Artificial intelligence Archives | Page 6 of 706

MIT develops safer way to detect harmful AI models

MIT researchers have developed a new auditing method to detect whether generative AI models have been adapted to produce child sexual abuse material without generating illegal content during testing.

The technique was developed with Thorn, a child safety nonprofit focused on protecting children from sexual abuse and exploitation online.

Traditional AI safety testing often involves prompting a model and checking its outputs, but that approach cannot be used for child sexual abuse material, which is illegal to generate in the US and many other jurisdictions.

MIT said the problem has become more urgent as open-source generative AI models become easier to download, adapt and redistribute.

The researchers’ method examines internal changes during fine-tuning, rather than testing the model by generating images.

In tests, the auditing procedure identified model variants adapted to generate child sexual abuse material with 100% accuracy.

MIT said hosting platforms could use the method to flag unsafe models, block uploads or remove harmful adaptations before they spread more widely online.

The researchers also plan to test whether the approach can detect harmful capabilities in a larger set of model variants and in base models before adaptation.

Why does it matter?

The research addresses a serious AI safety blind spot: some harmful model capabilities cannot be tested safely or legally by generating outputs. A non-generative auditing method could give hosting platforms, auditors and law enforcement a safer way to detect models adapted for child sexual abuse material before they are distributed. It also points to a broader governance challenge around open-source generative AI: platforms may need scalable tools to assess harmful adaptations without exposing reviewers to illegal or traumatic content.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

Saudi Arabia launches AI tool for national data insights

Saudi Arabia’s Ministry of Economy and Planning has launched the beta version of INSAIGHTS, an agentic AI tool integrated into the Data Saudi Platform. The system is designed to improve how users access, analyse, and interact with national economic and social data.

INSAIGHTS allows users to convert questions into instant insights and analytics by drawing on more than 7,500 indicators available through the platform. The tool aims to support decision-makers, researchers, analysts and the public with faster access to reliable information for data-driven decisions.

The launch forms part of Saudi Arabia’s wider digital transformation agenda under Vision 2030 and the National Transformation Program. The Ministry plans to continue expanding the tool’s capabilities while using emerging technologies to improve transparency, innovation and user experiences across its digital ecosystem.

Why does it matter?

The INSAIGHTS highlights Saudi Arabia’s growing focus on AI as a tool to improve public data accessibility and strengthen evidence-based decision-making. By combining AI capabilities with extensive national datasets, the platform could help organisations and individuals extract insights more efficiently.

The initiative also demonstrates how governments are increasingly adopting agentic AI systems to enhance digital services and economic planning. As the technology develops, platforms like INSAIGHTS may become important models for using AI to improve transparency, research capabilities and public-sector innovation.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our chatbot!

Music industry introduces labels for AI-generated songs

Major music industry organisations have announced a voluntary labelling system to help listeners understand when AI has been used in songs.

Groups including the RIAA, IFPI, the Recording Academy, SAG-AFTRA, IMPALA, WIN, A2IM and the Human Artistry Campaign back the initiative.

The labels will distinguish between ‘AI-generated’ and ‘AI-assisted’ music.

An ‘AI-generated’ label will apply when most or nearly all of the creative elements in a recording are artificial, such as an AI-generated lead vocal, instrumental performance or a song created from a prompt.

An ‘AI-assisted’ label will apply when humans mainly create a track but use AI for some expressive elements.

Industry leaders said the aim is to give fans clearer information about how music is made, while protecting human creativity, authorship and artistic intent.

The move comes as streaming platforms face rising volumes of AI-made content and low-quality uploads.

Deezer said in April that AI-generated tracks accounted for 44% of all new music uploaded to its platform each day.

Spotify also removed around 75 million spam tracks in 2025, as platforms increase efforts to identify low-quality, fraudulent or unauthorised content.

Why does it matter?

The labelling initiative shows how the music industry is trying to build transparency around AI use without waiting for formal regulation. Clear labels could help listeners distinguish fully synthetic music from human-led work that uses AI as a tool. They may also support copyright, attribution and platform-governance efforts as streaming services face growing volumes of AI-generated tracks, recycled content and spam uploads.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our chatbot!

Philippines uses AI and satellites to strengthen food security

The Philippine Department of Agriculture and the Philippine Statistics Authority are partnering to use AI and satellite technology to improve agricultural data collection, strengthen national food security and support more informed policymaking. Agriculture Secretary Francisco Tiu Laurel Jr. stated that enhanced data sharing between the two agencies would enable policymakers to make more informed decisions on food production, logistics and supply.

