Future of Smart Cities Security Challenges – Proactive Modelling & Identification Zlatogor Minchev1,2,* , Luben Boyanov3 1 Institute of ICT, Bulgarian Academy of Sciences, Acad. Georgi Bonchev Str., Bl. 25A, 1113 Sofia, Bulgaria 2 Institute of Mathematics & Informatics, Bulgarian Academy of Sciences, Acad. Georgi Bonchev Str., Bl. 8, 1113 Sofia, Bulgaria 3 University of National & World Economy, 8-mi Dekemvri Str. 19, 1700 Sofia, Bulgaria Abstract Joining technologies & people in future smart cities infrastructure by merging sensors, effectors and intelligence is going to create a rather challenging mixed reality transformation. In this sense, the competition between natural and artificial intelligence is inevitably establishing quite new and interesting society overlaying of humans and technologies with federated domination areas. The results are presently addressing the digital society transformation towards Society 5.0, whilst outreaching the next Society 6.0 expectations. The paper is going to outline a comprehensive analytical intelligence framework (i-framework) for studying the problem, adding a scenario-based proactive analysis, combined with system modelling and results hybrid multicriteria validation. The intelligent part comes from different AI models that are implemented in the process, giving supportive and generative added values. Finally, a concluding discussion on the outlined findings is presented. Keywords Future smart cities, digital society transformation, security challenges, scenario-based analysis, system modelling, hybrid multicriteria validation 1. Introduction divide between citizens in the smart urban reality and the rest of the populated areas. They are going to be consid- Digital transformation is expected to affect in practice all ered as a new digital class with advanced capabilities but fields of future society reality, including people and their will be also challenged via joint human-machine threats residence area, adding also biotope dynamics (to note: cli- in the smart habitat [6]. This definitely will affect future mate changes, species migration, natural disasters, etc.) jobs and culture transformation, together with deeper [1]. As for the new urban environment of future smart smart machines, sensors and algorithms integration in cities, the process is certainly expected to combine new the transformed people’s lifestyle and environment smart IoTs & AI, providing innovative commodities, services (to reshaping. In this context, the security, privacy and eth- mark: transportation, deliveries, education, governance, ical issues require an adequate and smart exploration media, energy supplies, assistance, medicine, economics) approach that is proactively organized as to be earlier and jobs for the citizens, aiming the horizon towards prepared as a civilization for this change. the year 2050 [2]. New smart gadgets’ autonomous in- Further, an exploration methodological approach in tegration (multifunctional robots, vehicles, etc.) with this context is going to be outlined, combining both nat- the vastly interconnected reality (due to broadband wire- ural and artificial intelligence with expert analytical sup- less meshes & optical network technologies enhanced port. usage) will additionally advance the new habitual digital landscape [3]. Thus, the digital change towards the post-information 2. Analytical "i-Framework" age is expected to have both - positive and negative trans- formational effects on the new Society 5.0 idea [4]. The Combining human & machine intellect into a joint ana- situation is getting even more complicated with Society lytical power is practically extending some of the ideas 6.0 transcends exploration [5], where AI and machine from [1, 7] into a new "i-framework" (see Figure 1), better singularity are expected to appear in practice. applicable to the digital future security dynamics, and So, new technologies are going to establish a digital comprehensive proactive exploration. The extension was organized due to the vast dynam- BISEC’23: 14th International Conference on Business Information ics’ escalation with AI & IoTs immersion in the post- Security, November 24, 2023, Niš, Serbia information age, as stated in the introductory part of the * Corresponding author. paper that is difficult to be easily handled due to the scale $ zlatogor@bas.bg (Z. Minchev); lboyanov@unwe.bg (L. Boyanov)  0000-0003-2479-5496 (Z. Minchev); 0009-0006-2292-0619 and