《安永（EY）：2025人工智能与社会流动性：释放青年群体发展潜能的机遇研究报告（英文版）（16页）.pdf》由会员分享，可在线阅读，更多相关《安永（EY）：2025人工智能与社会流动性：释放青年群体发展潜能的机遇研究报告（英文版）（16页）.pdf（16页珍藏版）》请在三个皮匠报告上搜索。

1、Artificial intelligence and social mobility:an opportunity to unlock the potential of young people January 20251|Artificial intelligence and social mobility:an opportunity to unlock the potential of young people 1.Introduction2.The opportunities of AI3.Mitigating the risks of AI4.Young people partic

2、ipation5.Next stepsContents2Artificial intelligence and social mobility:an opportunity to unlock the potential of young people|1.Googles impact on the UK 2023(2023),Public First2.Firm-level adoption of AI and automation technologies:Case Studies Report IFOW,Institute for the Future of Work,September

3、 20243.Innovation Generation:Next Steps for Social Mobility(2024),Social Mobility CommissionArtificial intelligence(AI)is transforming how we learn and work,with improvements in innovation and productivity potentially adding 400 billion to the UK economy by 2030.1 For young people,especially those f

4、rom low-income backgrounds,AI brings new opportunities,offering the chance to level the playing field by filling educational gaps,increasing accessibility and providing tailored learning experiences that were previously out of reach.It also holds the potential to eradicate the subtle implicit bias t

5、hat serves as a barrier for too many talented young people during the recruitment process.This isnt to suggest technology represents a magic wand that will make inequality disappear.But it is a powerful tool that can be used to unlock human potential irrespective of someones background.Harnessing th

6、is opportunity will often require targeted action and is accompanied by challenges that,if not addressed,could hinder progress and increase barriers to social mobility.The EY Foundation(EYF)is a charity that supports young people from a low-income background through employability skills training and

7、 is committed to ensuring AI becomes a force for positive change.Working at the intersection between young people,education and employment,we view AI not just as a tool for innovation,but a powerful means to advance social mobility,reduce barriers and open new pathways to success.Whilst the impact o

8、f AI on entry-level jobs,a crucial gateway for young people from low-income backgrounds,remains unclear,fears that AI will replace large sections of the workforce appear to be overstated.2 Concerns about bias,privacy and inequality are valid but having the right governance guardrails work towards th

9、eir mitigation.With the right approach,AI can uplift human potential rather than exclude or replace future generations.Collaboration and participation are key to realising AIs potential.Governments,businesses,educators and civil society must work together to create frameworks that make AI inclusive

10、and equitable.By engaging a wide range of voices,including young people and their communities,AI can be developed in a way that expands opportunities for all.Proactive efforts to guarantee transparency,fairness and accountability can mitigate the risks of deepening inequalities,amongst other potenti

11、al harms.By focusing on the practical implications and opportunities of innovation,EYF like other organisations in our sector is exploring how emerging technologies can enhance the support we provide to young people.And it is clear these tools offer exciting ways to engage,inspire and educate.Techno

12、logy can make learning more personalised and relevant for those who might struggle in traditional systems and it can be used to improve employability skills and provide targeted careers knowledge,guidance and support,addressing some of the most pressing challenges young people face today.This is an

13、approach that aligns with the Social Mobility Commissions position,as set out in its 2024 Innovation Generation3 and Innovation 1.Introduction 3|Artificial intelligence and social mobility:an opportunity to unlock the potential of young people and social mobility report:“The policy priority should b

14、e to leverage the power of innovation to enhance social mobility.Slowing down the pace of technology diffusion would be likely to reduce labour demand and generate unemployment,especially if other countries adopt innovations faster.”4However,to fully realise the promise of new technologies like AI,i

15、ntentional action is needed.Rather than simply naming the opportunities and risks,we must constructively address them together.By fostering collaboration across sectors,including young people themselves,we can shape AI to enhance human potential and drive social mobility.Through collaborative and pu

16、rposeful action,we can cement AI as a tool to help create a more inclusive and equitable future,where all young people,regardless of background,can thrive.EYF is committed to playing a central role in this mission,ensuring that AI serves as a catalyst for positive change,expanding opportunities for

