Meta is developing an advanced ‘agentic’ AI assistant designed to perform complex, multi-step tasks for consumers. The initiative reflects the company’s broader push to expand its AI capabilities beyond basic chat functions.
The planned assistant is intended to act more autonomously, helping users complete actions such as organising activities or managing digital tasks. Powered by a new internal model called Muse Spark, the assistant is still under development, and its rollout timeline depends on internal testing.
Meta’s strategy focuses on embedding these tools across its platforms, aiming to deepen user engagement and create more personalised digital experiences.
This marks a shift towards AI systems that can anticipate needs rather than simply respond to prompts. The move also signals intensifying competition among major technology companies in consumer AI.
The report highlights that Meta is positioning AI as central to its future growth, with a focus on making assistants more proactive and capable within everyday digital environments in the US.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
DeepSeek has again placed itself at the centre of the global AI race. After drawing worldwide attention with its R1 reasoning model in early 2025, the Chinese company has recently released DeepSeek V4, a new model designed to compete not only on performance, but also on price, openness and efficiency.
The hype around DeepSeek V4 is not based on a single feature. The model comes with a 1 million-token context window, open weights, two versions for different use cases and a strong focus on agentic workflows such as coding, research, document analysis and long-running tasks. In a market still dominated by expensive closed models, DeepSeek is trying to prove that powerful AI does not need to remain locked behind trademarked systems.
A model built for long memory
The most immediate difference between DeepSeek V4 and other models is context length. Both DeepSeek-V4-Pro and DeepSeek-V4-Flash support a 1-million-token context window, meaning they can process inputs far longer than those of older generations of mainstream models. According to DeepSeek’s official release, one million tokens is now the default across all official DeepSeek services.
For ordinary users, that may sound technical. In practice, it matters because a longer context allows models to work with large documents, long conversations, full codebases, legal materials, research archives or complex project histories without losing track as quickly.
That is why DeepSeek V4 is not just another chatbot release. It is aimed at the next stage of AI use, where models are expected to act less like question-answering tools and more like assistants that can follow long processes over time.
Two models for two different needs
DeepSeek V4 comes in two main versions. DeepSeek-V4-Pro is a larger and more capable model, with 1.6 trillion total parameters and 49 billion active parameters. DeepSeek-V4-Flash is a smaller model, with 284 billion total parameters and 13 billion active parameters, designed for faster and more cost-effective workloads.
That distinction is important. Not every user needs the strongest model for every task. A company summarising documents, routing queries or running basic support may choose Flash. A developer working on complex coding tasks, long-context agents or advanced reasoning may prefer Pro.
DeepSeek’s release reflects a broader trend in AI. The best model is no longer always the biggest one. Cost, speed, context size and deployment flexibility are now as important as raw benchmark performance.
Why the price matters
One reason DeepSeek attracts so much attention is its aggressive pricing. DeepSeek’s API page lists V4-Flash at USD 0.14 per 1 million input tokens on a cache miss and USD 0.28 per 1 million output tokens. V4-Pro is listed at USD 1.74 per 1 million input tokens and USD 3.48 per 1 million output tokens before the temporary 75% discount.
For developers and companies, that changes the calculation. High-performing AI models are useful only if they can be deployed at scale. If every long document, coding session or agentic workflow becomes too expensive, adoption slows down.
DeepSeek’s challenge to the market is therefore not only technical. It is economic. The company is pushing the idea that frontier-level AI should be cheaper to run, easier to access and less dependent on closed ecosystems.
The architecture behind the hype
DeepSeek V4 uses a mixture-of-experts approach, meaning only part of the model is active during each response. That helps explain why the model can be very large on paper, yet still more efficient to run than a dense model of similar overall size.
The more interesting part is how DeepSeek handles long context. NVIDIA’s technical overview explains that DeepSeek V4 uses hybrid attention, combining compression and selective attention techniques to reduce the cost of processing very long prompts. NVIDIA says these changes are designed to cut per-token inference FLOPs by 73% and reduce KV cache memory burden by 90% compared with DeepSeek-V3.2.
