How Dictators Control the Internet: A Review Essay
Open accessThe DOI matches its Crossref record, with a small difference in title or year; 61 of the 61 references it cites were found in a registry or an index; 14 of the 69 sources its claims cite were checked for support.
Abstract
A growing body of research has studied how autocratic regimes interfere with internet communication to contain challenges to their rule. In this review article, we survey the literature and identify the most important directions and challenges for future research. We structure our review along different network layers, each of which provides particular ways of governmental influence and control. While current research has made much progress in understanding individual digital tactics, we argue that there is still a need for theoretical development and empirical progress. First, we need a more comprehensive understanding of how particular tactics fit into an overall digital strategy, but also how they interact with traditional, “offline” means of autocratic politics, such as cooptation or repression. Second, we discuss a number of challenges that empirical research needs to address, such as the effectiveness of digital tactics, the problem of attribution, and the tool dependence of existing research.
Overview
CoChat Summary
TL;DR
This review essay synthesizes research on how autocrats interfere with internet communication, organizing the literature by infrastructure, network, and application layers. It argues that future work needs broader theories of digital strategy and better empirical tools to assess effectiveness, attribution, and tool-specific bias.
Key findings
- The paper identifies three layers of internet control: infrastructure, network, and application, and argues that autocratic interference operates across all three rather than only at the application layer.
- At the infrastructure layer, governments can allocate access unevenly, impose “just-in-time” shutdowns, and throttle bandwidth; the paper cites evidence that service provision is systematically lower for politically excluded ethnic groups and that shutdowns can disconnect parts or all of a population.
- At the network layer, autocrats use filtering, denial-of-service attacks, routing manipulation, and deep packet inspection; the paper notes that China’s “Great Firewall” blocks connections to websites and services to create a highly regulated “national intranet.”
- At the application layer, governments censor content, manipulate information, surveil users, and use e-governance to gather information; the paper notes that Chinese censorship can show a fictitious “internet police” message and that paid users or “fifty-cent army” tactics are used to flood the web with distracting or pro-government content.
Methods
- This is a review essay, not an original empirical study.
- It surveys the literature on autocratic interference in internet communication and structures the review around the internet’s layers: infrastructure, network, and application.
- It synthesizes prior studies to derive theoretical and empirical challenges, including the effectiveness of digital tactics, attribution problems, tool dependence, and the interaction between digital and conventional authoritarian tactics.
Limitations
- The paper itself states that current research is still limited by a lack of theoretical development and empirical progress, especially on how digital tactics fit into broader autocratic strategies.
- It explicitly notes three empirical challenges: effectiveness is often assumed rather than tested, attribution of interference is difficult, and much research is tool-dependent.
- Evident from the text: because this is a narrative review, it does not report a new dataset, sample, or causal estimates of its own.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
12 Valid · 8 Not valid · 37 Not checked · 19 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“However, even when the infrastructure is in place, there are other means to temporarily disable the entire infrastructure (Deibert & Rohozinski, 2010).”
What the sources say
- Supportedcites Deibert & Rohozinski, 2010 source
In Deibert & Rohozinski's 'Beyond Denial' chapter, the section on second-generation controls (p. 24-25) states that 'other, less subtle but nonetheless effective technical means include shutting down Internet access, as well as selected telecommunications services,' citing 'temporary outages of the Internet and SMS services' used by Belarus during the February 2006 presidential elections and similar reported instances in the 2007 Azerbaijani elections. The same chapter (pp. 6-7) describes 'just-in-time' blocking via computer network attacks/DDoS that 'take down' strategically important sources of information or services at key moments, and the index (p. 610) lists Burma's 'Internet shutdown (2007),' confirming that whole-network temporary disablement, not merely selective filtering, is treated in our copy. The claim's framing that these means exist alongside an already-built filtering infrastructure matches the chapter's account of next-generation controls superseding first-generation technical filtering. (LLM, conf=0.80)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“This means that in the case of political events that are potentially threatening to a regime, 'just-in-time' shutdowns allow governments to temporarily disconnect parts of or the whole population (Dainotti etal., 2014; Deibert etal., 2010; Freyburg & Garbe, 2018; Gohdes, 2015).”
What the sources say
- Not checked — no full text of the sourcecites Dainotti et al., 2014 source
- Partly supportedcites Deibert et al., 2010 sourceExperimental — accuracy not yet measured
Our copy does support the core idea: the Deibert & Rohozinski chapter (p. 6) describes controls applied 'just in time,' 'when the information being targeted has the highest value (e.g., during elections or public demonstrations),' and (p. 24) states that second-generation capabilities 'are calibrated to effect ''just-in-time'' or event-based denial of selected content or services.' It also documents temporary shutdowns tied to threatening political events: 'Other, less subtle but nonetheless effective technical means include shutting down Internet access, as well as selected telecommunications services such as cell phone services and especially short message services (SMS). Temporary outages of the Internet and SMS services were employed by Belarus authorities during the February 2006 presidential elections...' (p. 24). What the source does not support is the magnitude/scope in the claim: its 'just-in-time' framing targets selected content, sites or services rather than disconnecting 'parts of or the whole population,' and its documented shutdown cases are confined to the CIS (Belarus 2006, unverified reports for Azerbaijan 2007), not governments generally. (LLM, conf=0.75)
- Not matchedcites Freyburg & Garbe, 2018
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Every source the paper cites for this claim was checked and supports it.