17、all.Whilst responsible AI entails a consideration of the implications of the AI ecosystem on different aspects of diversity,equity and inclusion(DEI),there has been less focus placed on the specific impact that AI is having on social mobility.Using the core beliefs we set out,this report responds to

18、 this gap by combining our experience of delivering employability skills training to young people from a low-income background with a review of recent AI developments,research and policy initiatives from the UK government and more widely.Part two highlights the opportunities of AI to transform socia

19、l mobility,Part three identifies the risks,Part four considers how the insights of young people can be harnessed to establish AI as a technology that supports social mobility and we set out our next steps in Part five.With the speed of technological change,this short report can only provide a snapsh

20、ot of the situation at the time of writing(January 2025),but we hope the core beliefs guiding our approach and placing young people at the heart of the evolution of technology will remain critically important over the coming months and years.By bringing greater attention to this opportunity,our ambi

21、tion is to fast-track the action needed to ensure all young people can thrive in an increasingly technology dominated future.Our work in the field of AI is guided by three core beliefs:1.The design,development and deployment of AI can be shaped to help ensure it serves society 2.The development of A

22、I can augment human capacity 3.Social mobility can be enhanced,not diminished by AI4.Innovation and Social Mobility:Two sides of the same coin,(2024),Social Mobility Commission.4Artificial intelligence and social mobility:an opportunity to unlock the potential of young people|Whilst it is difficult

23、to predict what opportunities AI might bring in the future,the governments AI Opportunities Action Plan says it can drive economic growth,improve public services,address regional disparities,promote shared economic prosperity and increase personal opportunities.5 We see three core areas where AI cou

24、ld potentially help young people from low-income backgrounds.These revolve around the potential for personalised learning,a renewed focus on humancentric skills and the opportunity to increase social mobility.Some might argue that AI poses a threat to those already on the margins of society,who stru

25、ggle to achieve social mobility.However,there is also the real possibility that AI can have an outsized benefit for these sections of society if the opportunity is taken to develop the structure and tools that can make a positive difference.Personalised learningAI has the potential to democratise ac

26、cess to education in a way not seen since the introduction of universal education.AI can be used to personalise education,adapting learning materials to meet the unique needs of individual students.This could be particularly beneficial for learners who lack access to personalised support,such as pri

27、vate tutoring or tailored attention in smaller class sizes.AI-powered tools can provide students with real-time feedback,enabling tailored study at a pace designed for each student.These technologies can create individualised learning pathways that cater to each students strengths,challenges and lea

28、rning style,offering a level of personalisation that was previously unattainable for the vast majority of students.For students with disabilities or those learning in a second language,AI offers a wide range of tools that significantly enhance accessibility.Text-to-speech,speech recognition and adap

29、tive learning systems help break down barriers to education,enabling those with learning differences to better access and engage with educational content.In situations where schools lack the time or resources to assist students applying for jobs or further education,particularly for students requiri

30、ng tailored assistance,AI can help with the job application process,helping to draft and edit applications,identify transferable skills,prepare for interviews and clarify language.6 This ability to accommodate diverse learning needs makes education and the gateway to work more inclusive,allowing stu

31、dents who might struggle in conventional systems to excel.Creating time for humancentric learningEducation should remain centred on human connection,with AI serving as a tool to support teachers,not replace them.As AI becomes more integrated into classrooms and workplaces,the need for humancentric s

32、kills like empathy,communication,teamwork and cultivating a growth mindset to remain agile amidst workplace transformations,is increasing.These essential skills are best developed through direct,face-to-face engagement,yet many teachers,especially in under-resourced schools,are overwhelmed by large

33、class sizes and administrative burdens.If implemented effectively,AI can alleviate some of these challenges,offering meaningful support to both students and educators.AI can take 2.The opportunities of AI5.AI Opportunities Action Plan-GOV.UK(2025)6.Levelling the playing field(2024),Movement to Work.

