Digital standards

OpenAI unveils AgentKit for faster AI agent creation

OpenAI has launched AgentKit, a new suite of developer tools designed to simplify AI-powered agents’ creation, deployment, and optimisation. The platform unifies workflows that previously required multiple systems, offering a faster and more visual way to build intelligent applications.

AgentKit’s AI includes Agent Builder, Connector Registry, ChatKit, and advanced evaluation tools. Developers can now design multi-agent workflows on a visual canvas, manage data connections across workspaces, and integrate chat-based agents directly into apps and websites.

Early users such as Ramp and LY Corporation built working agents in just a few hours, cutting development cycles by up to 70%. Companies including Canva and HubSpot have used ChatKit to embed conversational support agents, transforming customer experience and developer engagement.

New evaluation features and reinforcement fine-tuning allow users to test, grade, and improve agents’ reasoning abilities. AgentKit is now available to developers and enterprises through OpenAI’s API and ChatGPT Enterprise, with a wider rollout expected later this year.

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

Bulgaria eyes AI gigafactory partnership with IBM

Bulgaria is considering building an AI gigafactory in partnership with IBM and the European Commission, Prime Minister Rosen Zhelyazkov announced after meeting with IBM executives in Sofia. The project aims to attract large-scale high-tech investment and strengthen Europe’s AI infrastructure.

The proposed facility would feature over 100,000 advanced GPU chips and require up to 500 megawatts of power. The initial phase alone is expected to need around 70 megawatts, highlighting the scale of the planned operation.

Funding could come through a public-private partnership, with the European Commission covering up to 17 percent of capital costs and EU member states contributing additional support for this Bulgarian project.

IBM is considered a strategic technology partner, bringing expertise in cloud computing, cybersecurity, and AI systems. The first gigafactories across Europe are expected to begin operations between 2027 and 2028, aligning with the EU’s plan to mobilise €200 billion for AI development.

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

AI maps over 1,300 mouse brain subregions with unprecedented precision

Researchers at UCSF and the Allen Institute have created one of the most detailed mouse brain maps. Their AI model, CellTransformer, identified over 1,300 brain regions and subregions, including previously uncharted areas. The findings were published in Nature Communications.

CellTransformer utilises spatial transcriptomics to define brain regions based on shared cellular patterns, rather than relying on expert annotation. Drawing city borders from building types reveals finer brain structures. This data-driven method provides unprecedented precision.

The model replicated known regions, such as the hippocampus, and revealed previously unknown subdivisions in the midbrain reticular nucleus. Researchers compared the leap from mapping continents to mapping states and cities. The tool provides a foundation for more targeted neuroscience studies.

Validation against the Allen Institute’s Common Coordinate Framework strongly aligned with expert-defined anatomy. The results gave researchers confidence in the biological relevance of the new subregions. Further studies will investigate their functions.

The model’s potential goes beyond neuroscience. Its methods can map other tissues, including cancers, by analysing large spatial transcriptomics datasets. However, this could support new medical research, helping uncover disease mechanisms and accelerate treatment development.

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

New report finds IT leaders unprepared for evolving cyber threats

A new global survey by 11:11 Systems highlights growing concerns among IT leaders over cyber incident recovery. More than 800 senior IT professionals across North America, Europe, and the Asia Pacific report a rising strain from evolving threats, staffing gaps, and limited clean-room infrastructure.

Over 80% of respondents experienced at least one major cyberattack in the past year, with more than half facing multiple incidents. Nearly half see recovery planning complexity as their top challenge, while over 80% say their organisations are overconfident in their recovery capabilities.

The survey also reveals that 74% believe integrating AI could increase cyberattack vulnerability. Despite this, 96% plan to invest in cyber incident recovery within the next 12 months, underlining its growing importance in budget strategies.

The financial stakes are high. Over 80% of respondents reported spending at least six figures during just one hour of downtime, with the top 5% incurring losses of over one million dollars per hour. Yet 30% of businesses do not test their recovery plans annually, despite these risks.

