On 8 October 2026 Legislative Decree No 179 of 7 October 2026 was published in Official Gazette (Gazzetta Ufficiale) No 234; it aligns national legislation with Regulation (EU) 2024/1689 (AI Act) “as regards the powers of the national Authorities and the use of artificial intelligence in education and training” (our translation) and enters into force on 23 October 2026 (Official Gazette (Gazzetta Ufficiale)). We reported on the content announced after final approval on 4 August in our commentary on the Italian implementation of the AI Act. Chapter III devotes Articles 36 to 39 to schools and adult education, and Article 36(5) requires educational institutions to carry out guidance initiatives to accompany “female and male students” (le studentesse e gli studenti) towards higher education pathways in STEAM disciplines, including those connected with artificial intelligence. Article 47 of the same decree makes AI training mandatory in the courses run by the professional bodies, a subject we dealt with in discussing mandatory AI training for professionals. We examine what the decree says, and what it does not say, about the gender gap in these pathways.
Article 36: STEAM modules, guidelines and the duty to provide guidance
The structure of Article 36 alternates between powers and duties. Paragraph 2 allows schools to promote, within their curricula, modules for strengthening “scientific, technological, engineering, mathematical and artistic (STEAM)” skills through the use of artificial intelligence tools in teaching. Paragraph 3 instructs the Ministry of Education and Merit (MIM) to update the guidelines for STEM disciplines adopted by Ministerial Decree No 184 of 15 September 2023, “fostering the integration of artificial intelligence into the scientific disciplines and into those in the artistic field”. Paragraph 4 calls on the Ministry of Labour and Social Policies to contribute to the analysis of the professional needs connected with STEAM disciplines and artificial intelligence, also for the purposes of the “informed guidance of students”.
Paragraph 5 is the central provision. Educational institutions “shall adopt specific guidance initiatives”, including within the formative guidance modules provided for lower and upper secondary schools, aimed at accompanying female and male students “in the informed choice of higher education and training pathways relating to STEAM disciplines” and at increasing knowledge of their applications, “including those connected with artificial intelligence and emerging technologies”. For the same purposes the Ministry ensures the revision and updating of the guidelines for guidance adopted by Ministerial Decree No 328 of 22 December 2022 (Official Gazette (Gazzetta Ufficiale)). Paragraph 2 says “may promote”, paragraph 5 says “shall adopt”: unlike the STEAM modules, guidance towards those disciplines is a duty of schools, albeit one to be exercised within their autonomy.
Gender in the decree: one textual formula and two indirect references
We searched the published text for the terms “studentesse” (female students), “genere” (gender), “parità” (equality) and “divario” (gap). The first appears, in the articles, only in Article 36(5); the second is found solely in the editorial notes reproducing other laws: the decree names female students in a single provision and never names gender. The formula “female and male students” takes up the delegation criterion of Article 24(2)(g) of Law No 132 of 23 September 2025, which called for the strengthening of STEM skills “in order to promote the choice by female and male students” of higher education pathways, “also through targeted personalised guidance activities”.
Gender enters the decree by two indirect routes. The first is Article 3, which declares the decree adopted “in compliance with the general principles contained in Article 3 of the AI law”: among those principles Law No 132/2025 lists “non-discrimination” and “equality of the sexes”. The second is the reference in Article 36(3) to the 2023 STEM guidelines, which make the gender gap an explicit objective. As for the equal opportunities clause in Article 38(3)(d), it concerns the reduction of the digital divide and access to emerging skills “with particular attention to persons in conditions of socio-economic disadvantage or with special educational needs”: gender is not among the criteria indicated.
The ministerial guidelines: the 2023 STEM document and the 2022 guidance document
Ministerial Decree No 184/2023 is among the measures implementing the National Recovery and Resilience Plan (PNRR) that the Ministry announced in Official Gazette (Gazzetta Ufficiale) No 88 of 15 April 2024 (Official Gazette (Gazzetta Ufficiale)). The annexed guidelines refer to the investment “New skills and new languages”, whose purpose is to develop STEM skills across all school cycles “with the objective of encouraging enrolment in tertiary STEM curricula, in particular for women”; they state that the funded actions pay “particular attention to overcoming gender gaps in access to STEM careers”; they allow actions addressed to families to be promoted, on the occasion of the International Day of Women and Girls in Science, to encourage participation “mainly of female pupils and female students, overcoming gender stereotypes”; and they affirm that it is through guidance actions towards those disciplines “that gender equality in the field of education can be promoted” (Ministry of Education and Merit).
The case of the 2022 guidelines for guidance, which Article 36(5) requires to be revised, is different: in the text published by the Ministry we found no explicit reference to gender, to female students or to stereotypes (Ministry of Education and Merit). The document to be revised is precisely the one that does not currently contain the gender objective, while the document that does contain it is to be updated with the sole mandate of integrating artificial intelligence into it.