In the paper
“In cases where domestic companies own those servers, governments often have legal and illegal means to compel providers to comply with governmental control practices (Deibert & Rohozinski, 2010).”
What the sources say
- Supportedcites Deibert & Rohozinski, 2010 source
Deibert & Rohozinski state the substance of the claim directly: their overview (p. 4) says next-generation controls 'employ the use of legal regulations to supplement or legitimize technical filtering measures, extralegal or covert practices... and the outsourcing or privatizing of controls to "third parties," to restrict what type of information can be posted, hosted, accessed, or communicated online.' The domestic-provider case is spelled out in the Russian/CIS chapter (pp. 22-25): formal and informal requests to ISPs, providers 'under constant pressure to comply with government requests or face any number of possible sanctions... from visits from the taxation police to revocations of their licenses,' alongside overt legal mechanisms (registration requirements, license revocation, takedown). P. 28 adds Chinese web hosting and social networking services 'routinely obliged to sign self-discipline pacts and follow rigid hosting protocols,' and p. 8 summarizes that 'In many countries, Internet companies are either pressured or legally compelled to censor their services or turn over user data.' The only slight drift is the claim's 'domestic companies own those servers' condition: the source's compulsion runs to any provider operating in the jurisdiction, including foreign-owned ones (Google, Microsoft, Yahoo! in China), so the domestic-ownership framing is narrower than, but not contradicted by, the source. (LLM, conf=0.82)
plainstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“In most autocratic regimes, governmental interference in digital infrastructure and communication is commonplace.”
What the sources say
- Not checked — no full text of the sourcecites Dainotti et al., 2014 source
- Partly supportedcites King et al., 2013 sourceExperimental — accuracy not yet measured
The Introduction describes extensive Chinese government censorship of online communication, and §3.1 describes website blocking. These passages support China as an example, but they do not establish that such interference is commonplace in most autocratic regimes. Page numbers are unavailable in our copy. (LLM, conf=0.86)
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“In addition to these refined techniques, so-called denial-of-service attacks provide a relatively cheap way of censoring information where automated control cannot be implemented easily.”
What the sources say
- Supportedcites Lutscher et al., 2020 source
In “A Tool for Censorship” (page not provided), the paper describes DoS attacks as relatively low-cost and easy to employ for temporarily restricting access to websites. It also says they offer an alternative for governments that lack the technical capabilities for more sophisticated censorship methods. (LLM, conf=0.91)
- Not matchedcites Nazario, 2009
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites H. Roberts et al., 2012
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
43 more claims · 3 with their cited sources checked
Create a free account to see all 76 claims and to verify the 2 references still open.
- It may state two separate claims as one. 20
- Probably not a substantial claim. 12
- Borderline: one of the tests for a substantial claim is unclear. 1
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The evolution of censorship throughout this time period underscores the importance of being able to track multiple methods of censorship to gain a holistic understanding of the digital strategy of autocrats, something that has as of yet been difficult to do at scale [42].”
From A multi-perspective view of Internet censorship in Myanmar (2021) · cites this paper as [42]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“As a result, some forms of digital repression (particularly forms of covert information coercion, see Table 2, cell 2) may be unavailable or operate at rudimentary levels until the regime’s technological infrastructure reaches particular thresholds (24).”
From Physical and virtual repertoires of contention (2022) · cites this paper as [24]
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Second, African governments have strong incentives to attract FDI in their domestic telecommunications sector to stimulate growth. 62”
From Authoritarian collaboration and repression in the digital age: balancing foreign direct investment and control in internet infrastructure (2026) · cites this paper as [62]
Experimental — accuracy not yet measured
Citations
Cites
How Censorship in China Allows Government Criticism but Silences Collective Expression
KING, G., PAN, J., & ROBERTS, M. E. (2013) · American Political Science Review · doi:10.1017/s0003055413000014 · Cited by 2,415 · cited 5×
“In China, for instance, two virtual police officers remind citizens that their browsing behavior is monitored (King et al., 2013).”
BackgroundInfluentialState Repression and Political Order
Davenport, C. (2007) · Annual Review of Political Science · doi:10.1146/annurev.polisci.10.101405.143216 · Cited by 1,564 · cited 1×
“Research has long argued that autocratic regimes select from a large repertoire of approaches to ensure political survival (Davenport, 2007; Gerschewski, 2013), for example by co-opting elites, increasing legitimacy, or by violently repressing dissent.”
