Bilingual

Anthropic embedded spyware in Claude Code – and attempted to hide it from you
23 days ago
- Since version 2.1.91, Claude Code checks if a proxy is enabled and covertly transmits data via invisible system prompt alterations, including detecting Chinese timezones, proxy URLs, and affiliations with Chinese AI labs.
- Anthropic uses subtle changes in date formats and apostrophes in the system prompt to embed this information, potentially for detecting unauthorized resale and distillation in China, while obfuscating the code in the binary.
- This covert transmission is criticized as a violation of user trust, raising concerns about potential misuse, such as remote code execution or data exfiltration, given the extensive system access developers grant Claude Code.
Age verification is just a precursor to automated attribution of speech
a month ago
- Age verification laws in various regions are framed as protecting children but function as identity attribution systems.
- These laws aim to tie digital accounts to real identities, making it easier for authorities to identify individuals.
- The state seeks to automate identity attribution, enabling swift action against individuals for online speech without requiring traditional investigation methods.
- Age verification could lead to automated surveillance and penalties for inconvenient speech, similar to copyright infringement notices.
- Resisting age verification and protecting anonymity are recommended, with suggestions to use privacy-focused payment methods if necessary.
"These cameras are just like the Eye of Sauron"
20 days ago
- The study investigates privacy risks for domestic workers (DWs) due to AI-driven smart home devices in both employer and personal households.
- Interviews with 18 UK-based DWs and threat modeling using Communication Privacy Management (CPM) reveal intensified surveillance in employer homes through AI features and opaque agency arrangements.
- In their own homes, DWs had more control but faced challenges like gendered administrative roles, unclear AI functionalities, and data retention uncertainties.
- The research proposes a sociotechnical threat model that identifies DW agencies as institutional adversaries and maps privacy risks across interconnected households.
- It outlines social and practical implications to strengthen DW privacy and agency, extending threat models beyond abstract adversaries and single-household contexts.
Microsoft Can Track Users via a Windows Device ID
19 days ago
- A teenager's arrest revealed Microsoft's use of a Global Device ID (GDID) to track Windows PCs and online activity, raising surveillance concerns.
- Microsoft provided data linking the hacker to crimes via GDID, which persists across Windows updates but can be reset with a reinstall.
- Experts worry GDID could enable widespread user tracking without cookies, prompting questions about similar identifiers on Apple devices.
- Cybersecurity professionals suggest using Linux or VPNs for anonymity, as GDID may be tied to hardware or other identifiers.
Green River Residents Shut Down License Plate Reader Camera Project
17 days ago
- Green River City Council rejected a $112,000 anti-terrorism grant for license plate reader cameras after strong public opposition.
- Residents expressed privacy and Fourth Amendment concerns, fearing third-party data storage and erosion of freedoms.
- The police chief defended the plan for criminal response but acknowledged civil liberties; council members apologized for earlier support.
- No one spoke in favor of the cameras during the meeting, highlighting proactive community resistance to surveillance technology.
Reframing smart glasses as 'pervert glasses'
17 days ago
- Smart glasses, equipped with cameras and microphones, raise significant privacy concerns due to their ability to covertly record audio and video without subjects' knowledge or consent.
- Tech giants like Meta and Snap are promoting smart glasses through fashion brands and celebrity endorsements, but consumer backlash has led to them being labeled 'pervert glasses' over surveillance fears.
- Risks include covert recording for harassment, stalking, or deepfakes, and broader structural dangers like data pooling by companies and access by police or employers, as highlighted by experts.
- An investigative report revealed that Meta's smart glasses expose users' personal information, with contractors accessing sensitive videos, raising awareness about data uploads and accessibility.
- From a security perspective, smart glasses lower the bar for cybercrime by enabling covert recording of screens or passwords, posing risks of data compromise or hacking.
- There are concerns about future privacy invasions, such as facial recognition integration, which could normalize expansive surveillance and further erode privacy, as demonstrated by hobby hackers.
- Consumers can reject these technologies to prevent normalization, as surveillance is not inevitable, and public distrust of tech giants fuels the reframing of smart glasses as surveillance devices.
Exit Chat Control
16 days ago
- EU Prüm II regulation allows facial image exchange across police forces, raising mass surveillance concerns.
- eIDAS 2.0 mandates European digital identity wallets by 2026 and forces browsers to trust state certificates.
- High-Level Group proposes 'lawful access by design' for police access to data, criticized as mass surveillance.
