Obsah obrázku socha, umění Popis byl vytvořen automaticky
24.10.2023
GLCb2028 Artificial Intelligence in Political Science and Security Studies
Jan KLEINER
jkleiner@mail.muni.cz
AI in Security: Applications and Ethics
Obsah obrázku text, Písmo, Grafika, snímek obrazovky Popis byl vytvořen automaticky

Obsah obrázku socha, umění Popis byl vytvořen automaticky
Presentation outline
•
•PolSci and AI generally and its origins.
•PolSci examples of communication research.
•AI and cybersecurity.
•LAWS.
•Migration.
•Wargames and theory (preparation for the incoming seminars).
•

Obsah obrázku socha, umění Popis byl vytvořen automaticky
PolSci and AI generally (Duffy & Tucker, 1995)
1.
1.
•
•Early applications of AI in research focused on constructing choice models in foreign-policy
decision contexts.
•Other applications:
•Production systems,
•computational text analysis,
•logic programming and computer learning,
•conflict simulation and predicting outcomes in international conflicts via machine learning.
•AI + computer vision + natural language processing + sentiment analysis à set to transform
society, the economy, and politics (Efthymiou-Egleton, Egleton & Sidiropoulos, 2020).
•AI can create new ways of (researchable) communication (alphabets, iconographics, languages etc.)
(Mueller & Massaron, 2021).
•
•

Computer vision is a field of artificial intelligence that focuses on enabling computers to
understand and interpret visual information from images or videos.
Natural Language Processing (NLP) is a field of artificial intelligence that focuses on enabling
computers to understand, interpret, and generate human language. It involves developing algorithms
and techniques that allow machines to process and analyze text or speech data. NLP encompasses
tasks such as language translation, sentiment analysis, text classification, information
extraction, and question answering. It aims to bridge the gap between human language and machine
understanding.

Obsah obrázku socha, umění Popis byl vytvořen automaticky
Three examples of PolSci (communication) research
•
•
•1. Can AI communication tools increase legislative responsiveness and trust in democratic
institutions? (Kreps &  Jakesh, 2023).
•Recent.
•
•2. Artificial intelligence and European identity: the European Commission’s struggle for
reconciliation (von Essen & Osseewarde, 2023).
•Recent.
•
•3. Rise of the Machines? Examining the Influence of Social Bots on a Political Discussion Network
(Hagen et al., 2022).
•Cited (30x – SCOPUS).

Obsah obrázku socha, umění Popis byl vytvořen automaticky
AI tools and responsiveness and trust in democratic institutions (Kreps & Jakesh, 2023)
1.
1.
•
•Legislative correspondence generated by AI with human oversight may be received favorably by
constituents and increase trust and legislative responsiveness compared to generic auto-responses.
•
•Poorly performing AI may damage confidence in legislators.
•Still unclear specific impact of AI to political communication.
•
•
•Technologies like ChatGPT could streamline democratic processes rather than destabilize them à
BUT: authors do not mention dis/mis/information or propaganda threats (cf. Hagen et al., 2022).
•
•HITL and SITL concepts (Rahwan, 2018).
•

Obsah obrázku socha, umění Popis byl vytvořen automaticky
EU´s approach to AI (von Essen & Osseewarde, 2023)
•
•
•The European Commission aims to develop European version of AI, but its communication efforts may
not be sufficient to generate trust in AI among the European public.
•
•The EC frames European AI as trustworthy and human-centric, based on European values and
historical success, but fails to connect its claims to specific European values
•
•
•

European AI refers to the version of artificial intelligence that is being developed and promoted
by the European Commission (EC) as distinctively European.
European AI is an approach to AI in general not a concrete application.

Obsah obrázku socha, umění Popis byl vytvořen automaticky
Social bots´ impact on political discussion network (Hagen et al., 2022)
1.
1.
•
•Social bots (automated accounts on social media), often utilize AI techniques to generate content,
interact with users, spread information etc.
•
•Social bots can significantly impact political discussion networks by creating the appearance of
virtual communities, attenuating the influence of traditional actors, and amplifying pro-Trump
messaging.
•
•Bots are often utilized by actors with ideological positions reflective of a small subset of the
public (e.g., the far-right).
•
•The potential for spreading misinformation, which undermines democratic processes.

Obsah obrázku socha, umění Popis byl vytvořen automaticky
AI and Cybersecurity (Bonfanti et al., 2021)
•
•AI as an underdeveloped field in social sciences (AI politics research years behind the
cybersecurity politics one).
•Inter and transdisciplinary (decisions and research in one discipline transpires into other ones).
•Well suited for cyber defense and offense + influence ops.
•„…in what ways will AI enhance the protection of individuals, organizations, nations, and their
cyber-dependent assets from hostile threat actors?
•How will it introduce novel vulnerabilities and enable additional typologies of actions?
•How will it induce cyber-security stakeholders to adapt to changing risk scenarios and
opportunities?“ (p. 226).
•

Obsah obrázku socha, umění Popis byl vytvořen automaticky
LAWS (Sauer, 2021)
1.
1.
•Lethal autonomous weapons systems.
•Autonomy vs. automation – no consensus on delineation à e.g., functionalists: machine instead of
human performing the task.
•„kill chain“ = finding, fixing, tracking, selecting, and engaging the target (+ assessing the
aftereffects).
•Autonomy incl. critical functions is not new, but AI scales it up heavily.
•Incentives – no fear, emotions, fatigue, mercy, speed of (re)action etc.
•Technological, ethical, legal, strategic criticism.
•E.g., „the accountability gap“ (p. 241) – someone has to be accountable for war actions.
•
•
•
•
•

Red = critical functions by ICRC.