BackgroundThe three pillars of stability: Legitimation, repression, and co-optation in autocratic regimes
Gerschewski, J. (2013) · Democratization · doi:10.1080/13510347.2013.738860 · Cited by 1,053
KING, G., PAN, J., & ROBERTS, M. E. (2017) · American Political Science Review · doi:10.1017/s0003055417000144 · Cited by 1,028 · cited 1×
“Oftentimes, this is done with the help of paid users (Han, 2015; King et al., 2017).”
BackgroundTilly, C. (2006) · doi:10.7208/chicago/9780226803531.001.0001 · Cited by 1,018
The Political Economy of Dictatorship
Wintrobe, R. (1998) · doi:10.1017/cbo9781139174916 · Cited by 1,008
Sources of Authoritarian Responsiveness: A Field Experiment in China
Chen, J., Pan, J., & Xu, Y. (2016) · American Journal of Political Science · doi:10.1111/ajps.12207 · Cited by 636
Lorentzen, P. (2014) · American Journal of Political Science · doi:10.1111/ajps.12065 · Cited by 602 · cited 1×
“These innovations further mitigate principal-agent problems between rulers and local officials when citizens help to detect corruption, which in turn is expected to alleviate griev-ances (Chen et al., 2016; Lorentzen, 2014).”
BackgroundGuriev, S., & Treisman, D. (2019) · Journal of Economic Perspectives · doi:10.1257/jep.33.4.100 · Cited by 560
Censored: Distraction and Diversion Inside China's Great Firewall
(2018) · Project Muse (Johns Hopkins University) · doi:10.23943/9781400890057/html · Cited by 508
Research works printed but not linked: 3
- Freyburg T., 2018, International Journal of Communication, 12, 3896
- Hellmeier S., 2016, Politics & Policy, 42, 635
- Roberts H., 2012, Access contested: Security, identity, and resistance in Asian cyberspace, 133
Cited by
Earl, J. (2022) · Science Advances · doi:10.1126/sciadv.abl8198 · Cited by 250
State of the world 2021: autocratization changing its nature?
Boese, V. A., Lundstedt, M., Morrison, K., Sato, Y., & Lindberg, S. I. (2022) · Democratization · doi:10.1080/13510347.2022.2069751 · Cited by 217
Repression in the Digital Age: Surveillance, Censorship, and the Dynamics of State Violence
Gohdes, A. R. (2024) · Oxford University Press · doi:10.1093/oso/9780197743577.001.0001 · Cited by 59
How African countries respond to fake news and hate speech
Garbe, L., Selvik, L.-M., & Lemaire, P. (2023) · Information, Communication & Society · doi:10.1080/1369118x.2021.1994623 · Cited by 57
How Authoritarianism Transforms: A Framework for the Study of Digital Dictatorship
Schlumberger, O., Edel, M., Maati, A., & Saglam, K. (2024) · Government and Opposition · doi:10.1017/gov.2023.20 · Cited by 48
Metadata
- Authors
- Keremoğlu, E., & Weidmann, N. B.
- Affiliations
- University of Konstanz (DE)
- Corresponding author
- Weidmann, Nils B.
- Venue
- Comparative Political Studies
- Publisher
- SAGE Publications
- Published
- 2020-03-23
- Volume
- 53
- Issue
- 10-11
- Pages
- 1690-1703
- ISSN
- 0010-4140 (ISSN-L)
- 1552-3829
- OpenAlex
- W3014033480
- Cited by
- 131
- References
- 48
- Citations by year
- 2026: 14 · 2025: 28 · 2024: 17 · 2023: 33 · 2022: 29 · 2021: 6 · 2020: 4
- FWCI
- 24.02
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- Comparative Political Studies (publishedVersion, cc-by-nc)
- KOPS (University of Konstanz) (submittedVersion, cc-by-nc)
- Funders
- Deutsche Forschungsgemeinschaft (402127652)
- Topics
- Social Sciences › Sociology and Political Science › Political Conflict and Governance
- Social Sciences › Political Science and International Relations › Cybersecurity and Cyber Warfare Studies
- Social Sciences › Sociology and Political Science › Media Influence and Politics
- Keywords
- internet censorship, authoritarian regimes, co-optation, repression, digital strategy, network layer
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (qwen3.8-omni-flash and gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: How Dictators Control the Internet: A Review Essay. https://app.cochat.ai/papers/10.1177/0010414020912278
- The paper
- Keremoğlu, E., & Weidmann, N. B. (2020). How Dictators Control the Internet: A Review Essay. Comparative Political Studies. https://doi.org/10.1177/0010414020912278
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.