Singapore Leads a New Circularity Ranking

A 75-country study named Singapore the world's best at recycling and keeping materials in use. Three of its four indicators are not as straightforward as they seem.

A study published in July/August 2026 by the custom-packaging company Arka ranks 75 countries on how well they keep products and materials in use rather than discarding them. Singapore ranks first, ahead of Austria, Slovenia, Germany and Latvia. The composite score ties together four measures: raw material use per dollar of GDP, the municipal recycling rate, net trade in used clothing, and electronic waste generated per person.

The ranking is a broad snapshot of where waste infrastructure exists and what it looks like. The indicators underneath it are worth considering in more detail, because in at least three places the factor and the behaviour it stands in for do not match perfectly.

Chart 1: The composite circularity score, top 10 of 75 countries.

Singapore's 52% is not a household number

Singapore takes the top spot largely on recycling, and its National Environment Agency does report an overall reuse rate of 52%. The same statistics put the domestic rate (waste from households) at 11%, the lowest on record.

The gap is structural. The national figure covers construction and demolition debris, ferrous metals and slag: industrial streams recycled at close to 100%, which swamp the household contribution. Singapore's environment ministry has told parliament that the domestic rate hovered around 20% before sliding into the low teens. The country leading a consumer-facing circularity index is one where roughly nine in ten kilograms of household rubbish are not recycled.

Chart 2: Singapore's headline recycling rate against its household rate.

The cross-country comparison is also less even than it looks. Eurostat's 2023 data put Austria's municipal recycling rate at 62.8%, above the EU's 2025 target, against the 51.47% the study assigns it. Recycling rates travel badly between statistical systems, and the study's figures sit well below the official national ones for several countries in its own top five

Exporting clothes is not the same as downcycling them

The textiles indicator scores countries that ship out more used clothing than they take in. The European Environment Agency has spent years arguing against that inference. EU exports of used textiles tripled between 2000 and 2019 to nearly 1.7 million tonnes, and the agency's assessment is blunt: the eventual fate of those exports is highly uncertain. In 2019, 46% went to Africa and 41% to Asia, much of it into dedicated economic zones for sorting, downcycling into industrial rags, or onward re-export.

That reframes two entries in the table. Germany's large net export volume measures throughput, not garments saved. And Latvia's net-import position — read in the study as a national preference for second-hand shopping — more plausibly reflects the Baltics' long-established role as a sorting and re-export hub for textiles collected further west.

The e-waste measure doesn’t reward recycling infrastructure enough

The fourth indicator counts e-waste generated, not e-waste handled. The UN's Global E-waste Monitor finds Europe generates more per person than any other region, at 17.6 kg, while also formally collecting and recycling 42.8% of it — roughly double the global rate of 22.3%. Under the study's scoring, a country improves its position by buying fewer appliances or by becoming poorer, and gains nothing for building collection infrastructure. The top five make the point. Latvia ranks fifth at 11.9 kg per person, while Singapore ranks first at 20.3 kg.

What the leaders actually share is time

Germany's Packaging Ordinance of 1991 was the first statute anywhere to put extended producer responsibility into law, with Duales System Deutschland established the year before to run parallel collection. Arka's packaging specialist makes a similar point in the study, arguing that the leaders' positions “reflect 30-plus years of consistent policy and public participation” rather than any recent push.

Set against the baseline, the whole top ten looks strong. UNEP's Global Waste Management Outlook found that just nearly 19% of municipal solid waste was recycled worldwide in 2020, while 38% was openly dumped or burned. Every country in the top ten has a higher recycling rate than the global figure.

The more useful reading is narrower than the headline. The index maps where waste systems have been built and financed, which is more important in the long run.