Eurostat data: female ICT specialists and female graduates in computing disciplines
According to Eurostat, in 2025 women accounted for 19.5% of ICT specialists employed in the Union, and in Italy there were 150,500 female specialists (Eurostat). The Italian share, according to the Eurostat estimate, is 16.6% in 2025, down from 17.1% in 2024 (Eurostat). On the education side, in 2024 tertiary education graduates in information and communication technologies numbered 8,787 in Italy, of whom 1,596 were women; in the Union 223,623, of whom 49,334 were women (Eurostat). Eurostat itself recalls that the European target of 20 million employed ICT specialists by 2030 is coupled with a greater convergence towards gender balance.
The OECD picture points in the same direction: we examined the data on female ICT specialists in OECD countries, on the use of artificial intelligence at work and on the programming skills of young people in presenting the OECD Forum 2026 document on artificial intelligence and gender gaps. We consider the link with the decree to be direct: paragraph 5 intervenes on the choice of higher education pathways, and the graduate data show the imbalance already on leaving those pathways, even before entry into employment.
Critical aspects: rebalancing objectives, indicators, resources and timing
The first aspect concerns objectives. Article 36 sets no objective for female participation in STEAM pathways, nor does it require guidance initiatives to take gender stereotypes into account. That is the legislative datum; the assessment is ours: the formula “female and male students” is an inclusive use of language, not a measure, and a provision that does not name the gap places on no one the obligation to reduce it. The link with the STEM guidelines makes up for this only in part, because paragraph 3 calls for their updating without saying whether the gender objective is to be retained.
The second concerns measurement. The only monitoring mechanism with indicators provided for in this area is that of Article 44(6) and (7), for universities and higher education institutions for art, music and dance (AFAM): annual monitoring by the Ministry of University and Research, with indicators defined by ministerial decree after consulting the National Agency for the Evaluation of Universities and Research Institutes (ANVUR). For schools and for the guidance under Article 36 the decree provides for no indicators, and in neither case does it require sex-disaggregated data. In our view the decree under paragraph 7 is the place to introduce them: without sex-disaggregated data it will not be possible to verify whether guidance has had an effect on the choices of female students.
The third concerns resources and timing. Article 54 provides that, without prejudice to Article 37(2), the administrations concerned carry out the activities provided for by the decree with the human, instrumental and financial resources available under current legislation and without new or greater burdens on public finances (Official Gazette (Gazzetta Ufficiale)). The exception concerns only the training plan for teachers on the use of social media and digital platforms, to which Article 37(2) allocates up to EUR 100 million from the National Programme “PN Scuola e Competenze 2021-2027”. Guidance towards STEAM disciplines has no funding of its own: the duty under paragraph 5 falls on schools without any additional expenditure. Nor does Article 36 set a deadline for updating the two sets of guidelines, unlike the ninety days within which Article 37 requires the teacher training plan to be launched; this also emerges from the overview of the decree’s deadlines proposed by Nicola Fabiano (NicFab).
A final observation concerns the tools. If the guidance under paragraph 5 makes use of artificial intelligence systems, the AI Act classifies as high-risk, in Annex III, point 3(c), those intended to be used for the purpose of “assessing the appropriate level of education that an individual will receive or will be able to access”, and recital 56 warns that such systems, when improperly designed and used, may “perpetuate historical patterns of discrimination, for example against women” (EUR-Lex). Nicola Fabiano brings within that point the guidance platforms that recommend pathways on the basis of the student’s profile (NicFab). The teacher training under Article 38 and the minimum university content under Article 44 include knowledge of the risks and “distortions” of the systems; we addressed the issue of algorithmic discrimination, and of the limits of the reservation of human decision-making alone in detecting it, with regard to Law 132/2025 and algorithmic bias. The decree, in our view, treats distortions as training content and links them neither to guidance tools nor to the composition of those who will design those systems.
Conclusions
From 23 October 2026 educational institutions are required to adopt guidance initiatives towards STEAM disciplines and artificial intelligence addressed to female and male students; the Ministry of Education and Merit must revise the guidelines for guidance and update those for STEM disciplines; the Ministry of University and Research must define the monitoring indicators for universities. What is missing is a rebalancing objective, gender-related indicators, resources dedicated to guidance and a deadline for updating the guidelines. We consider that the revision of the guidelines for guidance should expressly incorporate the gender objective of the STEM guidelines, that the updating of the latter should retain it, and that the decree on the indicators under Article 44 should provide for sex-disaggregated data. Schools, in designing the initiatives under paragraph 5, can already refer to the STEM guidelines in force, which identify overcoming gender stereotypes as one purpose of guidance.
Author: Valentina Grazia Sapuppo