The Philippine Statistics Authority has begun piloting AI and satellite imagery to estimate crop production, building on approaches already used in several countries. National Statistician Claire Dennis Mapa said the technology would become more accurate as the Department of Agriculture expands field verification to validate satellite-generated data. The agencies also agreed to broaden the use of digital technologies in agricultural statistics and strengthen the capacity of local government units.

Agriculture Secretary Tiu Laurel also renewed calls to rebuild the department’s network of agricultural extension workers, describing them as its missing ‘boots on the ground’. Expanding the field workforce would support near real-time data collection, improve production forecasts and enable faster responses to challenges affecting farmers and fisherfolk. He also welcomed this year’s national census, saying updated population data would improve food demand forecasting.

The partnership aims to shift the Department of Agriculture from reactive to proactive food security management. Updated agricultural and population data will help the government better estimate future food demand, refine production targets and improve budget planning. According to Tiu Laurel, data should help policymakers anticipate future challenges rather than simply document past events in the Philippines.

Why does it matter?

The initiative illustrates how AI and Earth observation technologies are becoming practical tools for agricultural governance. More timely and accurate data can help governments improve production planning, respond more quickly to climate-related disruptions and strengthen long-term food security.

The partnership also highlights that digital transformation depends on both technology and institutional capacity. By combining AI, satellite imagery and field verification through agricultural extension workers, the Philippines is seeking to build a more reliable and responsive agricultural information system that could serve as a model for other countries facing similar food security challenges.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

Malaysia launches consultations on AI Governance Bill

Malaysia’s Ministry of Digital has launched public engagement sessions on its proposed AI Governance Bill, marking the next step towards establishing the country’s first dedicated legal framework for AI governance.

The sessions, organised by the National AI Office throughout July 2026, aim to explain the proposed framework, encourage dialogue and gather feedback from government agencies, industry, businesses, academia and the public.

According to the Ministry, the AI Governance Bill will become Malaysia’s first horizontal legal framework dedicated specifically to AI governance. It is intended to establish common principles that complement existing legislation and create a more coherent governance framework across sectors.

The proposed legislation adopts a risk-based approach, recognising that AI-related risks can arise throughout a system’s lifecycle. Governance responsibilities would be allocated according to the roles and level of control exercised by different parties.

The Ministry also highlighted incident reporting, appropriate safeguards and regulatory sandboxes as key mechanisms for supporting responsible innovation and strengthening public trust.

Digital Minister Gobind Singh Deo told Parliament on 24 June that the bill would not regulate AI-generated content directly. Existing laws and regulatory frameworks would continue to govern illegal content.

A nationwide consultation through the Unified Public Consultation portal was expected to begin on 10 July 2026, providing another channel for individuals, businesses and organisations to submit feedback before the legislation is finalised.

The Ministry said the process is intended to be transparent, inclusive and consultative, supporting Malaysia’s ambition to become a trusted and globally competitive AI hub under the Towards an AI Nation 2030 roadmap.

Why does it matter?

The proposed AI Governance Bill would provide Malaysia with a single legal framework for AI governance rather than relying on sector-specific rules. By combining a risk-based approach with regulatory sandboxes, incident reporting and public consultation, the government is seeking to balance innovation with legal certainty and public trust.

The initiative also reflects a broader regional trend, with governments across Asia increasingly introducing dedicated AI governance frameworks to support investment while preparing for the safe deployment of increasingly capable AI systems.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

CERN scientists see AI shaping next-generation collider research

CERN researchers are expanding the use of AI across particle physics, deploying AI both during experiments and in the analysis of collision data.

According to CERN physicists, AI systems are already being used to help identify potentially significant particle collisions in real time, enabling researchers to process the enormous volumes of data generated by the Large Hadron Collider (LHC).

Scientists say AI’s role is expected to grow as CERN upgrades the LHC to the High-Luminosity LHC and develops plans for the Future Circular Collider (FCC), a proposed successor that is expected to begin operation in the 2040s. Beyond analysing experimental data, researchers are exploring how AI could support the design of detectors, optimise collider components and improve operational efficiency.

According to CERN physicist Maurizio Pierini, new AI techniques, including neural-network-based trigger systems and anomaly detection, could help identify unexpected events that existing theories do not predict. Researchers also expect AI to improve the precision of data analysis and assist with the design and operation of future particle physics experiments.

Why does it matter?

CERN’s work illustrates how AI is becoming part of the scientific research infrastructure rather than simply a tool for analysing results after experiments. The technology is increasingly being integrated into the design, operation and interpretation of large-scale scientific facilities, with potential implications for how future discoveries are made across research disciplines.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

AI for Good Global Summit 2026 concludes with a focus on implementation

The AI for Good Global Summit concluded after four days of discussions highlighting the practical deployment of AI across healthcare, agriculture, mobility, education, creative industries and public policy.