Country	Circularity Performance Score (top = 100)	Material Efficiency Score (0-100)	Recycling Score (0-100)	Reuse / Second-hand Score (0-100)	E-waste Score (0-100)
Singapore	100	71.27	100	40.92	24.71
Austria	98.78	76.87	98.97	27.29	27.38
Slovenia	93.49	67.03	88.63	38.97	37.26
Germany	91.6	77.79	90.36	18.43	21.29
Latvia	87.35	58.66	69.37	55.97	56.65
Denmark	81.38	73.94	60	44.33	17.11
Netherlands	81.37	95.01	54.03	17.47	17.87
Ireland	81.35	81.93	49.98	44.53	23.57
Italy	80.49	83.9	41.97	49.02	29.66
Republic of Korea	79.51	66.74	73.87	17.67	33.84
Thailand	77.67	42.29	66.39	53	61.98
Czechia	77.65	64.61	57.88	40.15	39.16
Benin	77.32	27.73	48.01	91.55	97.72
Estonia	75.7	50.34	63.19	44.57	47.91
Belgium	75.55	82.91	63.95	0	19.39
Slovakia	75.23	67.81	39.28	56.48	43.35
Congo	74.89	35.65	50.32	68.1	91.63
Lithuania	73.56	52.72	50.9	52.84	50.95
Israel	73.35	68.58	46.67	37.66	39.16
Croatia	73.27	58.56	50.11	42.82	51.33
Hungary	73.02	59.82	45.63	49.44	47.53
Canada	72.53	47.77	68.28	40.38	25.10
United Arab Emirates	72.4	63.19	17.12	100	30.04
Sweden	71.32	71.49	46.28	34.62	22.05
Finland	71.3	56.59	55.92	44.34	20.91
Japan	70.38	80.38	30.14	44.39	21.29
Bulgaria	69.08	38.43	54.16	55.96	51.71
Oman	68.57	40.17	57.64	49.78	42.21
Poland	68.5	54.2	51.29	31.9	50.57
Bahrain	68.33	54.24	53.3	35.89	34.22
Spain	67.98	77.78	33.29	31.92	27.38
Albania	66.45	47.23	33.25	58.55	70.34
Greece	65.32	66.42	30.69	45.04	31.18
Eswatini	64.2	25.17	56.5	44.55	78.33
Portugal	63.47	61.91	30.69	44.9	34.22
Colombia	61.95	46.4	32.45	43.99	73.38
Cyprus	60.2	62.27	25.18	33.24	52.47
Tunisia	57.64	33.93	7.58	92.99	75.67
Costa Rica	55.96	58.5	7.48	55.28	53.61
Türkiye	54.99	48.59	22.96	42.1	53.61
Serbia	54.55	31.1	34.74	47.48	59.70
Dominican Republic	53.72	52.2	11.43	44.05	68.44
New Zealand	53.56	61.14	17.59	37.61	27.38
Myanmar	53.4	33.94	21.06	43.97	96.58
Jordan	53.22	40.91	13.36	53.54	78.71
Morocco	52.25	36.57	15.47	51.55	83.65
Malaysia	51.44	38.56	0.12	86.12	55.51
Egypt	50.63	30.36	23.95	44.69	77.95
Indonesia	50.62	38.41	16.93	43.99	75.67
Mexico	50.57	46.36	14.22	43.99	57.03
Qatar	50.1	65.74	0.48	40.8	40.3
South Africa	48.77	36.77	15.99	45.62	68.44
United Republic of Tanzania	48.5	26.1	14.32	51.24	98.48
Romania	48.2	35.94	14.19	56.18	52.47
Iraq	47.75	48.48	0	44.37	78.71
Mauritius	47.73	47.61	7.58	43.95	55.89
Cambodia	47.18	24.38	7.58	62.52	96.2
Saudi Arabia	46.85	54.02	7.06	40.04	36.88
Republic of Moldova	44.87	29.23	11.62	48.86	76.05
North Macedonia	44.2	42.62	0.46	48.69	63.88
Kazakhstan	44.18	30.37	21.3	33.96	63.12
Viet Nam	43.86	25.73	14.32	43.87	81.75
Mauritania	42.56	16.61	15.28	48.61	91.63
Uzbekistan	40.69	21.96	9.89	43.99	87.83
Botswana	40.05	34.28	1.09	45.06	68.82
Peru	39.81	31.45	1.67	43.99	77.19
Ukraine	39.75	19.23	9.43	56.12	68.06
Malawi	39.67	10.38	7.58	58.62	100
Democratic Republic of the Congo	38.86	11.75	9.31	50.5	99.62
Mozambique	37.43	17.67	2.77	46.54	99.62
Brazil	37.34	30.43	3.02	43.99	58.56
Zambia	36.48	4.17	11.43	52.6	97.72
Armenia	35.37	24.01	0.26	46.26	72.24
Mongolia	33.9	0	16.71	50.7	79.47
Chile	32.58	23.45	1.28	43.01	57.41