- Romanian election annulled due to TikTok campaign; DSA proceedings opened against platform.
- Vietnam requires identity verification for online posting, linking speech to verified identity.
- UK demands iCloud backdoor; Apple removes encryption for UK users rather than comply.
- France rejects 'ghost participant' mechanism in encrypted messengers after cross-party opposition.
- EU's ProtectEU strategy includes roadmap for accessing encrypted communications and boosting Europol.
- ICE purchases Palantir's 'ImmigrationOS' for targeted surveillance of immigration populations.
- Switzerland's surveillance ordinance proposal stalls after opposition from companies like Proton.
- US requires student visa applicants to make social media public for ideological screening.
- EU schedules state decryption capabilities by 2030, including AI analysis of seized data.
- EU pilots age-verification apps, integrating into digital identity wallets by late 2026.
- France enforces age verification for adult sites, blocking access for non-compliance.
- China introduces centralized 'cyberspace ID' with face recognition, eroding online anonymity.
- EMFA bans spyware against journalists but includes exceptions for national security.
- Russia mandates pre-installed state-controlled messenger MAX on smartphones, with tracking capabilities.
- Russia penalizes searching for 'extremist' content online, including via VPNs.
- Over 500 scientists sign open letter against client-side scanning, citing technical infeasibility.
- UK plans mandatory digital ID for right-to-work checks, sparking public opposition.
- EU pulls mandatory scanning proposal after backlash, but it returns as 'voluntary'.
- EU Entry/Exit System collects biometrics for non-EU travelers, centralizing border control.
- UN Cybercrime Convention adopted, enabling cross-border surveillance with broad definitions.
- ECB advances digital euro development while legal safeguards remain undecided.
- Australia bans under-16s from social platforms, requiring age estimation and deactivating millions of accounts.
- France extends algorithmic CCTV 'experiment' to 2030, expanding surveillance scope.
- Former Polish minister granted asylum in Hungary amid Pegasus spyware scandal.
- Russia blocks WhatsApp after platform refuses data-sharing demands.
- Greek court convicts spyware vendors in Predator case, but buyers remain untraceable.
- France proposes expanding 'black boxes' to include full URL tracking for metadata analysis.
- German and French intelligence switch from Palantir to ChapsVision for AI-driven police analysis.
- US FISA Section 702 lapses but surveillance continues under existing certifications.
- EU Parliament advances digital euro with conditional payment features, despite non-programmability claims.
- Europol budget proposal increases to €3 billion with expanded mandate on encryption and AI.
- Pegasus spyware found on MEP in committee overseeing spyware, highlighting oversight failures.
- Decentralized social networks like Mastodon, Nostr, and Matrix offer alternatives to corporate platforms, emphasizing user control and privacy.
Russian Surveillance Software Suppresses Georgian Civilians Rights
16 days ago
- Georgia built a face recognition enforcement system using software from a Moscow-based company linked to the FSB, which suppresses protests through fines and bank freezes.
- Protesters like Nino and Luka were identified via surveillance cameras, fined heavily, and faced frozen accounts, leading many to stop demonstrating due to fear and financial pressure.
- The government escalated repressive laws: from 2023 to 2026, assembly rules tightened, face covering became criminalized, and online criticism was outlawed, eroding constitutional freedoms.
- The surveillance software, Polyface, uses automated identification, operator-directed searches, and watchlist alerts, integrated with civil registry data and social media for cross-platform monitoring.
- Psychological impact includes paranoia, route changes, and reduced protest participation, with a chilling effect fracturing society and creating moral divides among activists.
- Strategic risks involve Russian control over biometric data, potential integrity attacks, and long-term dominance threats to Georgia's national security and European integration path.
- Despite ongoing surveillance, hundreds still gather daily on Rustaveli Avenue, but the cost of protest has shifted from monetary fines to the loss of personal freedom.
AI Cameras on Garbage Trucks to Scan Properties for Code Violations
15 days ago
- Cities like Cape Coral, Dallas, and Stockton are implementing AI-powered cameras on garbage trucks to detect code violations.
- Fixed-pole surveillance systems like Flock Safety already exist, using AI to analyze vehicles and share data with law enforcement.
- AI on trash trucks scans properties for issues such as overgrown lawns and graffiti, with human verification before fines are issued.
- City officials claim efficiency benefits, including objective records and force multipliers for understaffed departments.
- Privacy concerns include false positives, function creep, and the normalization of infrastructure-level surveillance without consent.
Auburn California Flock Surveillance Cameras Stolen and Dumped in a Canal
15 days ago
- Surveillance cameras, specifically Flock Safety ALPRs, were forcibly removed and destroyed in Auburn, California.
- The cameras, used by law enforcement to log license plates and track vehicles, were either found water-damaged or remain missing.
