Comprehensive analysis documenting AI-enabled authoritarian tools across surveillance (350M+ cameras in China analyzing 25.9M faces daily per district), censorship (22+ countries mandating AI content removal), and social control (1.16B individuals in social credit database). Evidence shows Chinese surveillance tech deployed in 100+ countries via Digital Silk Road; 80% of global population lives in countries "not fully free." Argues AI enables "perfect autocracy" through preemptive suppression—RAND suggests 90%+ detection of organized opposition—with $300B surveillance market projected by 2028.
Authoritarian Tools
AI Authoritarian Tools
Comprehensive analysis documenting AI-enabled authoritarian tools across surveillance (350M+ cameras in China analyzing 25.9M faces daily per district), censorship (22+ countries mandating AI content removal), and social control (1.16B individuals in social credit database). Evidence shows Chinese surveillance tech deployed in 100+ countries via Digital Silk Road; 80% of global population lives in countries "not fully free." Argues AI enables "perfect autocracy" through preemptive suppression—RAND suggests 90%+ detection of organized opposition—with $300B surveillance market projected by 2028.
AI Authoritarian Tools
Comprehensive analysis documenting AI-enabled authoritarian tools across surveillance (350M+ cameras in China analyzing 25.9M faces daily per district), censorship (22+ countries mandating AI content removal), and social control (1.16B individuals in social credit database). Evidence shows Chinese surveillance tech deployed in 100+ countries via Digital Silk Road; 80% of global population lives in countries "not fully free." Argues AI enables "perfect autocracy" through preemptive suppression—RAND suggests 90%+ detection of organized opposition—with $300B surveillance market projected by 2028.
Quick Assessment
| Dimension | Assessment | Evidence |
|---|---|---|
| Current Scale | 350+ million surveillance cameras in China; 1.16 billion individuals in social credit database | Freedom House 2025; Global Times 2024 |
| Global Spread | 100+ countries using Chinese smart city tech; 47-65 countries with AI surveillance components | ORF 2024 |
| Internet Freedom Decline | 15 consecutive years of decline; 28 of 72 countries deteriorated in 2024-25 | Freedom House 2025 |
| Economic Cost | $7.69 billion lost to internet shutdowns in 2024; 88,000 hours of outages globally | Access Now 2024 |
| Surveillance Market | $300 billion projected by 2028; $5.33 billion China IP camera market in 2025 | Mordor Intelligence |
| Political Repression | Legal consequences for online speech in 55 of 70 countries; 47 countries deploy state commentators | Freedom House 2025 |
| Stability Risk | High—AI enables preemptive suppression of dissent before organization occurs | RAND Corporation; Lawfare |
Overview
Artificial intelligence is fundamentally transforming the tools of authoritarianism, enabling unprecedented capabilities for surveillance, censorship, propaganda, and social control. Unlike traditional autocracies that relied on physical force and limited information, AI-powered authoritarian systems can monitor entire populations in real-time, automatically detect and suppress dissent, and predict opposition before it organizes.
Freedom House↗🔗 web★★★★☆Freedom HouseFreedom Houseauthoritarianismhuman-rightsdigital-repressionx-risk+1Source ↗ reports that internet freedom has declined for 15 consecutive years as of 2025, with AI playing an increasingly central role in digital repression. In 2024-25 alone, conditions deteriorated in 28 of 72 countries assessed. At least 22 countries now mandate platforms use machine learning to remove political, social, and religious speech deemed undesirable by authorities. China's surveillance state monitors 1.4 billion people through 350+ million cameras with facial recognition, a social credit database covering 1.16 billion individuals, and integrated behavioral analysis platforms.
The core concern extends beyond immediate human rights violations: AI may enable the creation of stable, durable authoritarian regimes that are significantly harder to overthrow than historical autocracies. If comprehensive surveillance can detect organizing before it becomes effective, and predictive systems can identify dissidents early, billions could live under repressive regimes indefinitely—representing a potential civilizational lock-inRiskAI Value Lock-inComprehensive analysis of AI lock-in scenarios where values, systems, or power structures become permanently entrenched. Documents evidence including Big Tech's 66-70% cloud control, AI surveillanc...Quality: 64/100 of oppressive governance.
