Artificial intelligence Archives | Page 2 of 607

California challenges federal approach with new AI rules

The government of California is advancing a more interventionist approach to AI governance, signalling a divergence from federal deregulatory preferences.

An executive order signed by Gavin Newsom mandates the development of comprehensive AI policies within 4 months, prioritising public safety and protecting fundamental rights.

The proposed framework requires companies seeking state contracts to demonstrate safeguards against harmful outputs, including the prevention of child exploitation material and violent content.

It also calls for measures addressing algorithmic bias and unlawful discrimination, alongside increased transparency through mechanisms such as watermarking AI-generated media.

Federal guidance has discouraged state-level intervention, framing such efforts as obstacles to technological leadership.

The evolving policy landscape reflects growing concern over the societal impact of AI systems, including risks to employment, content integrity and civil liberties.

An initiative by California that may therefore serve as a testing ground for future regulatory models, shaping broader debates on balancing innovation with accountability in digital governance.

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

Future intelligent networks move towards 6G era

Digital infrastructure remains central to modern society, with mobile networks forming the backbone of global connectivity and technological progress. Research efforts are increasingly focused on ensuring that today’s 5G systems evolve into more advanced and intelligent network platforms.

The future 6G era is expected to go beyond traditional connectivity, enabling immersive communication experiences, intelligent machine interaction, and the development of large-scale digital twins.

Networks are anticipated to become cognitive systems, capable of learning, adapting, and making autonomous decisions in real time.

Alongside new capabilities, future networks will further strengthen core requirements such as security, privacy, reliability, and resilience. Advanced distributed processing will be embedded across network architecture to support real-time operations and system stability at scale.

The 6G vision is aligned with the 2030 timeframe, with development built on open and standardised ecosystems that encourage global collaboration. Interoperability will remain central, supporting innovation and universal connectivity across devices and services.

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

Software development dominates AI agent tool usage

Researchers analysed 177,436 AI agent tools created between November 2024 and February 2026 using Model Context Protocol repositories. The study examines how AI agents use external tools to access and modify digital environments.

The tools are grouped into perception, reasoning and action categories based on their function. Perception tools access data, reasoning tools analyse information, and action tools modify systems such as files, emails or external platforms.

Software development accounts for 67% of all tools and 90% of downloads. The findings show that AI agents are primarily used to support coding tasks and related workflows.

The share of action tools increased from 27% to 65% over the 16 months analysed. Most action tools focus on medium-stakes tasks, though some are used for financial transactions and other higher-stakes activities.

The study also outlines a method to monitor AI agent usage through tool-level analysis. This approach can support oversight of risks linked to AI deployment in practical applications.

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

Ray-Ban Meta Gen 2 AI glasses expands smart eyewear line

Meta has unveiled its first prescription-optimised AI glasses, expanding its wearable line with Ray-Ban Meta Gen 2 models for everyday vision correction. The launch targets users who already rely on prescription eyewear, offering a more integrated and comfortable experience.

The range includes Blayzer Optics and Scriber Optics with adjustable hinges, nose pads, and temple tips for a better fit. Pre-orders begin at $499 in the United States via Meta and Ray-Ban platforms, with wider availability in optical retailers and select global markets from 14 April.

Alongside the hardware launch, Meta is introducing new frame and lens colour combinations across its Ray-Ban Meta and Oakley Meta collections.

Additional AI-driven features are also rolling out, including hands-free nutrition tracking, WhatsApp message summaries, and improved on-device recall capabilities designed to enhance everyday communication.

Further software updates extend functionality with discreet handwriting input, in-lens navigation across US cities, and expanded media recording tools. The company positions its AI glasses as a multifunctional platform combining vision correction, connectivity, and real-time assistance.

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

Smart TV viewing upgraded with YouTube AI feature

YouTube has expanded its conversational AI tool to smart TVs, marking a significant step in making home viewing more interactive. Viewers can now engage with content directly from their television screens using voice-enabled queries.

Access to the feature is simple. While watching a video, users can select the ‘Ask’ option and activate their remote’s microphone button to interact with the AI. Users can ask about similar content or a creator’s catalogue in real time, with prompts available to guide new users.

Initial rollout of the tool took place last year across mobile and web platforms, where it quickly became a practical companion for deeper content engagement. Users already use it to analyse podcasts, explore destinations, and understand content without pausing videos.

