Artificial intelligence: education and impacts on children and young people

Ben Smith; Devyani Gajjar

Overview

Although still in an early and experimental phase, in the past few years there have been advancements in artificial intelligence (AI) tools that can be used in education, with some AI tools beginning to be used in UK schools (PN 712).[1]

These advancements have been partly driven by developments in AI to generate realistic content, such as text, images, videos, audio and other media, known as generative AI (PB 57).

Some experts responding to POST’s horizon scan highlighted how these technologies could present opportunities, such as AI providing different ways of learning, but also risks, including around copyright and reducing learners’ ability to problem-solve by themselves.

As AI becomes increasingly embedded in society, these experts also highlighted the growing importance of education about AI for those in education, and the general population, to safeguard against ethical harms, and to support people to gain AI skills increasingly needed for the workforce.

Challenges and opportunities

AI in education

AI has various opportunities to be used by teachers in education, such as:[2]^,[3],[4],[5]

drafting ideas for lesson plans and other activities
producing tailored learning information for students with different needs
automating repetitive administrative tasks
reducing workloads, including marking

AI also has various opportunities for learners, such as:¹^,²^,³^,[6],[7]

formulating ideas, such as creating essay structures
assisting with language barriers
providing accessible and affordable learning support including feedback
developing critical thinking, creativity, curiosity and collaboration skills

However, both teachers and learners may lack the necessary skills, support, and technologies to use AI in their work.²

AI tools have various limitations, such as some models being liable to generate inappropriate, politically and culturally biased or factually incorrect information (PN 712). Some stakeholders have expressed concerns that biased algorithms used for marking or moderating assessments could lead to unfair results.[8]^,[9] AI tools can make use of material that might not be cited, or cited incorrectly, and this could raise risks, such as copyright infringement (PN 708).

Education institutions can collect vast amounts of sensitive data about students, such as academic records, and AI systems may require access to this information to assist with tasks.[10]^,[11] This carries risks of data breaches potentially exposing personal information, or inappropriate uses of data, including for surveillance.⁶^,[12]

Some experts have raised concerns that AI could reduce learners’ ability to problem-solve by themselves and could stifle the development of some skills, for example the use of AI to generate programming code can benefit a learner if used to help support their understanding of how to code, but could inhibit learning if it instantly produces workable code.[13]^,[14]

Advances in generative AI’s ability to produce realistic sounding written text has led to widespread concerns about students using it in their work or in assessments (PN 712). Detecting the use of AI in students work is extremely challenging because whilst AI-detection tools are constantly improving, so are generative AI capabilities (PN 712).

Differing access to infrastructure needed to use AI tools, such as the internet or computers, known as ‘digital divides,’ could result in differing access to benefits of AI in education (PN 712).

Impacts on children and young people

Depending on how AI is used, some academics have also raised concerns around children and young people using AI frequently such as:⁹^,[15]

young people having fewer interactions with educators and classmates
AI exposing young people to inappropriate or harmful content
AI algorithms intensifying mental health issues such as addictions, loneliness, isolation and lack of connections

Some academics are researching social ways of using generative AI in education, that could aid learners to share goals, perform tasks together and converse to reach a mutual understanding.[16] Examples include generative AI generating alternative ways of expressing an idea or helping groups to research and solve problems together.¹⁶

Education about AI

Education and guidance about both the potential, and challenges, of AI could help people to safeguard against ethical harms, such as biased algorithms, accidental plagiarism or privacy concerns.[17]^,[18],[19]

Some universities and policy institutes have highlighted a need for AI skills development to be embedded across higher education to better prepare students for a workplace where they may increasingly experience it.[20]^,[21],[22] Some universities have also stated the importance of them issuing clear guidance to staff and students on how they should use AI.¹⁵

Education about AI could also support people to get the skills they need for the workforce and contemporary life in an increasingly digital society.⁷ There is a growing demand for specialised skills in AI in the UK workforce (PN 697), and 2023 data from the Office for National Statistics showed around one in six UK organisations using at least one AI technology.[23]

Some academics and education providers have highlighted how AI and digital literacy education for young children can help to prepare them to be informed digital citizens, use online information and resources responsibly, and foster creativity and problem-solving skills.[24]^,[25] Some academics have highlighted how early education about AI could tackle the ‘digital divide’, in which many people are increasingly being digitally excluded due to factors including poverty or a lack of infrastructure.¹⁸

Key uncertainties/unknowns

Adoption of AI in education is still in an early and experimental phase and there are many and evolving uncertainties:

There is uncertainty about the benefits and limitations and how AI might impact the future of education, teaching and learning (PN 712).
It is uncertain how students may be using AI in education, and how it may be impacting assessment results.
Whilst there are various studies examining the effectiveness of AI in specific scenarios in education,[26] there is a lack of reliable evidence on the effectiveness of wide-scale adoption of generative AI tools for education (PN 712).
There are uncertainties around what sort of relationships with technologies might be fostered in children and young people from using AI in education, and how this may vary across different contexts (PN 712).