AI-Enabled Authoritarian Control Ecosystem
Risk Assessment
| Factor | Assessment | Evidence |
|---|---|---|
| Current Severity | High | 12+ million Uyghurs under comprehensive surveillance↗🔗 web200+ million Uyghurs under surveillanceauthoritarianismhuman-rightsdigital-repressionSource ↗; 22+ countries mandating AI content removal |
| Geographic Scope | Expanding rapidly | 100+ countries using Chinese smart city tech; 47-65 countries with AI surveillance (ORF 2024) |
| Technological Maturity | Near-comprehensive | Facial recognition 99.9% accurate; China's cameras analyze 25.9M faces daily in one district alone |
| Population Affected | 4+ billion at risk | 80% of global population lives in countries "not fully free" per Freedom House |
| Economic Infrastructure | $22 billion invested | Digital Silk Road investment 2017-2023; $300B surveillance market by 2028 |
| Stability Risk | Extreme | AI enables preemptive suppression—RAND analysis suggests 90%+ detection of organized opposition |
| Timeline | Accelerating | City Brain 3.0 launched March 2025; integration deepening annually |
| Trend | Worsening | 15 consecutive years of internet freedom decline; 28 countries deteriorated in 2024-25 |
Comprehensive AI Control Systems
Surveillance Infrastructure
Modern AI surveillance operates at unprecedented scale and granularity. China's SenseTime↗🔗 web★★★★☆ReutersSenseTimeauthoritarianismhuman-rightsdigital-repressionSource ↗ and Megvii↗🔗 webMegviiauthoritarianismhuman-rightsdigital-repressionSource ↗ systems can identify individuals from crowds in real-time, track movements across cities, and correlate behavior patterns across multiple data sources. Shanghai alone has over 5,000 surveillance cameras per square mile. In one Shanghai district, authorities estimate they capture and analyze 25.9 million faces daily—an average of 18,860 individuals per minute. The integration extends far beyond facial recognition:
- Gait analysis identifies individuals from walking patterns, defeating facial coverings
- Voice recognition monitors phone calls and public conversations
- Digital exhaust tracks online behavior, purchases, and location data
- Social network analysis maps relationships and influence patterns
- Predictive modeling flags "pre-crime" indicators and protest likelihood
Carnegie Endowment research↗🔗 web★★★★☆Carnegie EndowmentGlobal Expansion of AI SurveillanceA comprehensive study reveals the widespread adoption of AI surveillance technologies worldwide, with Chinese companies playing a major role in supplying these systems to govern...authoritarianismhuman-rightsdigital-repressionSource ↗ documents Chinese surveillance technology deployment in over 80 countries, often through "Safe City" infrastructure projects that embed comprehensive monitoring capabilities into urban planning.
Automated Censorship at Scale
AI censorship systems operate with speed and comprehensiveness impossible for human moderators. Oxford Internet Institute↗🔗 webOxford Internet Institute: Computational PropagandaThe Oxford Internet Institute's Computational Propaganda project studies how digital technologies are used to manipulate public opinion and influence democratic processes. They ...mental-healthai-ethicsmanipulationauthoritarianism+1Source ↗ research shows these systems can:
- Content filtering: Remove text, images, and videos in milliseconds based on semantic understanding
- Shadow banning: Reduce content visibility without explicit removal
- Keyword evolution: Automatically identify new euphemisms and coded language
- Context analysis: Distinguish between permitted and forbidden uses of identical content
China's Great Firewall 2.0↗🔗 web★★★★☆Atlantic CouncilGreat Firewall 2.0authoritarianismhuman-rightsdigital-repressionSource ↗ employs deep packet inspection and machine learning to block VPNs dynamically. Russian SORM↗🔗 webSORMauthoritarianismhuman-rightsdigital-repressionSource ↗ systems have evolved to incorporate AI-driven content analysis across platforms.