11:11 Systems’ CTO Justin Giardina said organisations must adopt a proactive, AI-driven approach to recovery. He emphasised the importance of advanced platforms, secure clean rooms, and tailored expertise to enhance cyber resilience and expedite recovery after incidents.

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

Employees embrace AI but face major training and trust gaps

SnapLogic has published new research highlighting how AI adoption reshapes daily work across industries while exposing trust, training, and leadership strategy gaps.

The study finds that 78% of employees already use AI in their roles, with half using autonomous AI agents. Workers interact with AI almost daily and save over three hours per week. However, 94% say they face barriers to practical use, with concerns over data privacy and security topping the list.

Based on a survey of 3,000 US, UK, and German employees, the research finds widespread but uneven AI support. Training is a significant gap, with only 63% receiving company-led education. Many rely on trial and error, and managers are more likely to be trained than non-managers.

Generational and hierarchical differences are also evident. Seventy percent of managers express strong confidence in AI, compared with 43% of non-managers. Half believe they will be managed by AI agents rather than people in the future, and many expect to be handled by AI themselves.

SnapLogic’s CTO, Jeremiah Stone, says the agile enterprise is about easing workloads and sparking creativity, not replacing people. The findings underscore the need for companies to align strategy, training, and trust to realise AI’s potential in the workplace fully.

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

AI tools reshape how Gen Z approaches buying cars

Gen Z drivers are increasingly turning to AI tools to help them decide which car to buy. A new Motor Ombudsman survey of 1,100 UK drivers finds that over one in four Gen Z drivers would rely on AI guidance when purchasing a vehicle, compared with 12% of Gen X drivers and just 6% of Baby Boomers.

Younger drivers view AI as a neutral and judgment-free resource. Nearly two-thirds say it helps them make better decisions, while over half appreciate the ability to ask unlimited questions. Many see AI as a fast and convenient way to access information during car-buying.

Three-quarters of Gen Z respondents believe AI could help them estimate price ranges, while 60% think it would improve their haggling skills. Around four in ten say it would help them assess affordability and running costs, a sentiment less common among Millennials and Gen Xers.

Confidence levels also vary across generations. About 86% of Gen Z and 87% of Millennials say they would feel more assured if they used AI before making a purchase, compared with 39% of Gen Xers and 40% of Boomers, many of whom remain indifferent to its influence.

Almost half of drivers say they would take AI-generated information at face value. Gen Z is the most trusting, while older generations remain cautious. The Motor Ombudsman urges buyers to treat AI as a complement to trusted research and retailer checks.

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

Google unveils CodeMender, an AI agent that repairs code vulnerabilities

Google researchers have unveiled CodeMender, an AI-powered agent designed to automatically detect and fix software vulnerabilities.

The tool aims to improve code security by generating and applying patches that address critical flaws, allowing developers to focus on building reliable software instead of manually locating and repairing weaknesses.

Built on the Gemini Deep Think models, CodeMender operates autonomously, identifying vulnerabilities, reasoning about the underlying code, and validating patches to ensure they are correct and do not introduce regressions.

Over the past six months, it has contributed 72 security fixes to open source projects, including those with millions of lines of code.

The system combines advanced program analysis with multi-agent collaboration to strengthen its decision-making. It employs techniques such as static and dynamic analysis, fuzzing and differential testing to trace the root causes of vulnerabilities.

Each proposed fix undergoes rigorous validation before being reviewed by human developers to guarantee quality and compliance with coding standards.

According to Google, CodeMender’s dual approach (reactively patching new flaws and proactively rewriting code to eliminate entire vulnerability classes) represents a major step forward in AI-driven cybersecurity.

The company says the tool’s success demonstrates how AI can transform the maintenance and protection of modern software systems.

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

Harvard team builds quantum computer that runs continuously for over two hours

A team of Harvard physicists has built a quantum computing machine that can operate continuously without restarting, achieving a significant milestone in experimental quantum hardware.