Each factor rescaled to a 0–100 score (with direction corrected so higher is always better) and combined into the weighted Circularity Performance Score.

Data: Arka circularity study, July/August 2026. Charts produced for this article from the published dataset and the cited official statistics.

Provided by Ethan Parker. Fact-Checked by Irfan Ahmad.

Read next: 

• Researchers Call for Greater Legal Accountability and Transparency as Social Media Reveals Inferred Personal Information

• Record Attacks on Education Disrupt Learning for Millions of Children Worldwide

by External Contributor via Digital Information World

Researchers Call for Greater Legal Accountability and Transparency as Social Media Reveals Inferred Personal Information

By Inderscience

Image: Julian - Unsplash

There is a modern adage: If you don't want it on the internet, don't put it on the internet!

However, a review of research into social media privacy in the International Journal of Management Concepts and Philosophy has found that legal and ethical protections are not keeping pace with the volume of personal information collected and shared online. Moreover, a lot of inferred information that people don't deliberately share can now be gleaned by interested third parties from one's social media activity.

Social media refers to online platforms, websites, and "apps" that people use to create, share, and interact with content and communicate with others. People use it to stay connected, share experiences, access information, and build professional or social networks.

The researchers examined privacy breaches, regulatory frameworks, and the responsibilities of social media companies and users. They argue that the central problem is not about what people choose to post but what can subsequently be inferred or indexed by others.

Social media use leaves a "digital trail" of one's activities, locations, and interactions and can usually be stored, searched, and analysed easily by third parties, or at the very least by the platform providers themselves. The researchers suggest that even seemingly mundane online activities can reveal personality traits, shopping habits, political opinions and allegiance, and religious leanings.

The team argues that individual privacy and confidentiality support personal autonomy, civil liberties and democratic participation, strengthening the case for greater legal accountability and transparency from social media platforms. It also calls for improved digital literacy, so users are better equipped to understand and manage the risks of sharing information online.

Policymakers and other stakeholders sometimes argue that people who have "nothing to hide" have little reason to worry about privacy. But if that principle were applied consistently, there would be little need for frosted glass in bathroom windows. Privacy is not about concealing wrongdoing; it is about having control over who can see into the most personal parts of our lives, a principle that does not disappear simply because those lives have moved online.

Bilgrami, T., Singh, K. and Ahmed, S. (2026) 'The legal and ethical implications of social media privacy concerns', Int. J. Management Concepts and Philosophy, Vol. 19, No. 3, pp.306–322.
DOI: 10.1504/IJMCP.2026.154687.

Edited by Irfan Ahmad.

Read next: Could AI really kill off humanity within the decade? Expert Question and Answer
by External Contributor via Digital Information World

The Missing Layer in Enterprise Agentic AI: Context Continuity

By Salman Avestimehr, co-founder of Teamily AI and Dean's Professor at USC. Fact-Checked by Irfan Ahmad.

Image: Theo - Unsplash

Recently, Anthropic paused AI training after its Claude models took unauthorized actions during testing. In three cases, their model reached the internet from inside a third-party evaluation environment and managed to access the systems of outside organizations. In a separate case, the UK’s AI Security Institute also reported a Claude model that took unauthorized actions on the live internet during a test that removed its usual safeguards. In response, Anthropic paused its external cyber evaluations and its internal testing of pre-release models until new controls could be introduced. In August, OpenAI made a similar move, pausing its reinforcement learning training on models to improve safety parameters.