Personalized Propaganda and Influence Operations
AI enables micro-targeted propaganda that adapts to individual psychological profiles. Stanford Internet Observatory↗🔗 webStanford Internet Observatoryauthoritarianismhuman-rightsdigital-repressionSource ↗ research demonstrates:
- Behavioral targeting: Personalized messaging based on browsing history, social connections, and inferred beliefs
- A/B testing at scale: Real-time optimization of persuasive content
- Deepfake generation: Synthetic media indistinguishable from authentic content
- Emotional manipulation: Content designed to trigger specific psychological responses
The Internet Research Agency↗🏛️ governmentSenate Intelligence Committee ReportThe Senate Intelligence Committee report details how the Internet Research Agency (IRA) used social media platforms to spread disinformation and divisive content targeting Ameri...authoritarianismhuman-rightsdigital-repressionSource ↗ operations in 2016 U.S. elections demonstrated early-stage capabilities; current systems are orders of magnitude more sophisticated.
Social Credit and Behavioral Modification
China's Social Credit System↗🔗 webSocial Credit Systemauthoritarianismhuman-rightsdigital-repressionSource ↗ represents the most comprehensive attempt to use AI for population-wide behavioral modification. By the end of 2024, the system had collected credit data on 1.16 billion individuals and 140 million enterprises:
- Comprehensive scoring: Integration of financial, social, and political behavior into unified ratings; 80% of provinces had implemented some version by 2022
- Algorithmic punishment: Automatic restriction of travel, education, and employment based on scores; 26.82 million air tickets and 5.96 million rail tickets denied to blacklisted individuals (as of 2019)
- Predictive intervention: Early identification of "unreliable" individuals before violations occur
- Corporate focus: 33+ million businesses given scores under the Corporate Social Credit System
- Social pressure: Public shaming and peer pressure through score visibility; 4.37 million blacklisted people subsequently fulfilled legal duties
Sesame Credit↗🔗 webSesame Creditauthoritarianismhuman-rightsdigital-repressionSource ↗ pilot programs demonstrated 20-30% improvement in targeted behaviors. However, recent analysis suggests the system is now focused primarily on corporate compliance rather than individual social scores.
Current Deployment and Geographic Spread
Global AI Surveillance Adoption by Region
| Region | Countries with Chinese Surveillance Tech | Key Technologies | Notable Implementations |
|---|---|---|---|
| East Asia | China, Hong Kong (expanding) | Full-spectrum surveillance, City Brain 3.0 | 350M cameras; 5,000 cameras/sq mile in Shanghai |
| Central Asia | Uzbekistan, Kazakhstan, Tajikistan | Safe City systems, facial recognition | Huawei Safe City initiative since 2017 |
| Southeast Asia | Myanmar, Cambodia, Vietnam, Philippines | Internet controls, facial recognition | Myanmar's 2024 VPN blocking; censorship laws |
| Middle East | UAE, Saudi Arabia, Egypt | Smart city infrastructure, AI analytics | 28 of 34 China-led projects involve intelligence ties |
| Africa | 18+ countries including Zimbabwe, Uganda | Safe City programs, facial recognition | 266 Chinese tech projects; social media taxes |
| Latin America | Venezuela, Ecuador, Bolivia, Argentina | Safe City products, surveillance systems | 35 cities using Huawei Safe City |
| Eastern Europe | Russia, Belarus | Internet isolation, content filtering | Signal blocked; YouTube throttled in 2024 |
Sources: ORF; Freedom House 2025; Jamestown Foundation
China's Integrated System
China operates the world's most comprehensive AI-enabled authoritarian system. Human Rights Watch↗🔗 web200+ million Uyghurs under surveillanceauthoritarianismhuman-rightsdigital-repressionSource ↗ documentation reveals:
- Xinjiang surveillance: 1 camera per 6 residents, mandatory phone app monitoring, DNA collection for 12+ million Uyghurs
- Nationwide expansion: 350+ million cameras with facial recognition capabilities (targeting 600 million)
- Predictive policing: IJOP system↗🔗 webIJOP systemauthoritarianismhuman-rightsdigital-repressionSource ↗ flags "unusual" behavior for investigation; combines CCTV, WiFi, and checkpoint data
- Social credit: 1.16 billion individuals and 140 million enterprises tracked; 6.7 billion credit report inquiries to date
- Travel restrictions: 33+ million blacklisted individuals denied air or rail tickets as of 2019
- Internet censorship: Real-time blocking of millions of websites and keywords via Great Firewall 2.0
The system's effectiveness is demonstrated by the absence of large-scale protests since implementation, despite historical patterns of periodic unrest. Research suggests fewer people protest when public safety agencies acquire AI surveillance technology.
