Table 1 Government roles, challenges and... | Emerald Publishing

Table 1

Government roles, challenges and solutions to promote social equity within and through AI

Role	Challenges	Possible solutions	References
Regulator Ethics and normative oversight	Conflict between AI development and public values Perpetuation of biases and discrimination Lack of transparency and accountability Privacy and data protection Misrepresentation surrounding AI capabilities	Adopt a proactive, context-specific approach to AI regulation Establish robust accountability mechanisms (explainability, transparency, fairness) Engage disadvantaged groups in AI adoption and governance Implement legally binding regulations on social equity Promote accurate narratives reflecting AI’s capabilities and limitations	Abiteboul and Stoyanovich (2019), Aizenberg and Van Den Hoven (2020), Alnemr (2023), Aoki et al. (2024), Arnesen et al. (2024), Bodó and Janssen (2022), Busuioc (2021), Cabrera-Medina et al. (2024), Chakraborty and Bhojwani (2018), De Almeida et al. (2021), Delfos et al. (2024), Engstrom and Haim (2023), Gaozhao et al. (2023), Grewal et al. (2024), Grimmelikhuijsen and Meijer (2022), Grimmelikhuijsen (2023), Guenduez and Mettler (2023), Guevara-Gómez et al. (2021), Haraguchi et al. (2024), Hjaltalin and Sigurdarson (2024), Ingram (2020), James et al. (2023), Jobin et al. (2019), Jones and McKelvey (2024), Kaur et al. (2023), Khan et al. (2024), Law and McCall (2024), Mac (2024), Mahmoudi et al. (2025), Margetts et al. (2024), Marjanovic et al. (2022), Minow (2023), Mittelstadt et al. (2016), Nicolás and Sampaio (2024), Nwafor (2021, 2024), Nzobonimpa and Savard (2023), Oravec (2019), Plantinga (2024), Robinson (2020), Robles and Mallinson (2023), Ruschemeier and Hondrich (2024), Selten and Meijer (2021), Sha et al. (2024), Smith and Desrochers (2020), Smith and Miller (2023), Taylor et al. (2024), Turner Lee (2018), Ulnicane et al. (2021), Van Toorn and Carney (2024), Vandersluis and Savulescu (2024), Walker (2024), Zajko (2022)
User Equitable Integration	Erosion of human discretion and professional judgment Amplification of existing inequities Ethical concerns and value trade-offs Technological limitations and risks (misinformation, hallucinations, biases)	Maintain a hybrid human-machine model (human-in-the-loop) Prioritize citizen engagement in AI design and implementation Conduct impact assessments focused on social equity Navigate ethical concerns through stakeholder involvement and alignment with public values Promote algorithmic literacy and engage communities in AI design and oversight	Alon-Barkat and Busuioc (2023), Aysolmaz et al. (2023), Balayn et al. (2021), Baykurt (2022), Bélisle-Pipon et al. (2023), Chen et al. (2023), Criado et al. (2021), Dankloff et al. (2024), Dekker et al. (2022), Desiere and Struyven (2021), Fountain (2022), Hermstrüwer and Langenbach (2023), Hoff (2023), Horvath et al. (2023), Johnson et al. (2022), Kaplan and Haenlein (2019), Karippur et al. (2020), König and Wenzelburger (2021), Kuberkar et al. (2022), Levy et al. (2021), Li (2024), López et al. (2024), McNamara and Tikka (2023), Moon (2023), Nam and Bell (2024), Ozmen Garibay et al. (2023), Papalexopoulos et al. (2022), Park and Humphry (2019), Peeters (2020), Piñeiro-Martín et al. (2023), Purdy and Glass (2023), Ramos-Maqueda and Chen (2025), Ranerup and Henriksen (2022), Ratner and Thylstrup (2024), Reeves et al. (2023), Rodolfa et al. (2021), Roehl and Hansen (2024), Schiff et al. (2022), Selten et al. (2023), Shin et al. (2020), Sidhu et al. (2024), Van Toorn and Scully (2023), Varona et al. (2021), Waldman and Martin (2022), Wang et al. (2024), Young et al. (2019)
Enabler Inclusive Enablement	Opacity, complexity and accountability challenges Societal trust and ethical concerns Communication and interpretability of AI	Translate abstract principles into concrete, enforceable requirements Foster expertise in AI development and management Employ AI for policy simulation to anticipate societal needs	Arnaout et al. (2023), Bernhard and Wihlborg (2022), Bonomi Savignon et al. (2024), Buslón et al. (2023), Cole et al. (2022), Fernandez-Aller et al. (2021), Gibbons (2021), Jin and Ryu (2025), Kaun (2022), Madan and Ashok (2023), Margetts (2022), Pah et al. (2022), Saldanha et al. (2022), Williams et al. (2022)
Leader Participatory AI Leadership	Negative influence of large technology companies Ethical, legal and human rights challenges Need for culturally sensitive and contextual AI implementation	Decouple digitalization from marketization and regulate Big Tech influence Build trust with communities through collaboration and information with local organizations Ensure benefits accrue also to the Global South	Barn (2019), Dobell and Zussman (2018), Eom et al. (2024), Hülter et al. (2024), Manjarrés et al. (2021), Qu and Wang (2024), Rehill and Biddle (2024), Sharon and Gellert (2024), Taylor (2024), Trajkovski (2024), Valle-Cruz et al. (2024), Westerstrand (2024)

