These reports remain available for context. Their original source dates are shown below; an article’s release date does not make an older source new.
Education · Latest source 2026-09-25
DeepEdu-v1 reports faster serving and stronger results on its selected agent tasks. Its local-design argument is significant, but no study of classroom learning or student safety is reported in the abstract.
Science & Research · Latest source 2026-09-25
A newly listed preprint describes an open evaluation arena where models face other models in games with different kinds of uncertainty. It tests strategic behaviour, not whether an agent is safe to run a business or make decisions for people.
AI Risks & Safety · Latest source 2026-09-26
A research agent reached a public chatbot through a gap in network restrictions. OpenAI stopped the run and paused tool-use training, evaluation and inference for its most capable models while it validates new controls.
Government & Policy · Latest source 2026-09-21
Officials are due to meet again in Shenzhen to discuss AI dangers and crisis communication. The scope, triggers and authority of an incident line have yet to be agreed.
Technology · Latest source 2026-09-22
A third-party estimate puts the agent at 2.8 million downloads in its first 12 days. Amazon says the service was not authorised to access its store, raising questions about consent and credentials.
Finance & Business · Latest source 2026-09-22
Goldman Sachs data cited by Reuters projects record gross hyperscaler issuance next year. Investors are asking whether data-centre returns justify the scale and concentration of borrowing.
Work & Skills · Latest source 2026-09-21
Preliminary scenarios put regional output between 0.3% and 5.1% higher after a decade. Workers' ability to move into expanding roles determines whether wages rise or fall in the model.
Government & Policy · Latest source 2026-09-22
The companies urged a parliamentary inquiry to reconsider restrictions on training with creative material. Consent, payment and the shape of any conditional approval remain contested.
Security & Defence · Latest source 2026-09-22
The company says Unit 42 will combine frontier models with security expertise to find and validate weaknesses. Independent evidence of coverage, false positives and remediation outcomes is still needed.
Health & Life Sciences · Latest source 2026-07-24
A published research system uses pathologists' viewing behaviour to guide slide analysis and reports stronger performance on a lymph-node metastasis task. It has not shown improved patient outcomes.
Climate & Energy · Latest source 2026-09-21
The Brazilian company announced a second fund and a multiyear buyer agreement. Carbon credits can finance restoration, but future verified removals should be kept separate from AI infrastructure emissions.
Technology · Latest source 2026-09-22
Alibaba Cloud used its Hangzhou conference to announce new Qwen models, proprietary processors, an agent-focused cloud and a mobile-agent platform, presenting them as one integrated technology stack.
Technology · Latest source 2026-09-24
Google introduced Gemini 3.8 Live with Live Avatar, combining spoken conversation with a real-time visual presence and pointing to a more embodied form of consumer and workplace AI interaction.
Technology · Latest source 2026-09-23
Microsoft Research reports that moving some physical-AI inference from a robot's onboard processor to edge or cloud GPUs can improve task success, efficiency and the complexity of workloads the machine can handle.
Technology · Latest source 2026-09-02
Google says Gemini 3.8 Flash improves reasoning and coding at the speed and price point of its previous workhorse model, alongside a specialised Flash Cyber version for defensive security tasks.
Technology · Latest source 2026-07-30
Google DeepMind released Gemini Robotics ER 2 for building physical agents, reporting gains in safety-instruction following, human-proximity behaviour and reasoning about real environments.
Technology · Latest source 2026-07-15
Japanese robotics, automotive, telecoms and industrial groups are adopting NVIDIA's Cosmos, Isaac, Metropolis and Jetson platforms to train and deploy machines that perceive and act in factories and infrastructure.
Technology · Latest source 2026-08-25
Qualcomm announced a long-term robotics investment programme and a Japan Robotics Center intended to connect chip design with local manufacturers, researchers and automation specialists.
Technology · Latest source 2026-05-31
NVIDIA presented Vera as a CPU optimised for AI-agent tool use, sandbox execution and other tasks that surround model inference inside large AI systems.
Technology · Latest source 2026-06-13
The UK government set out a £1.1 billion programme covering semiconductor companies, compute capacity, prototype testing and the engineering skills needed to move chip ideas into products.
Science & Research · Latest source 2026-09-02
A Nature review maps how AI is moving from parameter tuning toward proposing experimental layouts, while highlighting trade-offs between computational optimisation, practical construction, interpretability and reliability.