34、5|Artificial intelligence and social mobility:an opportunity to unlock the potential of young people on time-consuming tasks such as lesson planning and administrative duties,enabling teachers to focus on the personal engagement crucial for emotional and social skills development.Research from the N

35、ational Foundation for Educational Research found that ChatGPT reduced lesson planning time by 31%with no noticeable reduction in quality in comparison to a non-GenAI control group,7 demonstrating the huge potential for saving valuable teacher time.The UK governments announcement of the AI Opportuni

36、ties Action Plan in January 2025 includes plans to do just that,utilising AI to“drive down admin for teachers so they can get on with teaching our children”.8 By automating administrative tasks,AI empowers teachers to spend more time where it matters most:fostering meaningful relationships and suppo

37、rting student growth.At the same time,The Alan Turing Institute is clear that we should be careful to guard against developing a path dependency on AI tools to plug gaps in under-resourced schools,which could lead to a two-tier system where some students are denied access to quality in-person teachi

38、ng.9 Increasing social mobility For young people who often face systemic barriers moving from education to employment and then onwards within the work environment,AI could offer a new set of tools to access opportunities that have previously been out of reach.As we explore AIs capacity to deliver pe

39、rsonalised learning experiences,we need to look at its application beyond the classroom.In some workplaces,research shows that AI training could significantly improve performance for those who have not had the opportunity to gain the skills needed to thrive in their role.10 Using AI can achieve an e

40、qualisation of outcomes,allowing a young person to develop at a disproportionately quicker rate than their previously higher performing peers.Augmented AI enhancing human capabilities using AI could be core to enabling all young people to fully unlock their potential in the workplace.AI also holds t

41、he potential to provide tailored advice for pathways to further education and future careers.By assessing individual strengths,interests and academic achievement alongside advice direct from professionals AI tools can suggest relevant career paths and specialised training and apprenticeship opportun

42、ities,aligned with emerging fields and suited to each young persons unique profile.For many young people from low-income backgrounds,one of the most significant barriers to social mobility is a lack of access to professional networks,personalised mentorship and guidance offered to more affluent stud

43、ents either in an elite school environment or through family networks.AI could potentially democratise access to this information offering all young people the same level of tailored support,fostering greater social mobility.One of the most promising uses of AI is its ability to forecast the evolvin

44、g skills landscape and map out future job opportunities.As industries increasingly integrate AI and automation,the demand for certain skills could shift dramatically.This will be a challenge for the education system,the labour market and young people,particularly those who already struggle to overco

45、me the barriers to greater social mobility.However,AI could hold a potential solution and AI tools are already being trialled that can process vast amounts of data from job advertisements,industry trends and public data to predict emerging fields and the competencies required for future employment.1

46、1 By analysing these patterns,AI could offer young people,educators and employers real-time insights into which skills 7.ChatGPT in Lesson Preparation Teacher Choices Trial(2024),National Foundation for Educational Research.8.Prime Minister sets out blueprint to turbocharge AI-GOV.UK(2025)9.Generati

47、ve AI in Education:Call for Evidence.The Alan Turing Institute.10.Workers with less experience gain the most from generative AI|MIT Sloan11.Expanding AI Adoption Can Help Create Jobs(2023),Stanford University.6Artificial intelligence and social mobility:an opportunity to unlock the potential of youn

48、g people|and areas are in most demand,ensuring they are well-positioned for the jobs of tomorrow.AI-enabled analysis of data can also be used to assess specific impacts on marginalised groups,identifying emerging patterns of bias as they develop that can then be rectified through data-informed actio

49、n.It is encouraging too that the governments AI Opportunities Action Plan highlights the need to assess and then fill the UK skills gap,drawing on successful practices from countries like Singapore and South Korea,which are incorporating AI,data,and digital literacy throughout their education system

50、s.The plan identifies Skills England and the Curriculum and Assessment Review as opportunities to potentially adopt these approaches,ensuring all young people are equipped with the skills they need to thrive.12It is also the case that AI could be deployed to reduce human bias and increase fairness i

51、n the processes of hiring and promotion.This could represent a huge victory for social mobility and level the playing field for young people from underrepresented backgrounds.It could also represent a boost for employers with more equitable recruitment processes,contributing to the wider success of

52、organisations that are hiring based on skills and competencies rather than traditional markers,such as educational background or socioeconomic status.12.AI Opportunities Action Plan-GOV.UK7|Artificial intelligence and social mobility:an opportunity to unlock the potential of young people Whilst AI o