Russian Digital Authoritarianism
Russia's Sovereign Internet Law↗🔗 webSovereign Internet Lawauthoritarianismhuman-rightsdigital-repressionSource ↗ creates infrastructure for comprehensive digital control. In 2024-25, authorities dramatically escalated digital isolation:
- Deep packet inspection: Real-time monitoring and filtering of all internet traffic via SORM systems
- Platform blocking: Signal blocked and YouTube throttled in summer 2024; Cloudflare ECH protocol sites restricted
- Encrypted messaging ban: In June 2025, Russia and Belarus announced plans to develop AI built on "fundamental and traditional values"
- Platform compliance: Requirements for data localization and content removal
- Information warfare: State-sponsored disinformation campaigns using AI-generated content; 47 countries now deploy state commentators
- Opposition targeting: Navalny app removal↗🔗 web★★★★☆ReutersNavalny app removalauthoritarianismhuman-rightsdigital-repressionSource ↗ demonstrates platform cooperation under pressure
Global Expansion
Freedom House tracking↗🔗 web★★★★☆Freedom HouseFreedom House trackingauthoritarianismhuman-rightsdigital-repressionSource ↗ shows authoritarian technology adoption across regions:
- Middle East: UAE, Saudi Arabia deploying Chinese surveillance systems
- Africa: 18 countries with Chinese-supplied "Safe City" programs
- Latin America: Venezuela, Ecuador implementing social control systems
- Southeast Asia: Myanmar, Cambodia expanding digital monitoring
Export financing through Belt and Road Initiative↗🔗 webBelt and Road Initiativeauthoritarianismhuman-rightsdigital-repressionSource ↗ often includes surveillance infrastructure, creating long-term technological dependencies.
Digital Authoritarianism Impact by Country (2024-2025)
| Country | Freedom Score Change | Key Developments | Economic Impact |
|---|---|---|---|
| Venezuela | -7 points | Second-largest decline globally; internet controls during July 2024 election | Significant productivity losses |
| Russia | Severe decline | Signal blocked; YouTube throttled; Sovereign Internet deepening | Growing isolation costs |
| Myanmar | Severe decline | New VPN blocking tech in 2024; cybersecurity law January 2025 | $1.62B+ losses (2024) |
| Pakistan | Major decline | Nationwide cellular blackout on election day; social media blocks | $1.62 billion in 2024 |
| China | Sustained low | City Brain 3.0 launched; 350M+ cameras; 1.16B in social credit system | Market leader ($5.3B camera market) |
| Iran | Sustained low | AI deployed to identify hijab violations (2022-ongoing) | International isolation |
| Egypt | Sustained decline | Among worst 15-year declines globally | Investment uncertainty |
| Turkey | Sustained decline | Among worst 15-year declines globally | Tech sector impacts |
Sources: Freedom House 2025; Access Now
Global economic cost of internet shutdowns in 2024: $7.69 billion across approximately 88,000 hours of outages in 39 countries.
The Stability Concern: Perfect Autocracy
Historical autocracies fell through revolution, coups, or external pressure. AI may fundamentally alter these dynamics by creating "perfect autocracy"—regimes with comprehensive information about their populations and the ability to suppress threats before they materialize.