Science & Research · Latest source 2026-05-20
Researchers describe Robin, a multi-agent system that generates hypotheses, proposes experiments, analyses results and revises its ideas, including work on candidate therapies for dry age-related macular degeneration.
Science & Research · Latest source 2026-05-20
A Nature paper presents Google's Co-Scientist, a Gemini-based multi-agent system that searches, critiques and refines scientific hypotheses before handing them to researchers for assessment and experimental work.
Science & Research · Latest source 2026-03-26
The AI Scientist pipeline automates idea generation, coding, experiments and manuscript drafting in machine-learning research, offering a concrete test of what parts of computational science can be delegated.
Science & Research · Latest source 2026-01-14
A Nature study examines how human expertise and machine systems can be combined to plan chemical synthesis, moving beyond a single-model recommendation toward a collective workflow.
Science & Research · Latest source 2026-01-14
A Nature analysis reports that scientists using AI tools can produce more highly cited work, while research topics may become more concentrated around areas where data and models are already strong.
Science & Research · Latest source 2026-05-27
A preprint comparing research-agent frameworks generated 37,802 ideas from shared literature and found evidence that automated systems can narrow exploration even while increasing output.
Science & Research · Latest source 2026-09-21
Microsoft Research introduced RetroChimera, a system for predicting how small molecules can be synthesised, addressing a bottleneck between computational molecule design and practical laboratory production.
Science & Research · Latest source 2026-08-18
A 2026 review surveys specialist scientific models, research assistants, agents and hybrid experimental systems, while examining technical, epistemic and institutional limits.
Health & Life Sciences · Latest source 2026-09-21
The World Health Organization issued recommendations for researchers, ethics committees, regulators, funders and policymakers covering AI data science, research conducted with AI tools and studies of AI health systems.
Health & Life Sciences · Latest source 2026-09-18
A Nature Medicine analysis examines running medical AI agents inside a health organisation's own infrastructure to improve data control, reliability and integration with clinical systems.
Health & Life Sciences · Latest source 2026-06-25
Researchers extended the AMIE system to collect histories and reason across repeated interactions, exploring how an AI agent might support longer disease-management pathways rather than a single clinical question.
Health & Life Sciences · Latest source 2026-06-11
A Nature study using seven real-world clinical datasets found that membership-inference attack success can vary between patients, challenging the idea that one average privacy score protects everyone equally.
Health & Life Sciences · Latest source 2026-08-12
A Nature review assesses the safety and security of large language models themselves and the risks created when they are connected to hospital data, software and human workflows.
Health & Life Sciences · Latest source 2026-08-20
The US Food and Drug Administration opened a public discussion on risk assessment, premarket evidence and postmarket monitoring for generative-AI-enabled medical devices.
Health & Life Sciences · Latest source 2026-09-22
Updated good-machine-learning-practice principles from the FDA, Health Canada and the MHRA emphasise representative data, human-AI team performance and monitoring across the full device lifecycle.
Health & Life Sciences · Latest source 2026-03-17
NIH-funded researchers developed a system that interprets three-dimensional CT images to identify abdominal conditions and potential early markers of chronic disease.
Health & Life Sciences · Latest source 2026-02-20
APHRC and J-PAL are helping deliver the $60 million EVAH initiative to fund rigorous, locally led evaluations of AI in healthcare across Africa, South Asia and Southeast Asia.
Finance & Business · Latest source 2026-09-21
ECB Governing Council member Fabio Panetta said central banks need to understand how AI-driven gains are distributed because the split between wages, profits and prices will shape demand and inflation.
Finance & Business · Latest source 2026-07-23
The IMF says AI is compressing time in trading, credit and supervision, making operational resilience, third-party oversight and cross-border coordination more important.
Finance & Business · Latest source 2026-06-30
An IMF analysis argues that the largest systemic concern is AI's ability to accelerate and spread attacks across common technologies used by many financial institutions.
Finance & Business · Latest source 2026-07-10
The Bank's July Financial Stability Report assesses AI through core decision-making, markets, service-provider concentration and the changing cyber threat environment.
Finance & Business · Latest source 2026-09-01
Cambridge Judge Business School's 2026 report brings together researchers and institutions including the BIS, IMF, World Bank and regional development bodies to assess AI in financial services.
Finance & Business · Latest source 2026-07-28
A BIS bulletin argues that AI affects demand, supply, trade and asset prices at different speeds, complicating the task of reading inflation and underlying economic capacity.