53、ffers huge potential,it also brings significant risks,both real and perceived.The first barriers a young person might encounter are access to technology and core digital skills.The affordability of digital tools,such as computers,internet access and software in addition to the skills,knowledge and s

54、upport needed is denying some the opportunity to reap the benefits of technology.One response,set out in a report by the Good Things Foundation and the University of Liverpool,proposes a Minimum Digital Living Standard13 and calls for public,voluntary and private sector collaboration to reduce the d

55、igital barriers many are confronted by.Looking beyond access,risks span privacy concerns,algorithmic bias and the potential to deepen existing inequalities,particularly for young people who are already the most monitored generation in history.14 As AI becomes more integrated into sectors like educat

56、ion and employment,it is crucial that these technologies are implemented responsibly,with a focus on equity and protection for vulnerable populations.Privacy and data protection for young peopleAIs ability to collect and analyse vast amounts of data raises significant privacy concerns,particularly f

57、or young people.Both parents and students are increasingly uneasy about how personal information might be collected and used,especially as young people and parents have limited understanding of how AI may impact them,both in the short and long term.15 Concerns about data misuse,whether by private co

58、mpanies or government bodies,are likely to persist.These fears,including the potential for long-term data tracking that could affect future opportunities,may slow AI adoption among the very communities that stand to benefit most.Addressing these concerns requires technical,ethical and regulatory sol

59、utions,alongside a commitment to collaboration and responsible development involving the communities that AI tools are designed to serve.Regulators in the UK and worldwide are addressing these issues.Probably the most comprehensive attempt to date is the EUs Artificial Intelligence Act,16 which impo

60、ses strict rules and some prohibitions on high-risk AI systems,demanding transparency about the data used,particularly for general purpose AI(GPAI).This safeguards sensitive information and maintains compliance with data protection standards like GDPR,which emphasises data minimisation,user consent

61、and clear communication about AI decisions.In the UK,regulators are prioritising transparency and accountability with the promise of greater regulation to come.Already,developers must explain how personal data is processed and comply with the UKs Data Protection Act.AI-driven systems must undergo da

62、ta protection impact assessments(DPIAs)17 to identify risks before deployment,with any issues reported to the Information Commissioners Office(ICO)for 3.Mitigating the risks of AI13.The Minimum Digital Living Standard|Good Things Foundation14.AI,Childrens Rights,&Wellbeing:Transnational Frameworks.(

63、2023)The Alan Turing Institute.15.Responsible Technology Adoption and Department for Education Research on Public Attitudes Towards the Use of AI in Education(2024),Thinks Insight&Strategy,Responsible Technology Adoption and Department for Education.16.Artificial Intelligence Act,Shaping Europes Dig

64、ital Future.The European Union.17.Data Protection Impact Assessments(DPIAs),Information Commissioners Office.8Artificial intelligence and social mobility:an opportunity to unlock the potential of young people|oversight.18 To further protect privacy,techniques like synthetic data generation and addin

65、g noise to datasets are being implemented.These methods,whilst not perfect if used poorly,enable AI systems to function effectively whilst minimising exposure to personal information.The action that is already underway is encouraging but not an end point.As activity grows in this area,it will be cru

66、cial to ensure it fully addresses the opportunity to accelerate social mobility.This is explored in more detail in part four below.Tackling algorithmic bias and inequalityAI has the potential to break down barriers to social mobility,reducing educational disparities and addressing biases that hinder

67、 hitherto excluded communities.However,there is growing evidence that AI,if not carefully managed,can also exacerbate inequalities through“algorithmic bias.”19 This occurs when AI systems are trained on incomplete or biased data,leading to the unintentional reinforcement of stereotypes or discrimina

68、tory practices.There are legitimate concerns that for students from marginalised racial,ethnic,or low-income backgrounds,algorithmic bias in automated or semi-automated assessments or admissions processes could result in unfair outcomes,favouring those better represented in data sets.Similarly,in re

69、cruitment and job placement,AI could perpetuate discrimination against underrepresented groups,deepening existing inequalities if not properly monitored.20 To address these risks,it is essential that AI systems are transparent,unbiased and subject to regular audits.Proper oversight is critical to ce