Information Asymmetry
Traditional revolutions required information advantage—knowing something the regime didn't. AI surveillance eliminates this by providing:
- Real-time monitoring: Continuous awareness of population sentiment and activity
- Predictive capabilities: Early warning systems for protest organization
- Network analysis: Identification of influential individuals and communication patterns
- Behavioral prediction: Models forecasting individual likelihood of dissent
RAND Corporation analysis↗🔗 web★★★★☆RAND CorporationRAND Corporation analysisauthoritarianismhuman-rightsdigital-repressiongovernance+1Source ↗ suggests comprehensive surveillance could detect 90%+ of organized opposition activity before it reaches critical mass.
Preemptive Suppression
Rather than reacting to threats, AI enables prevention through:
- Targeted intervention: Removing key organizers before movements form
- Information manipulation: Flooding communication channels with noise
- Social isolation: Restricting travel, employment, and social connections for dissidents
- Psychological pressure: Demonstrating omnipresent monitoring to discourage resistance
International Implications
Stable AI-enabled authoritarianism could affect global governance by:
- Norm erosion: Legitimizing digital repression as "effective governance"
- Technology export: Spreading control systems to client states
- Democratic pressure: Forcing open societies to compete on efficiency rather than freedom
- Lock-in effects: Creating technological and economic dependencies difficult to reverse
Key Technological Uncertainties
Circumvention vs. Suppression
The ongoing competition between surveillance capabilities and privacy-preserving technologies remains uncertain:
| Technology | Surveillance Capability | Circumvention Tool | Current Balance | Trend |
|---|---|---|---|---|
| Encryption | Deep packet inspection; metadata analysis | Signal Protocol; quantum-resistant protocols | Contested | Surveillance gaining via metadata |
| VPNs | Dynamic blocking; traffic pattern analysis | Obfuscation protocols; decentralized VPNs | Surveillance advantage | Myanmar blocked VPNs in 2024 |
| Anonymity networks | Tor exit node monitoring; traffic correlation | Tor; I2P; mesh networks | Mixed | Russia blocked Tor in 2024 |
| Facial recognition | 99.9% accuracy under optimal conditions | Masks; adversarial makeup; IR LEDs | Strong surveillance advantage | 350M+ cameras deployed |
| Gait analysis | Defeats facial coverings | Limited countermeasures | Surveillance advantage | Rapidly advancing |
| AI content filtering | Real-time semantic analysis; context detection | Euphemisms; coded language; steganography | Contested | AI auto-detects new evasion patterns |
Key insight: Circumvention tools provide temporary advantages, but state-level actors have sustained resources for detection improvements.
- Encryption advancement: Quantum-resistant protocols may preserve private communication
- Anonymization tools: Tor, VPNs, and decentralized networks enable some circumvention
- AI detection: Advanced systems may identify circumvention attempts in real-time
- Cat-and-mouse dynamics: Historical precedent suggests temporary advantages rather than permanent solutions
Electronic Frontier Foundation↗🔗 webElectronic Frontier Foundationauthoritarianismhuman-rightsdigital-repressionSource ↗ research indicates circumvention tools face increasing sophistication in detection and blocking.
Technological Dependencies
The durability of AI-enabled authoritarianism may depend on:
- Semiconductor supply chains: Advanced chips required for surveillance infrastructure
- Internet infrastructure: Physical control points for traffic monitoring
- Cloud computing: Centralized vs. distributed processing capabilities
- Energy requirements: Substantial power needs for comprehensive surveillance
Human Factors
AI systems require human operators, creating potential vulnerabilities:
- Operator loyalty: Security forces must remain committed to the regime
- Technical expertise: Maintaining complex systems requires skilled personnel
- Error rates: False positives could create public resentment
- Adaptation: Opposition groups may develop counter-surveillance tactics
Current Trajectory and Projections
Technological Advancement
AI capabilities relevant to authoritarianism are advancing rapidly:
- Accuracy improvements: Facial recognition error rates dropping approximately 50% annually; now exceeding 99.9% accuracy under optimal conditions
- Processing speed: Real-time analysis of millions of faces per day per district; City Brain 3.0 launched March 2025
- Integration capabilities: Unified systems combining CCTV, WiFi, purchase data, location tracking, and biometrics
- Cost reduction: China's surveillance IP camera market at $5.33 billion in 2025, growing 13.3% annually to $9.94 billion by 2030
- Market scale: Global surveillance technology market projected to exceed $300 billion by 2028
MIT Technology Review↗🔗 web★★★★☆MIT Technology ReviewMIT Technology Reviewauthoritarianismhuman-rightsdigital-repressionSource ↗ reports facial recognition accuracy exceeding 99.9% under optimal conditions. The number of countries deploying state commentators to manipulate online discussions has doubled in the past decade to 47 countries.