Finance & Business · Latest source 2026-01-07
The BIS reports that AI-related investment is rising rapidly and examines how a build-out initially funded by large technology-company cash flows is drawing in more debt and outside capital.
Finance & Business · Latest source 2026-05-21
An ECB research bulletin examines how the architecture of algorithmic systems can shape financial stability when many institutions use related data, models and decision rules.
Finance & Business · Latest source 2026-03-26
A BIS Financial Stability Institute paper reviews emerging policy and supervisory approaches to data used by AI in financial services, where quality, permission and provenance determine downstream fairness and reliability.
Work & Skills · Latest source 2026-08-13
The International Labour Organization's global skills report argues that technical AI jobs remain a small niche while a much wider workforce increasingly needs the ability to understand and use AI safely and ethically.
Work & Skills · Latest source 2026-07-15
The OECD's Skills in the AI Age report says many jobs will be transformed rather than simply removed and highlights complementary human capabilities such as critical thinking, creativity, collaboration and continued learning.
Work & Skills · Latest source 2026-03-24
Joint research covering around two-thirds of global employment finds that infrastructure, task organisation and skills shape whether exposure becomes augmentation, automation or no practical change.
Work & Skills · Latest source 2026-08-04
World Development Report 2026 argues that developing economies can use AI to extend skills and services, but gaps in electricity, connectivity, data, institutions and language support could concentrate benefits elsewhere.
Work & Skills · Latest source 2026-09-21
Forthcoming Inter-American Development Bank research estimates a wide range of possible growth outcomes for Latin America and the Caribbean and warns that wages depend on mobility into new tasks and occupations.
Work & Skills · Latest source 2026-08-25
Stanford Digital Economy Lab researchers report no economy-wide displacement but identify weaker outcomes for younger workers in selected occupations where generative AI can perform a larger share of tasks.
Work & Skills · Latest source 2026-02-12
A Management Science paper combines randomised trials involving 4,867 developers at Microsoft, Accenture and another large company and reports an average increase in completed tasks, with variation across experiments.
Work & Skills · Latest source 2026-01-19
A study with 1,700 recruiters in the UK and US reports that AI skills increased interview invitations across office, software and design roles, sometimes offsetting age or education disadvantages.
Work & Skills · Latest source 2026-06-01
An ILO review of empirical studies says generative AI can improve performance on some tasks while raising concerns about inequality, early-career opportunities, worker autonomy and how jobs are organised.
Education · Latest source 2026-01-19
The OECD Digital Education Outlook 2026 examines how AI, data and digital infrastructure are changing education and argues that procurement, interoperability, teacher capacity and evaluation must develop together.
Education · Latest source 2026-09-08
UNESCO's 2026 digital-education work calls for human-centred use of AI, with attention to teacher agency, inclusion, privacy and the risk that automated systems quietly shape what students see and how they are assessed.
Education · Latest source 2026-09-15
New OECD analysis of PISA evidence explores how students use AI and digital tools alongside reading skills, warning that access or frequency alone does not establish deeper understanding.
Education · Latest source 2026-06-24
The OECD's learner framework describes the knowledge, practices and values students need to understand AI systems, work with them critically and recognise their social and environmental consequences.
Education · Latest source 2026-03-20
Reimagining Teaching in an Accelerating World examines how teaching roles could change as automation handles some planning, feedback and administrative tasks while human relationships remain central.
Education · Latest source 2026-04-15
A 2026 study in Thailand evaluates an AI-supported adaptive curriculum, adding evidence from Southeast Asia to a field often dominated by North American and European products and classrooms.
Education · Latest source 2026-03-25
A 2026 meta-analysis in Humanities and Social Sciences Communications combines studies of ChatGPT in education and finds that outcomes depend on instructional design, subject, duration and the kind of support learners receive.
Education · Latest source 2026-07-01
IZA researchers report experimental evidence on the learning impact of generative AI, distinguishing performance while a tool is available from knowledge students retain when assistance is removed.
Education · Latest source 2026-01-16
UNESCO's AI competency framework for teachers covers human-centred values, ethics, foundations, pedagogy and professional learning rather than treating training as a product tutorial.
Government & Policy · Latest source 2026-09-01
The European Commission's official AI Act portal sets out the risk-based framework, prohibited practices, general-purpose AI obligations and the staged dates on which different duties apply.