For a non-technical audience, the point is simple. DeepSeek V4 is trying to solve one of the biggest problems in modern AI: how to make models remember and process much more information without becoming too slow or too expensive.
That is where much of the hype comes from. The model is not merely larger. It is designed around the economics of long-context AI.
Why NVIDIA is still in the picture
NVIDIA’s role in the DeepSeek V4 story is especially interesting. DeepSeek is often discussed as part of China’s effort to build a more independent AI ecosystem, but NVIDIA has also been quick to move forward to support developers who want to build with the model.
In its technical blog, NVIDIA describes DeepSeek V4 as a model family designed for efficient inference of million-token contexts. The company says DeepSeek-V4-Pro and V4-Flash are available through NVIDIA GPU-accelerated endpoints, while developers can also use NVIDIA Blackwell, NIM containers, SGLang and vLLM deployment options.
NVIDIA also reports that early tests of DeepSeek-V4-Pro on the GB200 NVL72 platform showed more than 150 tokens per second per user. That matters because long-context models place heavy memory pressure, as well as on compute and networking infrastructure. The model may be efficient by design, but serving it at scale still requires serious hardware.
So, DeepSeek V4 does not remove NVIDIA from the story – it complicates it. The model is part of a broader push towards more efficient AI, but the infrastructure race remains central.
The chip question behind the model
DeepSeek V4 also arrives at a time when AI infrastructure is becoming just as important as model performance. MIT Technology Review frames the release partly through that lens, noting that DeepSeek’s new model reflects China’s broader attempt to reduce reliance on foreign AI hardware and build a more self-sufficient technology stack.
That detail matters because the AI race is no longer only about who builds the most capable model. It is also about who controls the chips, software frameworks and data centres needed to run it.
Replacing NVIDIA, however, remains difficult. Its advantage lies not just in its chips, but also in the software ecosystem developers have built around its platforms over many years. Moving to alternative hardware means adapting code, rebuilding tools and proving that the new systems are stable enough for serious use.
DeepSeek V4, however, sits between two realities. It points towards China’s ambition to build a more independent AI stack, while NVIDIA’s rapid support for the model shows that frontier AI still depends heavily on established infrastructure.
Open weights as a strategic move
DeepSeek V4 is also important because the model weights are available through Hugging Face under the MIT License. That gives developers more freedom to inspect, adapt and deploy the model than they would have with a fully closed commercial system.
Open-weight models are becoming a major pressure point in the AI race. Closed models may still lead in some areas, especially in polished consumer products, enterprise support and safety layers. However, open models offer something different: flexibility.
For universities, start-ups, smaller companies and developers outside the largest AI ecosystems, that flexibility matters. It means advanced AI can be tested, modified and integrated without relying entirely on a handful of dominant providers.
Benchmarks need caution
DeepSeek presents V4-Pro as highly competitive across reasoning, coding, long-context and agentic benchmarks. Hugging Face lists results including 80.6 on SWE-bench Verified, 90.1 on GPQA Diamond and 87.5 on MMLU-Pro for DeepSeek-V4-Pro.
Those numbers are impressive, but they should not be treated as the full story. Benchmarks are useful, but they rarely capture every real-world use case. A model can score well on coding tests and still struggle with reliability, factual accuracy, safety or complex multi-step workflows in production.
That caution is important. The AI industry often turns benchmarks into headlines, while real performance depends on deployment, prompting, safety controls and the specific task at hand.
More than just another model release
DeepSeek V4 matters because it combines several trends into one release: long context, lower prices, open weights, agentic workflows and geopolitical competition. It also shows that the AI race is no longer fought only in labs, benchmarks and data centres. Visibility now matters too. Tools such as Diplo’s Digital Footprints show how digital presence shapes the way technology actors and media narratives are discovered, ranked and understood. At this stage, the competition is not only about who has the smartest model. It is also about who can make intelligence cheaper, more available and easier to deploy.
That does not mean DeepSeek has solved every problem. Questions remain around independent benchmarking, safety, data governance, infrastructure and the broader political context of Chinese AI development. Still, the release does show where the market is heading.