The programme also explored the technologies and infrastructure supporting AI deployment, including quantum technologies, robotics, international standards and digital infrastructure.

The closing programme featured the Robotics for Good Youth Challenge Grand Finale 2026, an ITU-led UN robotics competition for young people aged 10 to 18. Teams from more than 50 countries competed over four days to develop robotic solutions addressing this year’s theme of food security.

Venezuela won both competition categories, with participants using open-source software and hardware to design, build and programme robots focused on food security.

The final day also featured discussions on AI governance, standards and inclusion. A morning panel examined the underrepresentation of women in AI development and decision-making, bringing together representatives from standards bodies, industry, civil society, telecommunications and ITU networks.

Speakers argued that gender equity should be treated as a core element of AI standardisation rather than a secondary consideration, linking women’s leadership to education, research, technical standards and institutional decision-making.

Youth engagement remained another key theme. In the Youth Zone, participants took part in an AI safety challenge in which they role-played as developers designing safeguards for new AI applications, exploring security principles, ethical risks and responsible AI governance.

The AI for Good Global Summit also explored longer-term questions about AI, culture and education. Futurist Ray Kurzweil discussed the future trajectory of AI, while Arizona State University President Michael Crow and musician will.i.am, ITU Goodwill Ambassador for the AI Skills Coalition, examined AI’s growing role in education and future workforce skills.

The Centre Stage programme concluded with a presentation by Galaxy Corporation CEO Yong-ho Choi on the convergence of AI, robotics and entertainment, including a performance featuring both humans and robots.

The 2026 AI for Good Global Summit concluded with a clear message that the future of AI depends not only on technological progress but also on standards, skills, inclusion and international cooperation to support responsible deployment.

Why does it matter?

The 2026 AI for Good Global Summit reflected a broader shift in AI governance from discussing high-level principles to addressing implementation. Across healthcare, education, robotics, standards and digital infrastructure, the focus was increasingly on how AI can be deployed responsibly in practice rather than whether it should be adopted.

The summit also highlighted the growing importance of international standards, youth engagement and inclusive participation in shaping AI’s future. By bringing together governments, industry, researchers and civil society, the event reinforced the view that responsible AI will depend as much on cooperation and capacity-building as on technological innovation.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

UNESCO study highlights AI gender gap in South Asia

UNESCO has published the Outlook Study on Artificial Intelligence and Gender in South Asia, the first regional assessment of women’s participation in AI across Bangladesh, Bhutan, India, Maldives, Nepal and Sri Lanka.

Developed by the UNESCO Women for Ethical AI (W4EAI) South Asia Chapter, the report examines women’s representation across the AI ecosystem, from education and research to employment and entrepreneurship, highlighting the structural barriers that limit leadership opportunities.

The study finds that greater access to higher education has not translated into equal participation in AI-related fields. Women account for only around one-third of STEM students across the region and remain significantly underrepresented in computing, engineering and AI disciplines.

Although women contribute to AI research, they hold only 26% of corresponding or lead authorship positions, reflecting limited representation in research leadership. In the labour market, women remain concentrated in lower-value AI roles, while technical positions involving AI model development continue to be dominated by men.

Female entrepreneurs also face persistent barriers, including limited access to investment, technical expertise and institutional support.

Drawing on education and labour statistics, bibliometric analysis, LinkedIn Economic Graph data and interviews with women working in AI, UNESCO concludes that AI could either reinforce existing inequalities or become a catalyst for greater gender equality, depending on policy choices.

The report calls for greater investment in inclusive AI education, skills development, leadership opportunities and ethical AI governance to ensure women can participate fully in shaping the region’s AI future.

Why does it matter?

The report shows that gender inequality in AI extends well beyond education, affecting research leadership, employment, entrepreneurship and access to decision-making roles across South Asia. As countries invest in AI-driven economic growth, broader participation will be important not only for fairness but also for innovation and the development of AI systems that reflect diverse perspectives.

The findings also reinforce the growing link between AI governance and inclusion. Building ethical and trustworthy AI depends not only on technical safeguards but also on ensuring that women have equal opportunities to shape how AI technologies are designed, developed and deployed.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

WSIS Forum 2026 closes with call to turn digital commitments into action

The World Summit on the Information Society (WSIS) Forum 2026 concluded with a strong call for the next decade to focus on implementation, as leaders from governments, international organisations, the private sector, and civil society stressed that digital transformation must be measured by its impact on people’s lives rather than by technological progress alone.

Closing the week-long forum in Geneva, ITU Secretary-General Doreen Bogdan-Martin described the gathering as a historic milestone for the global digital community, while Forum Chair Raafat Hendy of Egypt urged stakeholders to ensure that commitments made under the renewed WSIS mandate translate into tangible outcomes by 2035.