Role	Challenges	Possible solutions	References
RegulatorEthics and normative oversight	Conflict between AI development and public values Perpetuation of biases and discrimination Lack of transparency and accountability Privacy and data protection Misrepresentation surrounding AI capabilities	Adopt a proactive, context-specific approach to AI regulation Establish robust accountability mechanisms (explainability, transparency, fairness) Engage disadvantaged groups in AI adoption and governance Implement legally binding regulations on social equity Promote accurate narratives reflecting AI’s capabilities and limitations	Abiteboul and Stoyanovich (2019), Aizenberg and Van Den Hoven (2020), Alnemr (2023), Aoki et al. (2024), Arnesen et al. (2024), Bodó and Janssen (2022), Busuioc (2021), Cabrera-Medina et al. (2024), Chakraborty and Bhojwani (2018), De Almeida et al. (2021), Delfos et al. (2024), Engstrom and Haim (2023), Gaozhao et al. (2023), Grewal et al. (2024), Grimmelikhuijsen and Meijer (2022), Grimmelikhuijsen (2023), Guenduez and Mettler (2023), Guevara-Gómez et al. (2021), Haraguchi et al. (2024), Hjaltalin and Sigurdarson (2024), Ingram (2020), James et al. (2023), Jobin et al. (2019), Jones and McKelvey (2024), Kaur et al. (2023), Khan et al. (2024), Law and McCall (2024), Mac (2024), Mahmoudi et al. (2025), Margetts et al. (2024), Marjanovic et al. (2022), Minow (2023), Mittelstadt et al. (2016), Nicolás and Sampaio (2024), Nwafor (2021, 2024), Nzobonimpa and Savard (2023), Oravec (2019), Plantinga (2024), Robinson (2020), Robles and Mallinson (2023), Ruschemeier and Hondrich (2024), Selten and Meijer (2021), Sha et al. (2024), Smith and Desrochers (2020), Smith and Miller (2023), Taylor et al. (2024), Turner Lee (2018), Ulnicane et al. (2021), Van Toorn and Carney (2024), Vandersluis and Savulescu (2024), Walker (2024), Zajko (2022)
UserEquitable Integration	Erosion of human discretion and professional judgment Amplification of existing inequities Ethical concerns and value trade-offs Technological limitations and risks (misinformation, hallucinations, biases)	Maintain a hybrid human-machine model (human-in-the-loop) Prioritize citizen engagement in AI design and implementation Conduct impact assessments focused on social equity Navigate ethical concerns through stakeholder involvement and alignment with public values Promote algorithmic literacy and engage communities in AI design and oversight	Alon-Barkat and Busuioc (2023), Aysolmaz et al. (2023), Balayn et al. (2021), Baykurt (2022), Bélisle-Pipon et al. (2023), Chen et al. (2023), Criado et al. (2021), Dankloff et al. (2024), Dekker et al. (2022), Desiere and Struyven (2021), Fountain (2022), Hermstrüwer and Langenbach (2023), Hoff (2023), Horvath et al. (2023), Johnson et al. (2022), Kaplan and Haenlein (2019), Karippur et al. (2020), König and Wenzelburger (2021), Kuberkar et al. (2022), Levy et al. (2021), Li (2024), López et al. (2024), McNamara and Tikka (2023), Moon (2023), Nam and Bell (2024), Ozmen Garibay et al. (2023), Papalexopoulos et al. (2022), Park and Humphry (2019), Peeters (2020), Piñeiro-Martín et al. (2023), Purdy and Glass (2023), Ramos-Maqueda and Chen (2025), Ranerup and Henriksen (2022), Ratner and Thylstrup (2024), Reeves et al. (2023), Rodolfa et al. (2021), Roehl and Hansen (2024), Schiff et al. (2022), Selten et al. (2023), Shin et al. (2020), Sidhu et al. (2024), Van Toorn and Scully (2023), Varona et al. (2021), Waldman and Martin (2022), Wang et al. (2024), Young et al. (2019)
EnablerInclusive Enablement	Opacity, complexity and accountability challenges Societal trust and ethical concerns Communication and interpretability of AI	Translate abstract principles into concrete, enforceable requirements Foster expertise in AI development and management Employ AI for policy simulation to anticipate societal needs	Arnaout et al. (2023), Bernhard and Wihlborg (2022), Bonomi Savignon et al. (2024), Buslón et al. (2023), Cole et al. (2022), Fernandez-Aller et al. (2021), Gibbons (2021), Jin and Ryu (2025), Kaun (2022), Madan and Ashok (2023), Margetts (2022), Pah et al. (2022), Saldanha et al. (2022), Williams et al. (2022)
LeaderParticipatory AI Leadership	Negative influence of large technology companies Ethical, legal and human rights challenges Need for culturally sensitive and contextual AI implementation	Decouple digitalization from marketization and regulate Big Tech influence Build trust with communities through collaboration and information with local organizations Ensure benefits accrue also to the Global South	Barn (2019), Dobell and Zussman (2018), Eom et al. (2024), Hülter et al. (2024), Manjarrés et al. (2021), Qu and Wang (2024), Rehill and Biddle (2024), Sharon and Gellert (2024), Taylor (2024), Trajkovski (2024), Valle-Cruz et al. (2024), Westerstrand (2024)

Source(s): Table by authors