70、menting AI as a tool for equity,not exclusion.Regulators around the world are working to mitigate these risks.The EUs AI Act mandates regular evaluations of high-risk AI systems to confirm they do not reinforce societal inequalities.Developers are required to use representative,unbiased data,particu

71、larly in sensitive application areas like education and employment.The Act also explicitly prohibits AI systems that exploit vulnerabilities related to age,disability,or socioeconomic status,safeguarding young people and vulnerable groups.In the UK,The Responsible Technology Adoption Unit(RTA)21,for

72、merly known as the Centre for Data Ethics and Innovation(CDEI)plays a pivotal role in promoting ethical AI development,with a focus on protecting marginalised communities.The UK regulatory framework emphasises that AI systems must be inclusive,transparent and sensitive to the needs of minority and v

73、ulnerable populations.Regulators are actively involved in ensuring AI developers consider the societal and ethical impacts of their technologies,with initiatives in place to assess and mitigate the negative effects of AI on these communities.By working together,regulators and AI developers in the pr

74、ivate sector can undoubtedly reduce the risks associated with AI now and in the future.However,the only true way of guaranteeing that AI will serve the interests of society is for society to be actively engaged in its development.18.Information Commissioners Office19.Measuring Implicit Bias in Expli

75、citly Unbiased Large Language Models(2024)Bal,X.Wang,A.Sucholutsky,I.Griffiths,T L.Cornell University.20.Using AI Responsibly in People Management(2023).CIPD.21.The Centre for Data Ethics and Innovation(CDEI)/Responsible Technology Adoption Unit(RTA)9|Artificial intelligence and social mobility:an o

76、pportunity to unlock the potential of young people In the absence of knowing fully how AI will evolve or the full purposes to which it will be deployed,we believe there is one area above all others that all groups working with marginalised communities should focus on and advocate for participation a

77、nd collaboration.This also applies to young people from low-income backgrounds across the UK.As stated in a publication by EY on AI in the context of the United Nations sustainable development goals“to harness AIs full potential for inclusive development,we must close the digital gaps and foster mea

78、ningful participation from marginalised communities.”22Looking more widely than technology,the UK government recognises that the views of young people should be put at the heart of decision making on areas of policy that impact them.To achieve this goal it has committed to establishing a National Yo

79、uth Strategy.23There is a compelling need for AI tools to be developed and delivered through a participatory process,especially when those tools are aimed for use by young people.It should become best practice that AI developers shape technology and its deployment by directly involving those for who

80、m the technology is intended.By establishing marginalised groups at the heart of AI design and delivery,we not only address issues of fairness but also increase the effectiveness of AI in solving real-world problems.The case for participation of young people in AI developmentThe importance of engagi

81、ng with children and young people in the development of AI tools has been raised by numerous organisations,notably The Alan Turing Institute,which in its research on AI,childrens rights,and wellbeing,emphasised the need for,“engagement with children to ensure that future approaches are informed by t

82、he real experiences,concerns and the interests of children and to maximise their relevance and value for children.”24 If this is true of children in general,it is even more the case for young people from low-income backgrounds,who often experience unique barriers that can only be effectively address

83、ed by those who understand them firsthand,as evidenced in Partnership for Young Londons review of participatory approaches in the youth sector.25 By involving young people directly in the design and delivery process,AI developers could create AI solutions that are more likely to meet these pressing

84、needs.Participation ensures that AI solutions are not just theoretically inclusive but practically relevant.This will also make sure variations in barriers and opportunities such as those connected to location and the intersectionality of other characteristics such as gender and ethnicity are recogn

85、ised and responded to.Without such involvement,there is a risk that AI could exacerbate existing inequalities by failing to consider these groups lived experiences.4.Young people participation22.Inclusive innovation:An inside look at AIs potential to achieve the SDGs(September 2024),Devex and EY.23.