dynamics with human only intellect. (L. Boyanov) While aiming proactive identification and comprehen- © 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). sive analysis of digital transformation security in general, CEUR Workshop Proceedings http://ceur-ws.org ISSN 1613-0073 CEUR Workshop Proceedings (CEUR-WS.org) CEUR ceur-ws.org Workshop ISSN 1613-0073 Proceedings Figure 2: Future smart cities security transcendents towards Figure 1: Analytical "i-framework" for digital future security year 2037. proactive exploration. representation in I–SCIP-SA environment [9] has been the current "i-framework" implementation is addressing performed. the future smart cities in particular. Being a broad land- Taking an aggregated analytical representation into a scape example for security dynamics studying with peo- "3D Sensitivity Diagram" – "3D SD" with Influence – x, ple & technologies digital transformation, the outlined Dependence – y & Sensitivity – z, due to relations i-fuzzy findings could be further used with the broader digital weights, concerning future smart cities security towards society evolution deeper exploration. year 2037 (simulated in 10 steps) is finally achieved with Further in the paper a detailed illustration of the trilat- 16 entities (addressing social – yellow, technical – blue eral "i-framework", concerning (2.1) Proactive Analytical & mixed – white aspects) & 41 bi-directional relations Modelling, (2.2) Hybrid Multicriteria Validation & (2.3) model (see Figures 3 and 4). Mixed Reality Assessment will be given for the future The resulting classification gives four classes for the smart cities’ comprehensive security challenges context model entities distribution (with two subclasses for each: exploration. Active – white & Passive – grey) in the 3D SD diagram as follows: 2.1. Proactive Analytical Modelling Buffering (in green): "Social Credits Score" – 16, being at the same time Passive. Achieving proactive analytical modelling is combining Active (in red): "Smart Infrastructure" – 5, "Trans- both morphological and system analysis approaches. A formed Life" – 8, "Mixed Intelligence" – 11, all being starting implementation of the scenario method, with Active. expert and reference data, towards the establishment Critical (in yellow): "Super Humans" – 3, "Human of plausible and implausible scenario combinations is Preservation" – 15, both being Passive & "Autonomous accomplished. The result is a cross-consistency matrix AI" – 2, "Data Leakages" – 6, "New Jobs" – 7, "Smart 𝑀 , containing three types of scenarios, in accordance Communication" – 9, "Criminal Activities" – 13 all being with their Relative Common Weight – 𝑅𝐶𝑊 : Active Active. (tangible), Passive (intangible) & Neutral (probably most Passive (in blue): "Privacy Concerns" – 1, "World i- uncertain) [8]. With the present study on future smart Domination" – 4, "AI Regulations" – 10, "Hardware Com- cities, the particular matrix context towards year 2037 promising" – 12, "Law Enforcement" – 14 all being Pas- and post-information society of 136080 scenarios [1] has sive. been zoomed for smart cities security topic, with a total Further, the presented quantitative classification re- scenario number 𝑁 * = 2880(𝑁 * = 5×3×4×3×4×4; sults could be aggregated in more detail, around several plausible – 𝑁 1* = 86 & implausible ones – 𝑁 2* = key findings: 2794; from 𝑁 1* are additionally selected: Active, i.e. – (i) That future smart cities will create numerous threats "tangible" (76, 𝑅𝐶𝑊 > 0) & Passive, i.e. – "intangible" and challenges from their critical infrastructure perspec- (10, 𝑅𝐶𝑊 < 0). tive [10], that could be attacked with different vectors: As this landscape shows quite an uncertain future with communicational, data & hardware ones. Apart of this, mostly implausible scenarios, a deeper analytical causal- the future superhumans will have a somewhat ambiguous ity exploration has been performed toward smart city role with new technologically extended capabilities. So, system sensitivity analysis. insider security threats are neither to be completely ex- A "system-of-systems" modelling paradigm over an cluded or added by default due to potential technological i-fuzzy weighted graph-based "Entity – Relationship" Figure 3: Future smart cities security transcendents of a discrete system-of systems model. scenarios with machine-controlled social credits and be- haviour that normally are a question of culture and social system