Expansion to smart TVs strengthens YouTube’s push to transform passive viewing into an interactive experience. Living room entertainment is increasingly shaped by AI-driven features, with real-time assistance now integrated directly into the home’s largest screen.

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

UN Global Mechanism on ICT security advances work, shifts focus to implementation

The United Nations Global Mechanism on developments in the field of ICTs in the context of international security and advancing responsible state behaviour in the use of ICTs held its second meeting, during which member states conducted a general exchange of views on the work of the newly established permanent forum.

The session, chaired by Ambassador Egriselda López of El Salvador, focused on agenda item four, during which 61 member states and three intergovernmental organisations delivered statements on priorities for the mechanism.

Delegations emphasised the transition from the previous Open-Ended Working Group (OEWG) to the new permanent mechanism, highlighting the need to build on existing agreements and move towards practical implementation. Several speakers stressed that the mechanism should focus on translating the agreed framework for responsible state behaviour in cyberspace into concrete outcomes, rather than negotiating new commitments.

Across statements, member states reaffirmed the five-pillar framework covering threats, norms and principles, the application of international law, confidence-building measures, and capacity development.

Capacity development was highlighted as a cross-cutting priority, particularly by developing countries and Small Island Developing States, which pointed to the need for demand-driven and sustainable approaches to strengthen cybersecurity capabilities. Delegations also noted challenges, including ransomware, threats to critical infrastructure, and the impact of emerging technologies such as AI.

Member states welcomed the establishment of two dedicated thematic groups, one addressing substantive ICT security challenges and another focused on capacity development, as a means to support more detailed discussions and implementation.

Several delegations reaffirmed that international law, including the UN Charter, applies to cyberspace and called for further work on its practical implementation. Many also emphasised the importance of maintaining a consensus-based, intergovernmental process, while enabling contributions from stakeholders, including the private sector, academia, and civil society, in line with agreed modalities.

The meeting forms part of the initial phase of the Global Mechanism’s work, following its establishment as a permanent UN forum on ICT security. The mechanism is expected to convene its first substantive plenary session in July 2026, alongside dedicated thematic group meetings scheduled for December 2026.

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

EU funding platform drives competitiveness in strategic technologies

The European Commission has highlighted the growing impact of the Strategic Technologies for Europe Platform (STEP), which has mobilised €29 billion to strengthen innovation and competitiveness across key sectors.

An initiative that supports the development and manufacturing of critical technologies, reinforcing the Union’s strategic autonomy.

Funding has been directed toward digital and deep-tech innovation, clean technologies, biotechnology and defence, combining resources from EU programmes and Member States.

Such a coordinated approach reflects efforts to reduce strategic dependencies instead of relying on fragmented investment strategies.

The platform has also improved access to funding, with hundreds of calls and projects supported across all Member States. Tools such as the STEP Seal and the planned AI-based access systems aim to simplify processes and attract further public and private investment into high-potential projects.

Looking ahead, the initiative is shaping broader reforms, including proposals for a European Competitiveness Fund. These developments signal a continued focus on streamlining funding mechanisms while supporting innovation ecosystems and long-term economic growth across Europe.

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

AI capacity partnership links UNDP and Intel in Lesotho and Liberia

The United Nations Development Programme and Intel are working together to expand AI training and digital skills in Lesotho and Liberia under a Memorandum of Understanding signed in March 2025. According to UNDP, the partnership is intended to combine global technical expertise with local leadership as both countries pursue broader digital transformation goals.

Lesotho and Liberia are approaching the issue from different starting points.UNDP says Lesotho is aiming for universal digital access by 2030, while Liberia is investing in AI in higher education and governance systems to prepare for the future digital economy. Through its partnership with Intel, the UN’s global development network says it is helping close gaps in AI literacy and capacity-building so communities can better understand how AI may affect everyday life.

In Lesotho, UNDP says it has already helped establish 40 Digital Skills Learning Labs and train 40 Digital Ambassadors, including teachers, religious leaders, and local influencers. Intel’s ‘AI for Citizens (AI Community Experiences)’ programme was introduced to provide locally relevant training materials for low-connectivity environments. UNDP says the onboarding included virtual sessions using games and storytelling, while analogue activities and puzzles were used to explain concepts such as computer vision.