Key questions for Parliament

What steps is the government taking around the use of AI in education?
How can AI tools in education be regulated to mitigate concerns and protect staff and students, such as from viewing inappropriate content, not widening the digital divide, and ensuring students’ personal data is protected?
Are there sufficient regulations to protect children and young people from the harms of AI? How can protection be strengthened?
How should oversight of potential AI usage by students during examinations and coursework be managed and by who, such as teachers, school administration or inspectors such as Ofsted?
Should learning about AI be part of the national curriculum, and if so, at what stage?
How can teachers be educated and kept up to date with developments in AI to help teach children and young people use it ethically and effectively?
Should the general population be educated about AI? If so, how and by who?
How should the effectiveness of AI technologies in education be evaluated and by who?

References

[1] Department for Education (2024). Generative AI in education: Educator and expert views.

[2] JISC (2024). Generative AI – a primer.

[3] Singh, K. (2024). What are the opportunities and challenges for AI in education? Birchwood University.

[4] University of Oxford (2024). Generative AI: the potential of AI in education. What does Claude* say?

[5] Kem, D. (2022). Personalised and Adaptive Learning: Emerging Learning Platforms in the Era of Digital and Smart Learning. International Journal of Social Science And Human Research, Vol5, 03, pages 385-391.

[6] Resnick, M. (2024). Generative AI and Creative Learning: Concerns, Opportunities, and Choices. MIT.

[7] Walter, Y. (2024). Embracing the future of Artificial Intelligence in the classroom: the relevance of AI literacy, prompt engineering, and critical thinking in modern education. International Journal of Educational Technology in Higher Education, Volume 21, 15.

[8] Hunton, S. (online). What the A Level Grade Scandal Can Show Us about Algorithmic Bias.

[9] Jones, E. et al. (2020). Can algorithms ever make the grade? Ada Lovelace Institute.

[10] McNulty, N. (2024). AI and Data Privacy in Schools: Safeguarding Student Information. Medium.

[11] European Commission (2022). Ethical guidelines on the use of artificial intelligence (AI) and data in teaching and learning for Educators.

[12] Buck, D. (2023). AI is a serious threat to student privacy. Thomas Fordham Institute.

[13] Kem, D. (2022). Personalised and Adaptive Learning: Emerging Learning

Platforms in the Era of Digital and Smart Learning. International Journal of Social Science and Human Research, Vol 5, 02, pages 385-391.

[14] Tobin, J. (2023). Educational technology: Digital innovation and AI in schools. House of Lords Library.

[15] Robinson, L. et al. (2024). Social Media and Mental Health.

[16] Sharples, M. (2023). Towards social generative AI for education: theory, practices and ethics. Learning: Research and Practice, Volume 9, 2, pages 159-167.

[17] Kandlhofer, M., et al. (2023). Education and Awareness for Artificial Intelligence. In: Pellet, JP., et al. (eds) Informatics in Schools. Beyond Bits and Bytes: Nurturing Informatics Intelligence in Education. ISSEP 2023. Lecture Notes in Computer Science, Vol 14296, Springer, Cham.

[18] UCL (online). Education and Work.

[19] Howell, C.H., et al. (2023). To Educate Students about AI, Make Them Use It. Scientific American.

[20] Kingston University (2023). Kingston University renews calls for urgent action to prepare students for workplace increasingly dominated by AI as it unveils latest Future Skills report.

[21] Brown, R. (2023). The AI generation: How universities can prepare students for the changing world. DEMOS.

[22] Russell Group (2023). Russell Group principles on the use of generative AI tools in education.

[23] Office for National Statistics (2023). Understanding AI uptake and sentiment among people and businesses in the UK: June 2023.

[24] Yang, W. (2022). Artificial Intelligence education for young children: Why, what, and how in curriculum design and implementation. Computers and Education: Artificial Intelligence, Volume 3.

[25] OneEducation (online). Nurturing Tomorrow’s Minds: What primary schools need to know about AI.

[26] Zheng, L. (2021). The effectiveness of artificial intelligence on learning achievement and learning perception: A meta-analysis. Interactive Learning Environments, Volume 3, 9, pages 5650–5664.