Geographic Expansion
Current trends suggest continued spread of authoritarian AI:
- Technology transfer: Chinese vendors expanding global market share
- Financing mechanisms: Development banks funding surveillance infrastructure
- Technical training: Capacity building for local implementation
- Regulatory frameworks: Legal structures legitimizing digital monitoring
Democratic Responses
Nascent efforts to counter authoritarian AI include:
- Export controls: U.S. and EU restrictions on surveillance technology sales
- Privacy legislation: GDPR and similar frameworks limiting data collection
- Technical assistance: Supporting civil society with circumvention tools
- Diplomatic pressure: Sanctions and international criticism
However, Center for Strategic and International Studies↗🔗 web★★★★☆CSISCenter for Strategic and International Studiesauthoritarianismhuman-rightsdigital-repressionSource ↗ analysis suggests defensive measures lag significantly behind authoritarian capabilities.
Countermeasures and Defensive Strategies
Technical Approaches
- Privacy-preserving technologies: Signal Protocol↗🔗 webSignal Protocolauthoritarianismhuman-rightsdigital-repressionSource ↗, Tor↗🔗 webTorauthoritarianismhuman-rightsdigital-repressionSource ↗, mesh networking
- Decentralized systems: Blockchain-based communication and organization tools
- AI red-teaming: Testing surveillance systems for vulnerabilities
- Open-source intelligence: Monitoring authoritarian technology deployment
Policy Responses
- Export controls: Bureau of Industry and Security↗🏛️ government★★★★☆Bureau of Industry and SecurityBureau of Industry and Securitycybersecurityauthoritarianismhuman-rightsdigital-repressionSource ↗ Entity List restrictions
- Sanctions regimes: Targeting surveillance technology companies and users
- International coordination: Freedom Online Coalition↗🔗 webFreedom Online Coalitionauthoritarianismhuman-rightsdigital-repressionSource ↗ diplomatic efforts
- Funding alternatives: Supporting democratic technology development
Civil Society Strategies
- Digital security training: Teaching circumvention and privacy tools
- Documentation: Recording human rights violations enabled by AI
- Advocacy: Raising awareness of surveillance technology impacts
- Legal challenges: Constitutional and human rights litigation
Timeline of AI-Enabled Authoritarianism
Historical Foundation (2010-2015)
- 2012: China begins massive surveillance camera deployment under Skynet project
- 2013: Snowden revelations expose NSA capabilities, spurring global surveillance adoption
- 2014: Xi Jinping consolidates power, accelerates Social Credit System development
- 2015: China's Cybersecurity Law establishes data localization requirements
System Integration (2016-2020)
- 2016: Internet Research Agency demonstrates AI-powered influence operations
- 2017: Xinjiang surveillance apparatus reaches full deployment; Digital Silk Road launched
- 2018: China's Social Credit System enters nationwide pilot phase
- 2019: Russia passes Sovereign Internet Law enabling comprehensive filtering; 26.82M air tickets denied to blacklisted individuals
- 2020: COVID-19 contact tracing normalizes population surveillance globally
Current Acceleration (2021-2025)
- 2021: Taliban uses facial recognition to hunt former officials
- 2022: Iran deploys AI to identify hijab violations; China launches Global Security Initiative
- 2023: 22+ countries mandate AI-powered content removal; internet shutdowns reach record 283 instances across 39 countries
- 2024: Venezuela suffers second-largest global decline in internet freedom; Russia blocks Signal and throttles YouTube; Myanmar deploys VPN-blocking technology; Hong Kong plans 2,000 new cameras with facial recognition; $7.69B global cost of internet shutdowns