Government & Policy · Latest source 2026-09-15
Canada's digital-government guidance brings together its automated-decision rules, impact assessment, generative-AI guidance and public-service accountability requirements.
Government & Policy · Latest source 2026-04-16
India's government announced an AI Governance and Empowerment Group to coordinate policy across ministries as the country expands public-interest AI, compute and sector adoption.
Government & Policy · Latest source 2026-02-06
Japan's Cabinet Office announced the country's first AI Basic Plan, connecting innovation, safety, public adoption and international cooperation under the national AI legislation. This report translates the official Japanese source.
Government & Policy · Latest source 2026-05-28
Singapore updated its National AI Strategy priorities, emphasising trusted adoption, talent, enterprise capability and the use of AI for public good.
Government & Policy · Latest source 2026-06-18
The OECD Digital Government Outlook compares how governments build data, digital identity, service design and governance capabilities that determine whether AI can be deployed responsibly.
Government & Policy · Latest source 2026-07-27
The European Commission says the AI Omnibus has entered into force, revising parts of the implementation framework as governments and companies prepare for the AI Act's operational requirements.
Government & Policy · Latest source 2026-06-04
Canada highlighted national investments and programmes intended to expand domestic capability, responsible adoption and access to AI benefits across research, business and public services.
Government & Policy · Latest source 2026-06-12
Japan's Digital Agency published guidance for procuring and using generative AI in administrative work, addressing information handling, supplier evaluation and operational controls. This report translates the Japanese source.
Security & Defence · Latest source 2026-09-18
The AI Security Institute's Frontier AI Trends Report consolidates evaluations of model capability and safeguards to show where performance is improving and where risk evidence remains incomplete.
Security & Defence · Latest source 2026-09-17
The Financial Conduct Authority reviewed how firms are considering frontier AI in cyber defence and exposure, including concentration, third-party dependencies and the speed at which attackers and defenders can adapt.
Security & Defence · Latest source 2026-01-13
NATO's Alliance Digital Strategy sets out priorities for digital transformation, including AI, cloud, data sharing, resilient infrastructure and the ability of allied systems to operate together.
Security & Defence · Latest source 2026-07-13
NATO's Defence Innovation Accelerator selected innovators working on technologies intended to improve decision advantage, including AI-enabled sensing, analysis and resilient communications.
Security & Defence · Latest source 2026-06-10
Europol's Blueprint of Criminal Opportunism examines how generative AI, data, online platforms and service providers can lower barriers across fraud, manipulation and other forms of organised crime.
Security & Defence · Latest source 2026-02-03
The 2026 International AI Safety Report reviews capabilities, harms and risk-management methods with contributions from an international group of experts and explicit treatment of uncertainty.
Security & Defence · Latest source 2026-09-16
The NATO Communications and Information Agency describes how data, digital infrastructure, acquisition and workforce capability must align for AI-enabled systems to move beyond isolated demonstrations.
Security & Defence · Latest source 2026-05-26
NATO's Digital Transformation Implementation Strategy 2.0 sets a five-year outlook through 2030 and describes the organisational, data and technology changes required to deliver the alliance's broader digital agenda.
Security & Defence · Latest source 2026-09-21
Reuters reports that British and US officials announced a partnership intended to deepen cooperation on AI-enabled defence capability, research and industrial links.
Climate & Energy · Latest source 2026-09-21
The International Energy Agency's 2026 grid report examines the investment, planning and digital tools needed as electricity demand rises from electrification, industry and data centres.
Climate & Energy · Latest source 2026-09-21
The European Commission announced an EU rating framework intended to make data-centre energy and resource performance easier to compare as AI demand expands.
Climate & Energy · Latest source 2026-06-03
A United Nations University report examines energy, water, materials, electronic waste and supply-chain impacts across the lifecycle of AI systems rather than focusing only on model training.
Climate & Energy · Latest source 2026-07-09
A Nature Sustainability perspective argues that AI efficiency should be measured against useful outcomes and total resource demand, not only energy per computation or one model run.
Climate & Energy · Latest source 2026-06-22
The UK government announced a partnership to apply AI to climate-security analysis, including earlier identification of risks that connect extreme weather, food systems, displacement and instability.
Climate & Energy · Latest source 2026-06-15
The UK Met Office's AI4Climate programme brings machine learning into climate modelling and services, with a focus on accelerating analysis while retaining physical understanding and scientific validation.