Liberia’s work has focused more on higher education and the public sector. UNDP says it supported the University of Liberia in designing its first Master of AI programme through six online sessions with global experts and in-person workshops involving 20 faculty members. The collaboration also extended to government, with targeted training for nearly 100 officials on how AI could improve public service delivery and inform policy decisions.

Anshul Sonak, Global Head of Intel Digital Readiness Programs, said: ‘We are deeply honoured to be a part of the AI training collaboration in Liberia with UNDP. Bringing AI skills and digital literacy to a country rich in history and potential was an amazing experience. We look forward to more collaborations in the future and finding more opportunities for Intel to be a player in the region.’

UNDP says future phases may include expanding training to more communities and countries, adapting content to local languages and contexts, and adding online components as connectivity improves. Dhani Spiller, Head of UNDP’s Digital Capacity Lab, said: ‘This partnership shows what’s possible when we combine UNDP’s development mandate with the innovation and technical depth of private-sector leaders.’

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

New quantum threat could weaken cryptocurrency encryption systems

A new warning from Google says advances in quantum computing could weaken widely used cryptographic systems protecting cryptocurrencies and digital infrastructure. A new whitepaper suggests future quantum machines may need fewer resources than previously estimated to break elliptic curve cryptography.

The research focuses on the elliptic curve discrete logarithm problem, which underpins much of today’s blockchain security. Findings suggest quantum algorithms like Shor’s could run with fewer qubits and gates, increasing concerns about cryptographic resilience.

To address the risk, the paper recommends a transition to post-quantum cryptography, which is designed to resist quantum attacks. It also outlines short-term blockchain measures, including avoiding reuse of vulnerable wallet addresses and preparing digital asset migration strategies.

Google also introduced a responsible disclosure approach using zero-knowledge proofs to communicate vulnerabilities without exposing exploitable details.

The company says this balances transparency and security, supporting coordinated efforts across crypto and research communities to prepare for quantum threats.

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

Cloudflare adds LLM layer to client-side security detection pipeline

Cloudflare has announced two changes to its client-side security offering, making Client-Side Security Advanced available to self-serve customers and offering domain-based threat intelligence at no extra cost to all users on the free Client-Side Security bundle. The update is focused on browser-based attacks that can steal data via malicious scripts without visibly disrupting a website’s normal operation.

Cloudflare says its client-side security system assesses 3.5 billion scripts per day and monitors an average of 2,200 scripts per enterprise zone. According to the company, the product relies on browser reporting, including Content Security Policy signals, rather than scanners or application instrumentation, and requires only that traffic be proxied through Cloudflare.

A central part of the announcement is a new detection pipeline combining a Graph Neural Network (GNN) with a Large Language Model (LLM). Cloudflare says the GNN analyses the Abstract Syntax Tree of JavaScript code to identify malicious intent even when scripts are minified or obfuscated. Scripts flagged as suspicious are then passed to an open-source LLM running on Workers AI for a second-stage semantic assessment intended to reduce false positives.

Cloudflare says the GNN is tuned for high recall to identify novel and zero-day threats, but that false alarms remain a challenge at internet scale. Internal evaluation results cited by the company show that the secondary LLM layer reduced false positives in the JS Integrity threat category by nearly three times across the total analysed traffic, lowering the rate from about 0.3% to about 0.1%. On unique scripts, Cloudflare says the false-positive rate fell from about 1.39% to 0.007%.

The company also describes a recent case involving a heavily obfuscated malicious script named core.js. According to Cloudflare, the payload targeted Xiaomi OpenWrt-based home routers, altered DNS settings, and attempted to change admin passwords. Cloudflare says the script was injected through compromised browser extensions rather than by directly compromising a website, and adds that its GNN detected the malicious structure while the LLM confirmed the intent.

Cloudflare argues that the two-stage design provides structural detection via the GNN and broader semantic filtering via the LLM, enabling the company to lower the GNN decision threshold without sharply increasing alert volume. Every script flagged by the GNN is also logged to Cloudflare R2 for later auditing, which the company says helps it review cases where the LLM overrode the initial verdict.

Domain-based threat intelligence is now being made available to all Client-Side Security customers, including those not using the Advanced tier. Cloudflare says the move is partly a response to attacks seen in 2025 against smaller online shops, especially on Magento, where client-side compromises continued for days or weeks after public disclosure. By extending domain-based signals more broadly, the company says site owners can more quickly identify malicious JavaScript or suspicious connections and investigate possible compromises.

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