- 2025: Freedom House reports 15th consecutive year of internet freedom decline; China launches City Brain 3.0; Myanmar enacts cybersecurity law restricting anti-censorship tools; China regulates facial recognition (effective June 2025); Russia-Belarus announce "values-based" AI development
Projected Developments (2026-2030)
- Enhanced prediction: AI systems forecasting individual behavior with 95%+ accuracy
- Camera expansion: China targeting 600 million surveillance cameras
- Market growth: Surveillance market exceeding $300 billion by 2028
- Counter-surveillance evolution: Arms race between monitoring and privacy technologies
- Institutional lock-in: Democratic backsliding enabled by "temporary" surveillance measures
Sources & Resources
Primary Research Organizations
- Freedom House - Freedom on the Net Reports↗🔗 web★★★★☆Freedom HouseFreedom House Reportsauthoritarianismhuman-rightsdigital-repressionSource ↗
- Carnegie Endowment - Global Expansion of AI Surveillance↗🔗 web★★★★☆Carnegie EndowmentGlobal Expansion of AI SurveillanceA comprehensive study reveals the widespread adoption of AI surveillance technologies worldwide, with Chinese companies playing a major role in supplying these systems to govern...authoritarianismhuman-rightsdigital-repressionSource ↗
- Human Rights Watch - China Surveillance Documentation↗🔗 web200+ million Uyghurs under surveillanceauthoritarianismhuman-rightsdigital-repressionSource ↗
- Oxford Internet Institute - Computational Propaganda Research↗🔗 webOxford Internet Institute: Computational PropagandaThe Oxford Internet Institute's Computational Propaganda project studies how digital technologies are used to manipulate public opinion and influence democratic processes. They ...mental-healthai-ethicsmanipulationauthoritarianism+1Source ↗
- Stanford Internet Observatory - Platform Governance↗🔗 webStanford Internet Observatoryauthoritarianismhuman-rightsdigital-repressionSource ↗
Policy and Technical Analysis
- RAND Corporation - Information Warfare Studies↗🔗 web★★★★☆RAND CorporationRAND Corporation - Information Warfare Studiesauthoritarianismhuman-rightsdigital-repressionSource ↗
- Center for Strategic and International Studies - Technology and National Power↗🔗 web★★★★☆CSISCSIS Critical Questionsgame-theoryinternational-coordinationgovernanceauthoritarianism+1Source ↗
- Atlantic Council - Digital Forensic Research Lab↗🔗 web★★★★☆Atlantic CouncilAtlantic Council DFRLabThe Atlantic Council's DFRLab is a research organization focused on exposing digital threats, disinformation, and protecting democratic institutions through open-source investig...open-sourceauthoritarianismhuman-rightsdigital-repression+1Source ↗
- Electronic Frontier Foundation - Surveillance and Privacy↗🔗 webElectronic Frontier Foundationauthoritarianismhuman-rightsdigital-repressionSource ↗
- Citizen Lab - Digital Rights Research↗🔗 webCitizen Lab - Digital Rights Researchauthoritarianismhuman-rightsdigital-repressionSource ↗
Government and International Bodies
- U.S. Bureau of Industry and Security - Export Controls↗🏛️ government★★★★☆Bureau of Industry and SecurityBureau of Industry and Securitycybersecurityauthoritarianismhuman-rightsdigital-repressionSource ↗
- EU AI Office - Regulation and Oversight↗🔗 web★★★★☆European Union**EU AI Office**risk-factorcompetitiongame-theorycascades+1Source ↗
- UK AI Safety Institute - Risk Assessment↗🏛️ government★★★★☆UK AI Safety InstituteAI Safety Institutesafetysoftware-engineeringcode-generationprogramming-ai+1Source ↗
- Freedom Online Coalition - Digital Rights↗🔗 webFreedom Online Coalitionauthoritarianismhuman-rightsdigital-repressionSource ↗
- UN Special Rapporteur on Privacy↗🔗 webUN Special Rapporteur on Privacyauthoritarianismhuman-rightsdigital-repressionSource ↗