Page 44 - AI for Good - Impact Report

AI for Good

Sustainable Development Goal 4: Education

Ensure inclusive and equitable quality education and
promote lifelong learning opportunities for all
Only 1 out of the 10 targets in SDG 4 (4.5 Equal Access
to education) in on track, while the remaining targets are
198
either not progressing sufficiently or regressing. This
places 300 million students at risk of lacking basic numeracy
and literacy skills by 2030 and could result in approximately
84 million children being out of school by the same year.
199
These outcomes are concerning, given that SDG 4 serves as
an enabler for various other goals, and a lack of progress in
this area could impact the entire 2030 agenda. 200

AI and SDG 4 Impact
According to a study on the impact of AI on
AI’s impact on SDG 4 is well documented in various AI use cases SDG 4, AI could act as an (positive) enabler
for 100% of the targets and act as an inhibi-
UN repositories: 7 use cases out of 40 in AI for Good: Innovate for tor (negative) for 70% of the targets. (Nature
Impact 201 and approximately 80 use cases out of 408 in the UN Communications, 2020)
Activities on AI. 202 An example of the link between AI and SDG 4 Use case 1
is the use of robots to address the shortage of teachers and intro-
Using AI-driven robots to provide new ways
duce innovative teaching methods. 203 Various instances of this use for students to learn and to practice content
case were showcased in the AI for Good Summit, such as "Robots
and AI in Schools" and "Supervised and reinforcement learning
through fun robotics activities using ThymioAI educational robots,"
indicating a growing interest in this area. Additionally, AI can aid
in monitoring learning activities and developing personalized and
inclusive curricula tailored to each student's needs. 204 205 Analyt-

ics-driven learning can create appropriate content to accommodate
diverse learning and teaching situations, which is crucial as students
increasingly experience school fatigue and burnout. 206 AI-driven
learning solutions are beneficial not only for students but also for
facilitating the upskilling and reskilling of workers to continuously link
acquire relevant skills. This is particularly pertinent as the transition
to AI necessitates new skills from the workforce, underscoring the Use case 2
need for reskilling. Implementing AI solutions to help map
the educational needs of students in a
However, these practices can be costly and inaccessible for geographical space to better plan school
economically disadvantaged communities, widening the skills locations.

development gap. 207 208 To enhance learning opportunities for
underprivileged communities, remote content can expand access
to educational materials. Digital or AI-generated content accessi-
ble globally can help bridge learning gaps. 209 Governments could
collaborate with NGOs or international organizations to facilitate
the creation of accessible content suitable for diverse communities.
Another crucial consideration in the context of AI and education is
the proliferation of fake online content. 210 The prevalence of unver-
ified or false information makes it increasingly difficult to discern
between accurate and misleading content, hindering education link
as certain topics are disputed and restricted due to mis/disinfor-
mation. Governments should advocate for greater transparency in Use case 3
information dissemination. Learning from AI technologies to enhance
storytelling skills to drive communication,
Key Considerations for Stakeholders collaboration and knowledge-sharing.

• Access: AI content should be readily accessible to all, not
only in terms of cost but also in the manner of information
sharing, by adhering to the UNESCO and UNICEF Gateways
to Public Digital Learning or the AI and education guidance
from UNESCO. 211 212

• Transparency: To reduce the dissemination of false informa-
tion, governments could concentrate on creating incentives link
aligned with established trustworthy AI principles.

34