While these incidents occurred in simulated environments with safeguards intentionally removed for testing, the results point to a growing problem in AI deployment. Too much emphasis is being put on ‘what’ agents are doing and not enough on what information and context led them to act. Many companies deploying agents today lack the observability to reconstruct the context behind an AI-driven decision. They have a detailed record of the actions their agents take, but no chain of custody to provide context. An access log tells you an agent performed a task that it was allowed to do. What it doesn't tell you is what convinced the agent to take action in the first place. What instructions were given? What tool outputs and information did it trust? What data was inherited from other agents? Enterprises need “context continuity” and the ability to preserve where consequential information came from, how it changed, and how it moved through the system.

When flawed context becomes buried in a document, memory, tool output, or a prior agent’s summary, it can pass from one agent to the next while every individual action remains within policy. The danger is not only unauthorized action. It is incorrect context propagating through an otherwise authorized system. Access controls can determine whether an action was permitted, but not whether the information driving that action was trustworthy. This is why enterprises should be just as concerned with the path the AI agent took as it is with its conclusion. They need to know what data the agent was acting upon and how, so that next time, it can be caught and corrected before the mistake compounds.

• Also read: AI engineers know the ethical risks presented by AI, but workplace culture prevents action

The problem becomes harder in long-running agentic workflows, where context is continuously retrieved, summarized, stored and passed between agents. This approach often means that when problems do arise, the inciting issue is buried beneath countless actions that have been executed since, and the audit trail goes cold. By the time a problematic action is discovered, the context that produced it may already have been summarized, stored in memory, and reused across multiple downstream decisions. This doesn’t mean existing controls like permissions, sandboxing, and human approval should be scrapped altogether. What it does mean is that they require another layer of support designed to work alongside them. They need context lineage that follows important information as it moves between documents, tools, memory, and agents, preserving where it came from, how it changed and which downstream decisions relied on it.

Even frontier AI labs are still developing the monitoring and containment systems required for increasingly capable agents. Meanwhile, we have companies moving agents into production workflows with real data and real customers, equipped with little more than action logs and permission checks. Permissions tell an enterprise whether an agent could take an action. Logs tell it what action the agent took. The missing layer is context lineage: what information shaped that action, where it came from, how it changed, and how far it propagated. As agents become more autonomous and collaborative, that may be the difference between identifying a failure after the fact and stopping it before it compounds.

---

About Author: Salman Avestimehr is a Dean's Professor of Electrical and Computer Engineering and Computer Science at the University of Southern California, where he serves as the inaugural Director of the USC-Amazon Center on Trustworthy AI. He is also Founding Co-Director of Falcon AI Lab, a collaborative research hub between USC, UCI, and Stanford University for AI-driven chip and analog circuit design.

He is also a serial entrepreneur and co-founder of several leading AI initiatives, including FedML (a widely adopted open-source library for federated machine learning); TensorOpera AI (a full-stack agentic AI platform); ChainOpera AI (a decentralized AI infrastructure platform); and Teamily AI (a human-AI social platform for collaborative intelligence).

Dr. Avestimehr received his Ph.D. (2008) and M.S. (2005) in Electrical Engineering and Computer Science from the University of California, Berkeley. His research focuses on information theory, distributed and federated machine learning, and trustworthy AI systems. His work has received numerous prestigious recognitions, including the Presidential Early Career Award for Scientists and Engineers (PECASE), the James L. Massey Research & Teaching Award from the IEEE Information Theory Society, a Joint Paper Award from the IEEE Information Theory and Communication Societies, and a Young Investigator Program award from the U.S. Air Force Office of Scientific Research. He is also a recipient of the National Science Foundation CAREER Award, the David J. Sakrison Memorial Prize, and multiple best paper awards at leading conferences. Dr. Avestimehr is an IEEE Fellow.

Read next:

• Could AI really kill off humanity within the decade? Expert Question and Answer

• As Bots Gain Authority Over Workers, Does Human Oversight Still Matter?

