseen.

The Tory passed law could be changed to make it less risky and draconian, but draconian infringements on personal rights are kind of a brand for Keir Starmer,  so I would expect any adjustments to make it worse.

Did Not Pass the Laugh Test

Verizon has been fined for selling its customers location data without the knowledge of consent of its customers.

Now a federal judge has called bullsh%$ on the telco's claims that it was legal.

Good. 

Verizon lost an attempt to overturn a $46.9 million fine for selling customer location data without its users' consent. The US Court of Appeals for the 2nd Circuit rejected Verizon's challenge in a ruling issued today.

The Federal Communications Commission fined the three major carriers last year for violations revealed in 2018. The companies sued the FCC in three different courts, with varying results.

AT&T beat the FCC in the reliably conservative US Court of Appeals for the 5th Circuit, while T-Mobile lost in the District of Columbia Circuit. Although FCC Chairman Brendan Carr voted against the fine last year, when the commission had a Democratic majority, his FCC urged the courts to uphold the Biden-era decisions.

………

"We disagree [with Verizon]," the 2nd Circuit ruling said. "The customer data at issue plainly qualifies as customer proprietary network information, triggering the Communication Act's privacy protections. And the forfeiture order both soundly imposed liability and remained within the strictures of the penalty cap. Nothing about the Commission's proceedings, moreover, transgressed the Seventh Amendment's jury trial guarantee. Indeed, Verizon had, and chose to forgo, the opportunity for a jury trial in federal court. Thus, we DENY Verizon's petition."

Needless to say, I expect SCOTUS to overturn this in an unsigned opinion in the shadow docket. 

 

Et Tu Grok(ey)?

Once again, one of Elon's projects screws up.

In this case, it's his AP program Grok, which leaked thousands of prompts.

This is not a surprise.  What is a surprise is that the chat-bot offered tips on how to assassinate ……… Elon Musk.

Gee, I guess that Grok is like all the rest of the Apartheid Era Emerald Heir Pedo Guy's employees.

Wall Street tech watchers that had only recently recovered from Elon Musk’s AI chatbot going rogue are now quietly reassessing the technology, after a new leak of thousands of user conversations show it teaching people how to make drugs, assassinate Musk himself, and build malware and explosives.

Luckily for xAI, the company that created Musk’s AI chatbot Grok, it is not a publicly traded company, so no public investor or shareholder backlash has forced down its share price or pressured its executives to address the public about privacy concerns.

Yeah, "Lucky," not, "Stupid and corrupt."

………

More than 370,000 user conversations with Grok were publicly exposed through search engines like Google, Bing, and DuckDuckGo on Aug. 21. That led to the posting of a wide range of disturbing content and sent xAI scrambling to contain the fallout and fix the malfunction that reportedly caused the leak.

What kind of disturbing content? Well, in one instance, Grok offers up a detailed plan on how to assassinate Musk himself, before walking that back as “against my policies.” In another exchange, the chatbot also helpfully pointed users to instructions on how to make fentanyl at home or build explosives.

Forbes, which broke the story, reports that the leak stemmed from an unintended malfunction in Grok’s “share” function, which allowed private chats to be indexed and accessed without user consent.

Needless to say, I am amused.

F%$# the Incumbent Carriers

In this case, it is T-Mobile, who just lost their appeal of $92,000,000.00 for selling customer location data without consent.

Their appeal was basically, "It's legal, because wer're T-Mobile."

The judges were less than receptive to this argument: 

A federal appeals court rejected T-Mobile's attempt to overturn $92 million in fines for selling customer location information to third-party firms.

The Federal Communications Commission last year fined T-Mobile, AT&T, and Verizon, saying the carriers illegally shared access to customers' location information without consent and did not take reasonable measures to protect that sensitive data against unauthorized disclosure. The fines relate to sharing of real-time location data that was revealed in 2018, but it took years for the FCC to finalize the penalties.

The three carriers appealed the rulings in three different courts, and the first major decision was handed down Friday. A three-judge panel at the US Court of Appeals for the District of Columbia Circuit ruled unanimously against T-Mobile and its subsidiary Sprint.

………

Until 2019, T-Mobile and Sprint sold customer location information (CLI) to location information aggregators LocationSmart and Zumigo. The carriers did not verify whether buyers obtained customer consent, the ruling said. "Several bad actors abused Sprint and T-Mobile's programs to illicitly access CLI without the customers' knowledge, let alone consent. And even after Sprint and T-Mobile became aware of those abuses, they continued to sell CLI for some time without adopting new safeguards," judges wrote.

……… 

Instead of denying the allegations, the carriers argued that the FCC overstepped its authority. But the appeals court panel decided that the FCC acted properly: 

Needless to say, I expect Trump's FCC is already looking for a way to let your cell phone provider spy on you again. 

Wolf Calls for Better Shepherds

I came across this headline today:

Tech giant Palantir helps the US government monitor its citizens. Its CEO wants Silicon Valley to find its moral compass

I can't even.

Support Your Local Police

In California, the Sacramento Municipal Utilities District (SMUD) was flagging customers with "suspiciously" high electricity usage and sending names and addresses to the cops so that the latter could harass people. 

Gee, surveillance state much?

The Electronic Frontier Foundation has advanced a lawsuit in which it alleges the City of Sacramento misused energy records to accuse residents of growing cannabis, often with disastrous results.

According to a statement from the digital rights group, local law enforcement authorities have worked with local power company, the Sacramento Municipal Utilities District (SMUD), to find households using a suspiciously high amount of energy.

The EFF commenced legal action against SMUD two years ago. This week it finished its discovery and filed a petition [PDF] with the judge in the case along with thousands of pages of evidence, allowing it to move forward with an October hearing.

According to the petition, the Sacramento Police Department (SPS) requested SMUD identify customers in specific ZIP codes whose energy consumption exceeded a monthly threshold, on grounds that the lamps used to grow cannabis indoors use a lot of electrical power. 

Specific ZIP codes?  Only specific ZIP codes?  From a police department that has been repeatedly and successfully sued for using ethnic profiling to harass people?

Yeah, this stinks to high heaven, particularly given the fact that LED grow lights sufficient for a basement pot operation would likely be less than 100W incandescent.

Sounds to me like they were looking for an excuse to hassle Asians:

………

The discovery documents also suggest Sacramento authorities had a preoccupation with race. While law enforcement systematically requested details of suspicious customers, SMUD analysts would apparently ask them to request records for specific addresses. One such text message read: “Send me a request for [two particular addresses]. One is 10k plus, and the other is 4k, Asian….”.

Analysts would also look at credit databases and make racially charged judgments, citing an “interesting thing” about one address was “the multiple Asians that have reported there…” according to one communication from an analyst to law enforcement cited in the petition.

The EFF filed the lawsuit with the Asian American Liberation Network, which advocates for Asian American rights.

One other co-claimant is Alfonso Nguyen, who said police officers visited him and behaved abusively after receiving records from SMUD. Nguyen said he used medical equipment at home to regulate his body temperature. Police called him a liar, the petition says.

As an FYI, Mr. Nguyen is in a wheel chair because of a spinal injury, and needs aggressive air conditioning to control his body temperature.

The EFF mentioned another police visit that saw a resident walk out in his underwear to the sight of police carrying guns. He had been mining crypto.

……… 

This isn't the first time that the SPD has been in court for allegedly targeting Asian Americans for cannabis grow-ops, the document adds. It points to another 2019 case, Wang v. City of Sacramento. 

Yep, seems to me that there is a lot of racial profiling going on here.