- Incidents of vandalism against such cameras are not isolated, raising debates over privacy versus public safety.
- The hardware's vulnerability, being easily accessible on poles, highlights a design flaw without announced anti-tamper measures.
- The situation underscores tensions between crime-fighting infrastructure and concerns over mass surveillance.
Flock threatens to sue private group in California that discusses diplomacy, law
15 days ago
- Flock is threatening to sue a private group in Southern California for holding discussions on diplomacy, law, and philosophy.
- The demand includes a prohibition against analyzing or speculating on confirmed security vulnerabilities related to government surveillance.
- The group holds regular Q&A sessions on these topics.
- The application is interactive and requires JavaScript, with simple HTML interfaces not being applicable.
LAPD lets contract with surveillance giant Flock expire
12 days ago
- LAPD will not renew its contract with Flock Safety due to civil liberties and privacy concerns.
- Flock's network includes at least 80,000 cameras that scan license plates for police and federal agencies.
- Other U.S. cities have also ended deals with Flock over privacy and surveillance issues.
- Flock cameras have faced backlash, including vandalism and unauthorized reinstallations.
- Errors with license plate readers have led to wrongful police stops, detentions, and arrests.
- Security lapses exposed Flock cameras and data, raising concerns about hacking and data privacy.
- Flock expressed surprise at the contract end, claiming it could address misconceptions.
Four awful new privacy-eroding features from Meta in a month
12 days ago
- Instagram planned to use public profile photos for AI-generated images on WhatsApp, but backtracked after backlash.
- Meta secretly included facial recognition in its AI app for Ray-Ban smart glasses, removed it after being exposed.
- Meta tested smart glasses that continuously record audio and take photos without user indicators.
- Meta accidentally released a map feature in Brazil showing users' exact locations, leading to investigations.
A new app alerts you if someone nearby is wearing smart glasses
11 days ago
- An Android app named Nearby Glasses can detect and alert users when nearby smart glasses or Bluetooth-enabled recording devices are present.
- The app scans for Bluetooth signals from manufacturers like Meta (Ray-Ban) and Snap, and users can add custom identifiers to detect other devices.
- It was created by Yves Jeanrenaud, inspired by reports on surveillance abuses, such as using smart glasses in immigration raids and harassment.
- Jeanrenaud criticizes smart glasses as privacy-invasive, especially citing Meta's default face recognition feature.
- The app may produce false positives, like mistaking Meta's VR headsets for smart glasses, but these are usually more noticeable.
- Testing showed the app works effectively, detecting Apple devices via a custom Bluetooth identifier.
- Jeanrenaud views the app as a 'desperate act of resistance' against social problems amplified by technology.
- Snap claims its glasses have privacy features like LED indicators, but did not detail prevention of unwanted recordings.
- Meta did not comment on the app or privacy concerns.
The Increasing State of Surveillance Simultaneously Around the World
11 days ago
- Global surveillance is increasing through multiple independent proposals that collectively create a comprehensive system, despite lacking a central plan.
- Governments are establishing identity layers via SIM card KYC, app store age verification, OS-level age attestation, developer verification, and foreign service KYC to make devices identifiable extensions of individuals.
- Governments are pushing for communications layer surveillance through client-side scanning of encrypted messages, CA mandates, TLS interception, network-level filtering, and VPN restrictions, though some efforts have been resisted.
- Device layer surveillance is advancing via OS-level content scanning mandates, mandatory preinstalled software, and potential locked bootloader requirements, enabling control within the cryptographic boundary.
- Reporting and accountability layers are emerging, with trusted flagger infrastructure, lawful intercept laws, and compelled data exfiltration potentially turning corporate tools into government surveillance pipelines.
- Corporations are enabling surveillance through features like on-device AI monitoring, always-on listening devices, cloud sync defaults, and telemetry, often justified as conveniences.
- Platform control by corporations includes app store gatekeeping, browser engine restrictions, attestation primitives, and CDN/DNS chokepoints, reducing alternatives and reinforcing surveillance.
- Indirect pressure from insurance requirements, banking exclusions, and captcha walls creates economic and functional barriers to privacy-respecting tools.
- Resistance has succeeded in defeating some proposals, such as Kazakhstan's CA mandate, Apple's NeuralHash scanning, and EU Chat Control 1.0, showing that organized efforts can be effective.
- To counteract surveillance, individuals can use Linux or custom-ROM Android, support diverse browsers like Firefox, self-host services, fund advocacy groups, and develop alternative software to maintain viable options.
Flock CEO Apologizes for Calling Activists 'Terrorists'
8 days ago
