人工智能时代的性别能量关系：挑战与机遇

Abbreviations Acknowledgements This report was written by Ava Strasser and Rosemary Idem. It wascommissioned by the Sustainable Energy for All (SEforALL) Gender &Youth Team. The SEforALL Gender & Youth Team is led by RosemaryIdem, with Akil Callendar, Charles Mankhwazi and Ava Strasser. This report received insightful comments from peer reviewers. We wouldlike to thank Nathyeli Acuna Castillo of the Energy Sector ManagementAssistance Program (ESMAP), Sheila Oparaocha and Elizabeth Cecelskiof ENERGIA and Birouke Teferra of World Resources Institute. Valuable guidance and oversight were provided by Damilola Ogunbiyi,CEO and Special Representative of the UN Secretary-General (SRSG) forSustainable Energy for All. We would like to thank SEforALL staff for their support: Kanika Chawla,Neil Claydon, Tracey Crowe, Brian Dean, Cristina Dominguez, RosaGarcia, Stephen Kent, Divya Kottadiel, Mikael Melin, Emi Mizuno, JennyNasser, Nishant Narayan, Luc Severi and Elisabeth Strasser-Müller. We acknowledge with gratitude the financial support provided by theAustrian Development Agency and the Ministry for Foreign Affairs ofIceland. Table of Contents Executive Summary4 Challenges11 Opportunities & responsibilities23 Diversifying AI development teamsAccelerating women’s representation in STEMImplementing gender-responsive decision-making & policy standardsApplying a gender lens to AI & energy solutionsUnlocking gender & energy financing2324272830 Key recommendations32 Executive Summary The intersection of gender, energy and Artificial Intelligence(AI) presents both challenges and opportunities forachieving gender equality and sustainable development.AI can be a critical enabler in accomplishing 134 of the169 targets under the framework of the SustainableDevelopment Goals (SDGs), with over 600 AI-enableduse cases identified.1However, the impact of AI at theintersection of SDG7 (Affordable and Clean Energy) andSDG5 (Gender Equality) requires significant attention. While AI shows positive potential for supporting SDG7by ensuring universal access to affordable, reliable,sustainable and modern energy for all, SDG5 has thelowest number of AI-enabled use cases, with only 10out of approximately 600 cases identified.2This disparityis concerning considering that lack of energy accessdisproportionately affects women and girls.3UN Womenhas reported that if current trends continue, by 2030,an estimated 341 million women and girls will still lackelectricity, with 85 percent of them in Sub-Saharan Africa.4 KEY CHALLENGES: These funding gaps limit the potential for AI to addressgender-specific needs in the energy sector, risking the exclusion ofwomen from technological advancements and hindering progresstowards both SDG7 and SDG5. —Energy access gap:Currently, 685 million people worldwide lackaccess to electricity, with Sub-Saharan Africa most affected.Additionally, 2.1 billion people lack access to clean cookingtechnologies.5The lack of modern energy access disproportion-ately affects women and girls, impacting their health, education,economic opportunities and ability to leverage AI applications. —Diversion of energy from communities:AI data centres located inlow-resource areas can exacerbate existing inequalities, as thesecommunities often lack the political and economic power to advo-cate for their energy and water needs. This perpetuates energypoverty and marginalization, disproportionately affecting womenwho manage household energy needs and are more vulnerable toenergy scarcity. —Gender data deficit:The lack of sex-disaggregated data in theenergy sector results in AI applications that fail to adequatelyaddress the specific needs and experiences of women. —Digital gender divide:Women are disproportionately affected bylimited access to digital technologies, the internet and electricityrequired for online connectivity. Men are 21 percent more likely tobe online than women, with the gap widening to 52 percent inleast developed countries.6 There are policy opportunities that can ensure an equitable and genderjust energy transition supported and facilitated by AI. These include:expanding the representation of women in STEM roles within the energyand AI-related sectors; creating diverse and gender-balanced AI devel-opment teams; increasing women’s participation in energy policy anddecision-making; establishing environmental regulations and ethicalframeworks for the use of AI in energy; ensuring AI and energy solutionsare developed and implemented through a gender-responsive lens; andapplying a gender lens to investment in the energy and technology sectors. —Workforce gender imbalance:Women are underrepresented inthe energy and AI sectors, especially in technical and leadershiproles, and overrepresented in care work and the informal econ-omy. —AI development bias:The lack of diversity in AI developmentteams leads to technologies that reflect and perpetuate genderbiases. By implementing these recommendations and ensur