86、National Youth Strategy to Break Down Barriers for Young People GOV.UK24.AI,Childrens Rights,&Wellbeing:Transnational Frameworks(2023),The Alan Turing Institute.25.Involving Young Londoners:A Review of Participatory Approaches in the Youth Sector(2020),Partnership for London.10Artificial intelligenc

87、e and social mobility:an opportunity to unlock the potential of young people|Many of the current concerns about the risks AI could pose could be alleviated through a collaborative approach.If young people are invited to contribute to the design and development of AI systems,they could highlight blin

88、d spots that developers,businesses and educators may overlook.Bringing young people into the process in a meaningful way is crucial to solidify AI as a force for empowerment not division,ensuring AI becomes an opportunity to address past exclusion and overcome the barriers to social mobility.The nee

89、d for participation and a collaborative approach in the field of AI is not limited to the sphere of education but is also firmly on the agenda in the workplace.Organisations like The Institute for the Future of Work(IFOW)and Movement to Work,highlight the critical importance of employee engagement i

90、n the development and deployment of AI tools in the workplace.Building trust through inclusive participation As surveys have revealed,trust is a fundamental factor in the success and adoption of AI technologies,26 particularly among communities that have historically been marginalised.For AI to be t

91、ruly effective,especially when deployed in sensitive areas like education or employment,the people who will use or be affected by these systems need to trust the technology.Building confidence,trust and transparency in AI processes is essential to safeguarding vulnerable populations and ensuring tha

92、t AI solutions are embraced rather than resisted.Connected to these points is the significance of intrinsic motivation,which means making sure all young people feel that AI is relevant to someone like me and feel empowered to fully engage in the new opportunities it might provide.This was identified

93、 as a potential barrier for young people from a low-income background in our 2023 report,Social Mobility in the Metaverse.27Including these groups in the development of AI tools could foster trust by giving them a sense of ownership over the technology.When people see the input of their community re

94、flected in a tools design,they are more likely to trust and adopt it.This participatory approach also builds long-term relationships between developers,implementors and communities,creating a feedback loop for ongoing refinement of AI systems based on real-world experiences.Trust is built not just t

95、hrough the accuracy of the technology but through transparency in the development process and responsiveness to the needs of those who are most affected.A model for inclusive participation in AI developmentThere are a growing number of organisations that are thinking about how civil society,right do

96、wn to the grassroots community level,can and should be engaged in the development and implementation of AI tools.The steps set out below,are informed by this thinking,in particular the work undertaken by the IFOW and its publication,The good work algorithmic impact assessment,a partnership approach,

97、28 and by Nestas paper,Participatory AI for humanitarian innovation.291.Identify and engage key stakeholders:Young people,especially those from a low socio-economic background,should be engaged through representative groups as primary stakeholders in the AI development process.This could involve dir

98、ect outreach to youth groups and community organisations working 26.Responsible Technology Adoption and Department for Education Research on Public Attitudes Towards the Use of AI in Education(2024),Thinks Insight&Strategy,Responsible Technology Adoption and Department for Education.27.Social Mobili

99、ty in the Metaverse(2023),EY Foundation.28.The Good Work Algorithmic Impact Assessment A Partnership Approach(2024),Institute for the Future of Work.29.Participatory AI for humanitarian innovation(2021),NESTA.11|Artificial intelligence and social mobility:an opportunity to unlock the potential of yo

100、ung people with these populations.We will take an active convening role to bring these organisations together to act as intermediaries,helping to identify young people whose voices are often ignored and including them in the process.This wont require technical expertise from young people,instead the

101、 focus will be to highlight how developments in AI might impact their development from education into employment.2.Establish meaningful participatory channels:There is a growing body of research into how participation can occur at multiple levels,from consultation to full co-creation.EYF takes an ac

102、tion-led approach to generating insights,with examples including the Social Mobility in the Metaverse and Breaking Barriers30 reports.The former explored how virtual reality focused technology could help young people to thrive in the workplace and the latter brought young people and employers togeth

103、er to identify practical ways to overcome the specific employment barriers facing young people from a low-income background.A structured approach could involve:Consultation:Initial consultations could take the form of focus groups or interviews,where young people are asked to share their perspective

104、s on the challenges they face and the role AI could play in solving them.This stage is crucial for identifying the key problems that need addressing.Collaboration:Young people could be actively involved in specific stages of the AI development and deployment process,such as data collection or user t

105、esting.Their input at this stage ensures that the AI tools being developed are grounded in the realities of their lives,and helps to nurture trust regarding the use of data.Co-creation:In the most comprehensive form of participation,young people could work alongside developers in designing the AI sy