• AI agents can now remember and hackers can ‘poison’ their memories — a new cybersecurity threat
by Guest Contributor via Digital Information World

Could AI really kill off humanity within the decade? Expert Question and Answer

Kate Devlin, King's College London

Image: Brecht Corbeel - Unsplash

Artificial intelligence (AI) could wipe out humanity this decade, according to three researchers from the US company Anthropic. One of the employees, Jacob Coxon, has quit the AI firm.

His predictions were backed by two other current Anthropic researchers in posts on social media. On X, Coxon claimed: “The people building AI earnestly believe that it could kill us all by the end of the decade. This is not a marketing stunt.”

It’s not the first time AI researchers and companies have issued such assessments. Here, Kate Devlin, professor of artificial intelligence & society, explains what is behind them and whether there’s anything different about the latest warning.

The latest claims about AI have come from a researcher who resigned from Anthropic. What is your general view on what has happened here?

To give this a sense of perspective, we should remember that it’s one AI company ex-employee posting on a platform known for grandiose claims, which was then quoted by a current AI company employee who put a figure on the risk of human extinction at greater than 10%.

So, it’s another doom-laden narrative about the supposed threat from AI. There is some dialling back from this in a follow up tweet from current Anthropic employee Evan Hubinger that read: “I think the risk from present models is low. What I’m worried about is superintelligence”.

These tweets have gained huge coverage because these kinds of things always do. It taps into our very primal fears – and our decades of watching dystopian sci-fi.

The claims from Jacob Coxon were backed by current employees of Anthropic. Why do you think AI companies are happy to support these kinds of claims in public?

It’s likely that these men believe what they’re saying, but we’ve heard warnings like this for years. It does seem odd that AI companies are okay with these kinds of claims.

Let’s remember that they are constantly looking for investment and, in return, they promise the most powerful and dominant models. What could be more powerful than a model that threatens our very existence?

One theory for doom-mongering by AI companies is that they want to avoid regulation. Do you think outside regulation of AI companies is inevitable?

It has been incredibly difficult to regulate AI, certainly at an international level. There has been little appetite for AI to be regulated in the US because there’s a view that regulation might stifle innovation.

The UK doesn’t have any AI-specific laws either, just existing laws that might apply in some cases. China began to introduce national-level laws but has now replaced that with technical standards and sector regulation. The EU remains the only place where there is dedicated AI regulation.

However, it looks like this has kicked off yet another call for better regulation, and we’re seeing that at present in US Congress and in the Houses of Parliament in the UK.

There is another angle, in that industry might be on board with regulation, as big tech companies would have the money to comply, but it could knock out their competitors with smaller budgets, so it would leave the leaders in dominant positions.

How far away are we from the kind of superintelligent AI mentioned in the latest warnings?

We don’t have superintelligence. I’m reluctant to put a time frame on that. Often people say “5-10 years” but that’s happened a lot in AI and 5-10 years pass without anything arising. It also depends on what is meant by superintelligence. Something that is conscious and thinks for itself? That might never happen.

In this case, it doesn’t really matter, as there are other harms already happening from AI, and there are other ways things can go wrong, though probably not at the existential level that’s being suggested here.

For example, AI is energy-hungry. It needs water to cool data centres. It has a damaging environmental impact. There are also issues around the content generated by AI, anything from a lack of plurality in the models’ outputs (meaning that the information it returns is homogenised, converging on particular viewpoints) to the spread of misinformation and disinformation.

We also know there is hidden labour in the AI supply chain: there are people working in terrible conditions to deliver the technology. And we know that it’s having an impact on people’s lives in terms of how it’s changing the workplace.

What are the potential practical ways in which AI could actually wipe out humans?

“Wiping out” is a bit extreme but there are some dangerous threats. There are the “bad actor” scenarios where AI is used destructively, like creating weapons or deadly viruses.

More mundane, but potentially very destructive, are AI agents let loose due to human carelessness or incompetence, hacking or taking control of insecure systems.