Honouring the past while looking ahead

Before reviewing the forum’s achievements, Bogdan-Martin paid tribute to three long-time contributors to the WSIS process who had recently passed away, Rita Goulous of Tunisia, Yuri Grin of the Russian Federation, and Gary Fowley of Canada. She recognised their lasting contributions to building the multistakeholder framework that has guided WSIS for more than two decades.

Turning to the forum itself, Bogdan-Martin highlighted its unprecedented scale, with more than 12,000 participants from over 170 countries attending Geneva Digital Week, including the AI for Good Global Summit, the first Global Dialogue on AI Governance, and the WSIS Forum.

She also celebrated major milestones achieved during the week, including the Partner2Connect initiative surpassing its original target by securing more than USD 120 billion in connectivity commitments worldwide. Another key outcome was the publication of the final report of the International Advisory Body on Submarine Cable Resilience after two years of global collaboration.

Looking ahead, Bogdan-Martin outlined the next phase of the WSIS process, noting that Action Line facilitators will submit implementation roadmaps to the UN Commission on Science and Technology for Development (CSTD) in 2027, directly linking WSIS commitments with the Sustainable Development Goals (SDGs) and the Global Digital Compact.

A decade of implementation

Forum Chair Raafat Hendy reflected on the week’s discussions, identifying digital inclusion, AI for development, digital finance, youth participation, and closing the gender digital divide as the central priorities that emerged throughout the forum.

He argued that connectivity alone is no longer sufficient, stressing that people also need affordable access, digital skills, relevant content, trusted services, and meaningful opportunities to participate in the digital economy.

‘Success should be measured not by technology deployed, but by lives improved,’ Hendy said, urging governments and stakeholders to focus on practical outcomes rather than technological achievements.

The WSIS Prizes were highlighted as examples of that approach, with 18 winners and 72 champions selected from 1,595 submissions representing 122 countries. According to Hendy, the projects demonstrated that digital technologies are already improving education, healthcare, inclusion, and community development worldwide.

Global cooperation remains essential

Representatives from Malaysia, South Africa, and the Republic of Korea reaffirmed their commitment to the WSIS vision of a people-centred, inclusive, and development-oriented information society.

Malaysia pledged continued cooperation to advance universal connectivity, trusted digital governance, and resilient digital infrastructure, while South Africa emphasised the importance of maintaining Africa’s active role in shaping the global digital agenda. The Republic of Korea highlighted the need to move beyond basic connectivity towards meaningful digital use supported by affordability, digital skills, accessibility, and trust.

Throughout the ceremony, speakers consistently stressed that digital transformation cannot be achieved by governments alone. Instead, they pointed to the multistakeholder model that has defined WSIS since its inception, bringing together governments, UN agencies, the private sector, academia, civil society, and the technical community.

Closing the forum, Bogdan-Martin described the renewed WSIS mandate as an opportunity to move from discussion to delivery. With implementation roadmaps due in 2027 and the mandate extended to 2035, she said the coming years should focus on translating political commitments into measurable improvements for people everywhere.

‘The countdown starts now,’ she concluded.

Track all key moments from the WSIS Forum 2026 on our dedicated WSIS page.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

EU-Australia Digital Dialogue focuses on AI and online safety

The EU and Australia have reaffirmed their digital partnership during the third EU-Australia Digital Dialogue, advancing cooperation on AI, cybersecurity, digital policy and secure infrastructure.

The online meeting was co-chaired by Renate Nikolay, Deputy Director-General for Communications Networks, Content and Technology at the European Commission, and Helen Wilson, Deputy Secretary of the Science and Technology Group at Australia’s Department of Industry, Science and Resources.

Discussions covered critical technologies, secure connectivity and digital infrastructure, cybersecurity, online safety and data policy. The two sides also exchanged views on their respective priorities for AI infrastructure, AI capabilities and AI safety.

The dialogue also addressed secure international connectivity and the importance of resilient digital infrastructure. Both sides reviewed progress on online safety cooperation, with particular attention to protecting children online.

The EU and Australia agreed to continue discussions across these areas and explore further opportunities for collaboration, including through Australia’s association with the EU’s Horizon Europe research programme.

Why does it matter?

The dialogue reflects the growing strategic importance of digital partnerships between like-minded countries. As AI, cybersecurity, digital infrastructure and data governance become central to economic competitiveness and national security, international cooperation is increasingly focused on aligning policies as well as developing joint research and technology initiatives.

The reference to Horizon Europe also highlights the practical dimension of the partnership. Beyond policy discussions, cooperation could expand into collaborative research, innovation and technology development, strengthening ties between the EU and Australia’s digital ecosystems.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!