- Will Freeman leads DeFlock, an organization opposing Flock Safety's surveillance cameras, mapping over 80,000 camera locations.
- Flock CEO Garrett Langley apologized for calling DeFlock 'terroristic,' acknowledging valid criticisms and advocating for a balance between safety and privacy.
- Public backlash against Flock grew due to reports of ICE access, a creepy Super Bowl ad with Amazon Ring, and police misusing cameras to stalk women.
- Flock denies working with ICE, retains vehicle data for 30 days (21 days in Washington state), and welcomes privacy regulations.
- Despite vandalism of cameras and activist efforts, Flock's business is growing, with annual revenue reaching $500 million and half from products like drones and AI software.
- DeFlock's apps have 350,000 downloads, and Freeman is suing for records and organizing a National Week of Action Against ALPRs across 102 cities.
We want Texans to know their rights
6 days ago
- Texas used automated license plate readers to track a woman suspected of self-managed abortion, showing ALPRs can weaponize healthcare surveillance.
- Mayday Health launched billboards in Houston to warn Texans about state tracking of pregnancies and inform them about abortion pill options.
- Mayday Health emphasizes privacy on its website, avoiding data collection and directing users to privacy resources for reproductive healthcare.
- Digital rights and reproductive health rights are linked, as anti-abortion efforts use surveillance to restrict access to healthcare.
- Abortion pills are FDA-approved, safe, effective, and available by mail, providing options despite clinic bans and surveillance.
New Orleans Cops Published Policy Document Allowing Weaponized Drones
5 days ago
- The New Orleans Police Department (NOPD) published a draft drone manual allowing weaponized drones with superintendent approval, but later replaced it with a version banning weapons.
- NOPD claims the weaponized drone rules were an early draft, but the incident highlights the department's history of pushing surveillance boundaries and evading public scrutiny.
- Federal Aviation Authority (FAA) regulations prohibit arming drones with dangerous weapons, but experts note these rules are outdated and often ignored by law enforcement.
- Drone manufacturer Skydio, which supplies NOPD, initially opposed weaponizing drones but its CEO later stated it's not their place to restrict military or police use.
- Experts warn that arming drones could lower the threshold for police use of force, making it easier to deploy lethal force remotely with less emotional burden.
- The city council approved $250,000 for expanding Skydio drones in the French Quarter, despite concerns about surveillance and weaponization, with NOPD promising no weapons or facial recognition.
The creepiest 'sales demo' of all time
5 days ago
- A Flock vice president accessed a camera in a children's gymnastics room through the Dunwoody PD's system on Sept. 30, 2025.
- The community center shared camera access to the police under the condition it was only for real-time critical incident response.
- Flock and Dunwoody officials claim the access was a 'sales demo' under a 'demo partner agreement,' but the city cannot provide any evidence of such an agreement.
- The sole camera viewed was in a children's gymnastics room, raising questions about the justification and appropriateness of the demo.
90k Flock cameras have gone up in the US: What they track and how to check
4 days ago
- Flock cameras are installed across the US without public consent, tracking vehicles via license plate recognition.
- The cameras capture vehicle details like plate, make, model, color, and direction; data is stored in the cloud and searchable across a network.
- Risks include AI-driven false alerts, cloud security vulnerabilities, unauthorized access by multiple agencies, and potential misuse for profiling.
- A false alert from Flock led to police surrounding an innocent driver, highlighting the need for regulations and human oversight.
- Residents can find cameras via DeFlock.org, then request contracts, retention rules, and data-sharing agreements from operators to push for public discussion.
- Saranac Lake successfully removed Flock cameras after public protest, showing the power of community action.

surveillance

Anthropic embedded spyware in Claude Code – and attempted to hide it from you

Age verification is just a precursor to automated attribution of speech

"These cameras are just like the Eye of Sauron"

Microsoft Can Track Users via a Windows Device ID

Green River Residents Shut Down License Plate Reader Camera Project

Reframing smart glasses as 'pervert glasses'

Exit Chat Control

Russian Surveillance Software Suppresses Georgian Civilians Rights

AI Cameras on Garbage Trucks to Scan Properties for Code Violations

Auburn California Flock Surveillance Cameras Stolen and Dumped in a Canal

Flock threatens to sue private group in California that discusses diplomacy, law

LAPD lets contract with surveillance giant Flock expire

Four awful new privacy-eroding features from Meta in a month

A new app alerts you if someone nearby is wearing smart glasses

The Increasing State of Surveillance Simultaneously Around the World

Flock CEO Apologizes for Calling Activists 'Terrorists'

We want Texans to know their rights

New Orleans Cops Published Policy Document Allowing Weaponized Drones

The creepiest 'sales demo' of all time

90k Flock cameras have gone up in the US: What they track and how to check