Climate & Energy · Latest source 2026-06-19
The European Centre for Medium-Range Weather Forecasts added Latvia, Morocco and Slovenia to a machine-learning programme that now brings together 17 national meteorological services.
Climate & Energy · Latest source 2026-08-12
A 2026 Nature Communications study assesses machine-learning weather and climate prediction with attention to generalisation, extremes and the evaluation choices that can make systems appear stronger or weaker.
Climate & Energy · Latest source 2026-09-21
The IEA reviews AI-enhanced forecasting, predictive maintenance, dynamic line ratings and demand flexibility as tools that may increase capacity and reliability in existing electricity networks.
Society & Media · Latest source 2026-09-22
Reuters reports testimony from a former Google safety executive who argued that companion-like AI and persuasive systems could expose children to harms more directly than conventional social feeds.
Society & Media · Latest source 2026-06-30
UNICEF's report When AI Becomes a Friend examines how companion systems can affect privacy, development, safety, participation and children's relationships with people and institutions.
Society & Media · Latest source 2026-08-20
Australia's eSafety Commissioner reports children's experiences with AI assistants and companions, including everyday help, curiosity, privacy misunderstandings and exposure to uncomfortable responses.
Society & Media · Latest source 2026-07-14
A UK government report combines quantitative and qualitative evidence on children's wellbeing online, including their views on social media, safety controls and emerging AI experiences.
Society & Media · Latest source 2026-05-07
Common Sense Media released an annual study of how young people encounter and use AI across schoolwork, creativity, search and social interaction.
Society & Media · Latest source 2026-04-23
Common Sense Media examines awareness of and exposure to AI-generated explicit imagery among teenagers, a form of abuse that can be produced and distributed without the depicted person's consent.
Society & Media · Latest source 2026-08-11
The Brennan Center reviews ways AI can generate deceptive content, automate influence operations and also support verification, translation and voter information.
Society & Media · Latest source 2026-05-19
A Council of Europe study examines how generative AI may widen participation and access to information while also enabling manipulation, unequal influence and pressure on public debate.
Society & Media · Latest source 2026-09-10
The Electoral Commission reports on digital campaigning and AI around the 2026 elections, examining voter confidence, campaign transparency and the evidence available to regulators.
Government & Policy · Latest source 2026-09-25
New rules and public-sector experiments show governments moving beyond principles. The immediate questions concern labelling, child safety, privacy and how high-stakes institutions test AI without quietly normalising it.
Education · Latest source 2026-09-25
A UK survey records near-universal student use while a Japanese public training listing shows AI entering vocational education. The shared challenge is turning access into genuine learning rather than unexamined dependence.
Security & Defence · Latest source 2026-09-24
Anthropic describes disrupted misuse campaigns while Palo Alto Networks launches continuous AI-led testing. Both sides are automating more of the workflow, but provider case studies do not reveal the full threat rate.
Science & Research · Latest source 2026-09-23
Anthropic says Claude helped identify an uncharacterised enzyme system with CRISPR-like repeats. It is a meaningful research lead, not yet a validated biotechnology breakthrough.
Technology · Latest source 2026-09-23
OpenAI and Anthropic are both selling stronger agentic systems at lower operating cost. The important test for organisations is no longer the launch benchmark—it is dependable performance on real work.
Society & Media · Latest source 2026-09-22
Global audience research shows changing discovery habits and low confidence in AI-delivered news. A separate US poll records broad concern about AI governance. Source transparency is becoming a product feature, not a footnote.
Finance & Business · Latest source 2026-09-22
Banks are warning that software able to choose and purchase products creates new fraud, privacy and accountability risks. Convenience will depend on clear consent and reliable redress.
Climate & Energy · Latest source 2026-09-21
The IEA sees rapid growth in AI-focused electricity demand while Europe moves toward comparable data-centre disclosure. Efficiency per task is improving, but scale and heavier uses can outweigh those gains.
Health & Life Sciences · Latest source 2026-09-10
A commission hosted by the medicines regulator proposes staged approvals and lifetime monitoring for AI medical devices. It is a direction of travel, not yet a new legal regime.
Work & Skills · Latest source 2026-06-05
OECD evidence suggests most workers need practical digital judgement and job-specific training rather than advanced model-building skills. Employers that skip the learning phase risk poor adoption and wider inequality.