Until now, quantum computing systems have typically run only for milliseconds or seconds before decoherence or atom loss forces a reset. But in a new setup, the team sustained operation for more than two hours, and they claim that, in theory, it could run indefinitely.

The breakthrough depends on a design that uses an optical lattice conveyor belt together with optical tweezers. These tools allow the system to replenish qubits (atoms) in real time, injecting new atoms at a rate of 300,000 per second into a 3,000-qubit array, to counteract atom loss and maintain quantum information.

Overcoming atom loss has been one of the biggest bottlenecks in scaling quantum computers. Without that fix, durability and error accumulation limit usability. With this experiment, the researchers demonstrate a path toward more robust, always-on quantum platforms.

Mikhail Lukin, who leads Harvard’s quantum research, said that while scaling remains challenging, the approach appears compatible with larger systems. Collaboration with MIT physicist Vladan Vuletić suggested that machines capable of indefinite operation could be within reach in as little as three years.

Applications in cryptography, materials simulation, finance, and medicine could benefit enormously if quantum machines can reliably operate over long durations. The new design resets a key assumption in quantum systems, shifting focus from short bursts of computation to sustained, fault-tolerant operation.

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

Policy hackathon shapes OpenAI proposals ahead of EU AI strategy

OpenAI has published 20 policy proposals to speed up AI adoption across the EU. Released shortly before the European Commission’s Apply AI Strategy, the report outlines practical steps for member states, businesses, and the public sector to bridge the gap between ambition and deployment.

The proposals originate from Hacktivate AI, a Brussels hackathon with 65 participants from EU institutions, governments, industry, and academia. They focus on workforce retraining, SME support, regulatory harmonisation, and public sector collaboration, highlighting OpenAI’s growing policy role in Europe.

Key ideas include Individual AI Learning Accounts to support workers, an AI Champions Network to mobilise SMEs, and a European GovAI Hub to share resources with public institutions. OpenAI’s Martin Signoux said the goal was to bridge the divide between strategy and action.

Europe already represents a major market for OpenAI tools, with widespread use among developers and enterprises, including Sanofi, Parloa, and Pigment. Yet adoption remains uneven, with IT and finance leading, manufacturing catching up, and other sectors lagging behind, exposing a widening digital divide.

The European Commission is expected to unveil its Apply AI Strategy within days. OpenAI’s proposals act as a direct contribution to the policy debate, complementing previous initiatives such as its EU Economic Blueprint and partnerships with governments in Germany and Greece.

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

EU digital laws simplified by CEPS Task Force to boost innovation

The Centre for European Policy Studies (CEPS) Task Force, titled ‘Next Steps for EU Law and Regulation for the Digital World’, aims to refine and simplify the EU’s digital rulebook.

This rulebook now covers key legislation, including the Digital Markets Act (DMA), Digital Services Act (DSA), GDPR, Data Act, AI Act, Data Governance Act (DGA), and Cyber Resilience Act (CRA).

While these laws position Europe as a global leader in digital regulation, they also create complexity, overlaps, and legal uncertainty.

The Task Force focuses on enhancing coherence, efficiency, and consistency across digital acts while maintaining strong protections for consumers and businesses.

The CEPS Task Force emphasises targeted reforms to reduce compliance burdens, especially for SMEs, and strengthen safeguards.

It also promotes procedural improvements, including robust impact assessments, independent ex-post evaluations, and the adoption of RegTech solutions to streamline compliance and make regulation more adaptive.

Between November 2025 and January 2026, the Task Force will hold four workshops addressing: alignment of the DMA with competition law, fine-tuning the DSA, improving data governance, enhancing GDPR trust, and ensuring AI Act coherence.

The findings will be published in a Final Report in March 2026, outlining a simpler, more agile EU digital regulatory framework that fosters innovation, reduces regulatory burdens, and upholds Europe’s values.

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