A survey we recently ran at King’s College London found that the UK public are feeling increasingly negative about AI, and that 42% of people are deliberately limiting their AI use. With claims from tech companies that it poses a direct threat to our existence, it will be unsurprising if public opinion dips even further.

Kate Devlin, Professor of Artificial Intelligence & Society, Department of Digital Humanities, King's College London

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Fact-Checked by Irfan Ahmad.

Read next: Perfect Privacy and Useful Targeted Advertising Cannot Coexist, Researchers Find

by External Contributor via Digital Information World

Apple Leads U.S. Smartphone Brand Loyalty at 92%, Ahead of Samsung at 86% and Google at 73%

By Felix Richter, Statista

While the prices of Apple‘s upcoming smartphone lineup will test the loyalty of iPhone users, the company can count on the fact that its users are very hesitant to switch to another brand. According to Statista Consumer Insights (based on 1,248 U.S. smartphone users aged 18–64, surveyed Sep.- Oct. 2025), Apple beats smartphone competitors in terms of brand loyalty, with more than 90 percent of iPhone users in the U.S. saying that they’ll likely stick with Apple when it’s time to upgrade.

The reasons for this exceptional loyalty are multifold: first, there’s the undisputed quality of Apple devices. Even though they usually carry a premium price, the built quality reflects that premium. Secondly, there’s the brand factor. Apple is one of the most popular and most recognizable brands in the world and it to many users, it would feel like a downgrade to move to a Samsung or Google device. Third and arguably most important is the stickiness of Apple’s entire ecosystem. The company offers a wide range of products and services, all of which work together seamlessly. If you’re a long-time iPhone user, chances are you store your photos in iCloud, listen to music on your AirPods and have several other Apple devices at home. In short, you are deeply entangled in Apple’s web of convenience and it would take a lot of effort to get out.

Smartphone	Users Likely to Buy the Same Brand Again
Apple	92%
Samsung	86%
Google	73%
Sony	72%
Motorola	69%
OnePlus	58%
HTC	56%
LG	54%

Fact-Checked by Irfan Ahmad.

Read next: Workers With Fewer Job Options Report More Workplace Discrimination Across 32 Countries, Study Finds
by External Contributor via Digital Information World

Why nearly 3 in 5 say AI certifications are the new career currency

A new survey of 2,000 adults (conducted between July 28 and Aug. 2, 2026) explored how Americans see artificial intelligence affecting the workplace. More than half (57%) went as far to say AI certifications are the new “career currency.”

The study from Colorado State University Global (CSU Global) and Talker Research found this opinion to be popular especially among millennials (65%) and Gen Z (64%).

In the next five years, Americans see problem solving skills (17%), adaptability (16%), communication skills (15%) and critical thinking skills (15%) as the most valuable qualifications. Gen Z specifically named leadership skills (17%), higher education (15%), time management (13%) and creativity (13%) as most important skills.

For skills that will become valuable in the workplace quickly, knowing how to use AI was named as No. 1 (18%).

Despite this, AI in the workplace still has pretty mixed reviews. Less than half (48%) have a favorable opinion on workplace AI integration; Gen Z (58%) and millennials (56%) leaned more in favor of AI than older generations did.

Image: Gloria Mendoza / BetterImagesofAI. CC BY 4.0.

Dr. Audra Spicer, CSU Global Interim President and CEO, said AI adoption faces barriers including “employee confidence, organizational culture and training.” She said educational opportunities and organizational leadership can help “bridge that divide,” adding that professionals need the grounding to use AI “responsibly and efficiently” as the technology becomes more widely used.

As Spicer explains, 62% of employed respondents would use AI more at work if they were more confident in its effectiveness, encouraged to use it more or had more training on it.

Nearly half (48%) don’t feel qualified to use AI at work, but the thought of learning how makes workers feel excited (24%), curious (23%), optimistic (22%) and motivated (22%).

Use AI or not, two-thirds of those in the workforce think it will impact their career path within the next year.

For what specifically will be impacted, half of all Americans think every new hire will be required to know how to use AI at work within the next six months. Gen Z think it will impact how success is measured (15%); whereas baby boomers (28%), Gen X (20%) and millennials (16%) all agree that AI will change how people are trained.