respect, whilst trying to keep a non-authoritarian but secured future urban reality. As the presented expert findings are mostly based on expert analytical beliefs, further AI-assisted validation and assessment will also be given, trying to achieve a comprehensive urban security landscape exploration. An overall evaluation of the exercise has been per- formed by the participants (with Positive either Indefinite judgment marks) via a q-based survey, giving feedback for: reality, scenario & interawaring complexity, AI & human factor roles, training satisfaction (see Figure 6). Being somewhat subjective the obtained results could Figure 4: Resulting 3D SD analytical assessment towards be also enriched with biometric & simulation assessments year 2037. (see [1, 9]) and finally combined within the system model. More details on these ideas will be given further. influence. 2.2. Mixed Reality Assessment (ii) New jobs and ambitions for intelligent world domi- nation will certainly appear and progress both from pos- The mixed reality assessment of the accomplished ana- itive and negative (criminal, manipulative) perspectives, lytical findings (see Section 2.1) was further conducted, adding AI & IoT with new capabilities and challenges; using a transformed reality interactive simulation, orga- (iii) Mixing artificial and natural intelligence for future nized in the framework of CYREX 2023 [11]. Assuming smart cities’ security could be quite beneficial except a fictitious scenario events script (generated with human if a superintelligence with negative objectives manages intellect guiding & tailoring Open AI ChatGPT results), to compromise the system due to emergency external interactively played (for about 180 minutes) from the influences (natural or man-made disasters). trainees in several multirole teams, an exploration of (iv) Keeping privacy and humanity present understand- future smart cities security transendents was performed. ings will be quite different for the future as the role of AI The main objective of CYREX 2023 exercise was to test & IoT transformation will also demand new ethical and the human-machine inter-awareness in an imaginary social boundaries. context, concerning the future smart cities mixed real- All these findings will hopefully omit the dystopian ity from different aspects (both utopian and dystopian Figure 7: Socio-technological S-shaped curve quantum repre- sentation, after [1]. people (Y- & Z- generations), joining both human and machine intelligence in the process of decision-making and scenario development, while using a role-based orga- nization, multiple smart gadgets (smartphones, laptops, advanced PCs, smart TVs & interactive screens) and plat- forms (Windows, Android, iOS). The training was illus- trated, combining results of artificially generated images, videos, sounds and popular multimedia clips (assisted with Gencraft, beatovenAI & invideoAI). The approach allows studying of complex security transcendents dy- namics in a futuristic mixed reality smart ecosystem, giving excellent training feedback results, especially at organizational and operational levels. 2.3. Hybrid Multicriteria Validation The main idea for this section is to combine both hu- Figure 5: Selected moments (a, b) & organizational architec- man and AI expectations for the future of smart cities’ ture (c) of CYREX 2023 [11]. security, taking as a base the presented in 2.1. Analyt- ical i-Framework findings in the dynamic context. So, a joint smart approach has been further accomplished, adding human beliefs vs machine-adaptable smart multi- criteria optimization [10]. In this manner, it is possible to get a feasible evaluation of potential future expec- tations, taking into account trends S-shaped dynamics, but with unplanned jumps (see Figure 7) that could be best explained with quantum tunnelling effect stochastic socio-technological modelling of system model relations dynamics [1]. Selected multicriteria near future illustrative critical entities (towards the year 2037, see Figure 8), concerning Figure 6: Aggregated participants’ q-based assessment of the system-of-systems discrete model of smart cities se- CYREX 2023. curity transcendents (see Figure 3) are further presented, taking the risk-assessment approach from [12], but using a percentage-based measuring scale. ones), concerning the cyber security area different as- So, taking society the system-of-systems model ideas pects: technological, social, infrastructural, security, po- is quite intriguing as normally the model is both subjec- litical, governance, diplomatic. The idea was to develop tive to the expert beliefs and at the same time – limited and test a set of morphological and system models for the due to the preliminary analytical assessment. In this not-so-far future (10-15 years from now), among young sense, the validation process has been extended with helpful with the analytical models’ more detailed assess- ments. As for the role of AI, it is still at an early stage of development and the dream for a “General AI” is still quite limited. However, it should be honestly marked that the generative AI on the other hand, is quite sup- portive in the analytical assessment and experimental issues. Thus, providing both a neutral opinion advisor and a rapid prototyping tool that facilitates the explo- ration efforts’ proactive nature a lot. This clearly shows a positive technological trend for the not so far digital future new social evolution. Acknowledgment The results presented in this study are due to the tech- nological, industrial and expert support obtained in the framework of the international forum initiative “Securing Digital Future 21” with more than sixty countries now, spread around the world, https://securedfuture21.org/ References [1] Z. Minchev, et al., Digital Transformation in the Post-Information Age, 1st Edition, SoftTrade & In- stitute of ICT, Bulgarian Academy of Sciences, Sofia, 2022. [2] United Nations, Envisaging the Future of Cities, World Cities Report 2022, UN Human Settlements Figure 8: Selected multicriteria smart cities security future entities illustrative assessment (after Figure 3) from human – Programme, https://unhabitat.org/, 2022. H & machine – M perspectives towards 2037. [3] L. Boyanov, igital World – The Change, 1st. ed., Avangard Prima, Sofia, 2021. [4] G. Wahlers (Ed), The Digital Future, International Reports, Konrad Adenauer Stiftung, No. 1, https: "Other" additional relations and entities. This additional //goo.gl/8CLcvn, 2018. extension could be produced from both human and gen- [5] S. Bousri, Embracing Society 6.0: A Technological erative AI hints and supplementary human discussions. Renaissance for Human Living and Economies, Elit With the presented illustrative examples an added hu- Web3 Solutions, https://www.linkedin.com/pulse/ man opinions value has been taken from SRS’2023 young embracing-society-60-technological-renaissance-human-living-bousri-1f/, international participants in the context of Society 6.0 2023. [13]. The machine-generated added value was produced [6] United Nations, Addressing the Digital Di- by taking Open AI ChatGPT feedback with the human vide Taking Action towards Digital Inclu- responses’ extension, towards machine domination, con- sion, United Nations Human Settlements cerning smart cities’ security near future evolution. Programme (UN-Habitat), https://unhabitat.org/ programme/legacy/people-centered-smart-cities/ addressing-the-digital-divide, 2021. 3. Conclusion [7] Z. Minchev, Proactive identification of future cyber Proactively identifying the future smart cities’ security threats, in: Proceedings of the 13th BISEC Confer- challenges is a quite complex task that could benefit from ence, 2022, pp. 42–49. both human and machine intelligence joint efforts. Go- [8] Z. MINCHEV, Future transformational outlook for ing deeper in the problem normally requires a suitable the digital society and economy, Romanian Cyber framework as has been already shown in the present Security Journal. Fall 2 (2021) 25–37. study. Whilst human intelligence is always subjective [9] Z. B. Minchev, Human factor role for cyber threats by nature responsive biometric feedback could be quite resilience, in: Handbook of Research on Civil So- ciety and National Security in the Era of Cyber Warfare, IGI global, 2016, pp. 377–402. [10] Z. Minchev, Security challenges to critical infras- tructure of future smart cities, in: Proceedings of the 9th BISEC Conference, 2019. [11] CYREX 2023 Multimedia Clip, https://youtu.be/ m7mTfvtmtFc, 2023. [12] Z. Minchev, Malicious Future of AI: Transcendents in the Digital Age, in: Proceedings of the 12th BISEC Conference, 2019, pp. 18–22. [13] International Research Summer School on Math- ematics & Informatics - SRS’23, Multimedia Report, https://www.globaldiplomatic.eu/post/ international-research-summer-school-on-mathematics-informatics-srs-23, 2023.