Obsah obrázku socha, umění Popis byl vytvořen automaticky
Migration (Everuss, 2021)
•
•
•New fields like digital migration studies.
•Digitization of borders historically led by USA and EU.
•Biometrics à „…actionable inferences about personality, intent, emotional state, social
conformity, sexual orientation, and many other… attributes“ (Crampton, 2019: 55).
•

Obsah obrázku socha, umění Popis byl vytvořen automaticky
AI and Wargames (Knack and Powell, 2023)
•
•
•Red Teaming in general (political/security/other simulations, table-tops -> identification of gaps
in a strategy, SWOT analyses, policy analyses etc.).
•Narrow (safe) usage: Repetitive tasks within sims and wargames (background info creation,
automatic translation/transcription, textual data analysis, visuals etc.).
•High-risk usage: Red team, game manager etc.
•Low cost/questionable reliability.
•Better on tactical/operational level than on the strategic one.
•

Obsah obrázku socha, umění Popis byl vytvořen automaticky
Wargame theory – introduction I (Appleget et. al, 2020)
1.
1.
•Usually a sponsor – sets goals and timeframe.
•Sole purpose is to collect analytic data to answer sponsor´s (research) questions – data determine
wargame´s success à well tought-out data collection plan is needed!
•Roadmap = data collection and management plan (DCMP).
•Not just for combat/conflict scenarios, but for Analysis of alternatives (AoA) – e.g., M1A2 Abrams
and its replacement options.
•+ pedagogic, research tool.

Obsah obrázku socha, umění Popis byl vytvořen automaticky
Wargame theory – introduction II (Appleget et. al, 2020)
1.
1.
•Course of action wargaming.
•BOGGSAT = "bunch of guys and gals sitting around a table„.
•Vs.
•Seminar wargames - designed around the DCMP (Decision-Centric Methodology Process) and have a
structured approach.
•Quantitative/qualitative/hybrid models.
•Strong role of probability and chance (dice rolls) + conditioned probability (e.g., missile
interception of Iron Dome AA system – informed by statistics).
•
•
•

Source: Appleget et al. (2020, p. 73).



The Operational Wargame Series: The best game not in stores now – No Dice No Glory
Let´s do some BOGGSAT wargame!


References I
•Appleget, J. Burks, R. & Cameron, F. (2020). The Craft of Wargaming – A Detailed Planning Guide
for Defense Planners and Analysts. Annapolis: Naval Institute Press. ISBN 9781682473771
•Bonfanti, M. E., Cavelty M. D., Wenger, A. (2021). Artificial intelligence and cyber-security. In:
Elliott, A. (Ed.). (2021). The Routledge Social Science Handbook of AI (1st ed.). Routledge.
https://doi.org/10.4324/9780429198533.
•Duffy, G., & Tucker, S. A. (1995). Political Science: Artificial Intelligence Applications. Social
Science Computer Review, 13(1), 1–20. https://doi.org/10.1177/089443939501300101.
•Efthymiou-Egleton, I. P., Egleton, T. W. E., & Sidiropoulos, S. (2020). Artificial Intelligence
(AI) in Politics: Should Political AI be Controlled?. International Journal of Innovative Science
and Research Technology, 5(2).
•Everuss, L. AI, Smart Borders and Migration. In: Elliott, A. (Ed.). (2021). The Routledge Social
Science Handbook of AI (1st ed.). Routledge. https://doi.org/10.4324/9780429198533.
•
Obsah obrázku socha, umění Popis byl vytvořen automaticky

References II
•Hagen, L., Neely, S., Keller, T. E., Scharf, R., & Vasquez, F. E. (2022). Rise of the Machines?
Examining the Influence of Social Bots on a Political Discussion Network. Social Science Computer
Review, 40(2), 264–287. https://doi.org/10.1177/0894439320908190
•Knack, A. & Powell, R. (2023). Artificial Intelligence in Wargaming An evidence-based assessment
of AI applications. The Alan Turing Institute.
•Kreps, S., & Jakesch, M. (2023). Can AI communication tools increase legislative responsiveness
and trust in democratic institutions? Government Information Quarterly, 40(3), 101829.
https://doi.org/10.1016/j.giq.2023.101829
•Mueller, J. P. & Massaron, L. (2021). Artificial Intelligence for Dummies. Hoboken: John Wiley and
Sons, Inc.
•Sauer, F. (2021). Lethal autonomous weapons systems In: Elliott, A. (Ed.). (2021). The Routledge
Social Science Handbook of AI (1st ed.). Routledge. https://doi.org/10.4324/9780429198533.
•von Essen, L., & Ossewaarde, M. (2023). Artificial intelligence and European identity: the
European Commission’s struggle for reconciliation. European Politics and Society, 1–28.
https://doi.org/10.1080/23745118.2023.2244385
•
Obsah obrázku socha, umění Popis byl vytvořen automaticky

Obsah obrázku socha, umění Popis byl vytvořen automaticky
Questions?
Jan KLEINER
jkleiner@mail.muni.cz
Thank you for your attention.
Obsah obrázku text, Písmo, Grafika, snímek obrazovky Popis byl vytvořen automaticky