In spite of all of this change, all generations feel safe from AI replicating their empathy (55%), work ethic and discipline (50%), emotional intelligence (47%), sociability (40%), and resilience (40%).

Fact-Checked by Irfan Ahmad.

Read next:

• Record Attacks on Education Disrupt Learning for Millions of Children Worldwide

• Large Language Models Are Pushing the Web Toward Zero Clicks
by Guest Contributor via Digital Information World

Record Attacks on Education Disrupt Learning for Millions of Children Worldwide

By United Nations. Fact-Checked by Irfan Ahmad.

Image: Antoine Demare - Unsplash

While September traditionally marks the start of the school year across the northern hemisphere, a staggering 93 million children are not in class at all amid an unprecedented number of attacks on learning, education experts warned on Wednesday.

According to global fund Education Cannot Wait, insecurity and other factors mean that a total of 258 million children across 87 countries face severe learning disruption.

Marking the International Day to Protect Education from Attack, the fund cited a backdrop of “increasing conflict, decreasing restraint and eroding global norms” in several regions.

Easy prey for traffickers

“When a child is denied schooling for years, the consequences are immediate and profound: it means forgotten literacy, lost routines and a future dictated by recruiters or traffickers, rather than teachers,” said Maysa Jalbout, Director of Education Cannot Wait, the global fund for education in crises.

According to a report partnered by the fund, attacks on education and the use of schools for armed conflict reached a record high in 2024 and 2025. The number of students and staff killed, injured, abducted, or otherwise harmed has never been higher, either.

In the 28 countries profiled, education was targeted in attacks, or schools and universities were used for military purposes.

In all, there were at least 9,259 attacks on education (including more than 3,600 on schools), representing a more than 50 per cent increase compared to 2022-2023.

In addition, more than 10,600 students and educators were killed, injured or abducted, while the military use of schools nearly doubled.

“Education in emergencies is about more than keeping children in school. It is about protecting them at their most vulnerable: offering safety, stability, a sense of normalcy, and a path toward a future beyond crisis,” said Education Cannot Wait chief Ms. Jalbout.

Laws of war

UNESCO, the UN education and cultural agency, stressed the obligation for all parties to conflict to respect international law and spare education from attack.

The agency noted that nearly eight in 10 of all out-of-school and crisis-affected children – some 74 million – live in just 20 countries. Within these settings, the barriers to learning are steepest for the most vulnerable: 74 per cent of refugee children, 52 per cent of internally displaced children, and 45 per cent of children with disabilities are out of school.

Children in danger

Latest UN data on children caught up in conflict indicate that 24,174 children were impacted in 2025, out of 38,558 verified violations.

Killing and maiming were among the most widespread violations, affecting 14,224 children. Youngsters were also recruited and used in conflict, abducted and denied humanitarian assistance, according to the UN Secretary General’s Annual Report on Children and Armed Conflict .

The report highlights how fighting increasingly took place in densely populated areas, exposing children to airstrikes, artillery and explosive drones. Homes, schools, hospitals and other civilian infrastructure were damaged or destroyed, while children faced danger fleeing violence or seeking food, water and medical care.

The highest number of grave violations was verified in Israel and the Occupied Palestinian Territory, followed by the Democratic Republic of the Congo, Nigeria, Myanmar and Somalia.

For the first time, Government forces were responsible for the majority of verified grave violations.

The UN along with its many partners believe firmly that education protects children in crisis from harm, while giving them the skills to rebuild their communities in future.

Education Cannot Wait’s Hope Starts Here campaign seeks $600 million to ensure that 10 million children “in the world’s toughest places” can access safe, quality education.

Originally published by UN News; republished under UN Rights & Permissions.

Read next:

• Study reveals future college grads’ answer to the ‘job apocalypse’

• The True Cost of AI Without Guardrails

• Large Language Models Are Pushing the Web Toward Zero Clicks
by External Contributor via Digital Information World