106、stems from the ground up.This process might involve hackathons,innovation labs,or regular workshops where developers and young people work together to develop AI solutions.3.Provide training and build capacity:Not every young person needs to be an expert in the building and design of AI systems,but

107、all should have the opportunity to acquire the skills they need to engage with AI tools and ensure they are the masters of,not the servants to,this technology.Barriers to digital literacy need to be dismantled.For young people to participate meaningfully in AI development,investing in their digital

108、skills is required.This means providing affordable access to the internet,computers and digital training programs.Organisations like EYF already provide AI skills training for young people and other community groups are increasing access to digital devices and connectivity such as the Digital Povert

109、y Alliance.31 Working together,organisations in this space can play an important role,offering training workshops and resources to make sure marginalised young people can engage effectively in AI discussions.4.Develop feedback loops and continuous engagement:To establish participation as meaningful

110、and not tokenistic,it is essential to create ongoing mechanisms for feedback.These could include:Community audits:After deployment,the groups to whom the AI tool is aimed could help audit the performance of AI tools to ensure they are delivering on their intended outcomes without introducing new for

111、ms of bias.This would provide a mechanism for continuous improvement.30.Breaking Barriers:Increasing Employment Opportunities for Young People from Low-income Backgrounds in Greater Manchester(2023),EY Foundation.31.Digital Poverty Alliance.12Artificial intelligence and social mobility:an opportunit

112、y to unlock the potential of young people|Advisory panels:Setting up advisory boards that include young people representatives would cement their ability to continue to shape AI governance and oversight.5.Build long-term collaboration:Building trust and making sure that AI serves marginalised commun

113、ities requires long-term collaboration.Partnerships between AI developers,third-sector organisations and sections of society should be seen as long-term engagements,not one-off projects.Regular check-ins,ongoing training and collaborative refinement of AI systems and deployments are essential to mai

114、ntaining trust and ensuring that the technology continues to meet the evolving needs of the communities it serves.13|Artificial intelligence and social mobility:an opportunity to unlock the potential of young people AI will reshape the educational and employment landscape for young people and this c

115、an be to the benefit of young people from low-income backgrounds.However,the potential to personalise learning,foster social mobility and break down systemic barriers,must be the path that is actively chosen over a version of AI that would pose risks to privacy,algorithmic bias and the possibility o

116、f deepening existing inequalities.To fully realise positive impacts of AI,collaboration across government,educators,employers,civil society and young people themselves,is essential.In addition to the development of a model for young people participation,we will be taking two immediate next steps in

117、response to the insights captured in this report.Intrinsic motivation researchWe are working with the Institute for the Future of Work,young people from a low-income background and employers,to gain a deeper understanding of the practical training and support needed to develop humancentric skills an

118、d increase intrinsic motivation.AI and social mobility grand challengeThis will be a purpose-driven competition that invites entrants from the tech sector and more widely to tackle a specific question designed to identify how AI can unlock improvements in educational,skills and employment outcomes f

119、or young people from a low-income background.It will be a collaborative project,bringing together researchers,tech developers,educators and policymakers to forge a future where technologies and humans work in concert to unlock new realms of potential.5.Next stepsPlease get in touch to find out more

120、about:1.Bringing the perspective of young people into the development of AI.2.Our research into humancentric skills and intrinsic motivation.3.Supporting an AI and social mobility grand challenge.We welcome the opportunity to work with organisations who share our determination to ensure technology i

121、s harnessed to drive an increase in social mobility across the UK.Email:14Artificial intelligence and social mobility:an opportunity to unlock the potential of young people|About the EY FoundationThe EY Foundation is a UK registered charity that works directly with young people,employers and social

122、entrepreneurs to create or support pathways to education,employment or enterprise.EY Foundation operates and is incorporated independently of EY and is governed by a separate trustee board.The EY Foundation is a charitable company registered in England and Wales and Scotland with registered charity number 1157154 and SC045076.It is also a member firm of Ernst&Young Global Limited.The EY Foundation,1 More London Place,London SE1 2AF 2025 The EY Foundation.Published in the UK,All Rights RED NoneUKC-037230.indd(UK)01/25.Artwork by Creative UK.