The next phase of AI may not be defined solely by the most powerful model. It may be defined by the model that is powerful enough, affordable enough and open enough to change how people build products, services and tools with AI.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
Kraken has entered a strategic partnership with MoneyGram to enable crypto-to-cash withdrawals in more than 100 countries. The integration links digital asset infrastructure with MoneyGram’s global network, allowing users to convert crypto into hundreds of fiat currencies through physical and digital payout channels.
The service is intended to address one of the main barriers to crypto adoption by improving access to reliable off-ramps. Users will be able to transfer funds to their accounts and receive near-instant cash payouts through MoneyGram’s retail network and regulated payment infrastructure.
Both companies highlighted the importance of interoperability between traditional finance and digital assets in driving practical adoption.
Kraken stressed the value of connecting liquidity and compliance systems with established payment rails, while MoneyGram presented its global distribution network as a bridge between digital value and everyday financial use.
The rollout will begin across the United States, Europe, Latin America, Africa, and parts of Asia-Pacific, with plans to expand further into local bank deposits and additional payment services as the partnership develops.
Why does it matter?
The partnership addresses one of the main friction points in crypto adoption: converting digital assets into usable cash at scale. By linking crypto infrastructure with a global payout network, it strengthens the practical use of digital assets beyond trading and speculation.
More broadly, it reflects a gradual convergence between traditional financial rails and crypto-native systems, with interoperability becoming increasingly important to how value moves across borders.
It may also support financial inclusion by expanding access to cash-out services in regions where banking infrastructure remains limited or uneven.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
Meta has introduced USDC payouts for selected Facebook creators in Colombia and the Philippines, marking another step towards using blockchain-based payment rails for creator earnings. The programme allows eligible users to receive funds directly into crypto wallets using Polygon or Solana as settlement networks.
Creators receiving USDC on Polygon can move funds through supported wallets or exchanges and convert them into local currency where off-ramp services are available. The model reduces reliance on traditional cross-border payment channels and is intended to give creators faster and more flexible access to dollar-denominated earnings.
Polygon has been included alongside Solana as part of the payout infrastructure, with Polygon arguing that its network already handles a large share of global USDC transfer activity. Low transaction costs and broad wallet and exchange support are presented as key reasons stablecoin rails are becoming more attractive for recurring digital payouts.
Why does it matter?
The significance of the move lies less in crypto branding than in payment infrastructure. Meta is testing whether stablecoin rails can make creator payouts faster, more flexible, and less dependent on the frictions of traditional cross-border transfers. If this model scales, it would suggest that blockchain networks are becoming useful not only for trading or speculation, but for mainstream platform payments where speed, settlement, and access to dollar-denominated value matter.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
The UK government is stepping up efforts to better understand the structure and growth of its AI sector through an updated national survey led by the Department for Science, Innovation and Technology.
The research, conducted by Ipsos and supported by Perspective Economics, aims to gather direct insights from businesses operating in the UK AI ecosystem. The findings are expected to inform future government policy on AI and sector development.
Participation is voluntary and confidential. Respondents are drawn from senior leadership roles, including chief executives, chief technology officers, company directors, and senior members of AI or data science teams. The survey focuses on business activity, products and services, and longer-term growth plans across the sector.
Fieldwork is taking place between late April and the end of May 2026 using online questionnaires and telephone interviews. Each session is expected to last around 15 to 20 minutes, allowing businesses to contribute structured input without significant disruption to normal operations.
The initiative reflects a wider UK policy priority: ensuring that government strategy keeps pace with developments in AI innovation and commercial growth. By drawing on direct industry evidence rather than relying only on secondary analysis, policymakers are trying to build a more accurate picture of the country’s evolving AI landscape. This last sentence is an inference based on the survey’s stated purpose of informing government AI policy.
Why does it matter?
AI policy is much easier to design in theory than in a market that is changing quickly and unevenly. If the government lacks current information on how AI firms are growing, what products they are developing, and where the main constraints lie, it risks shaping policy based on outdated assumptions. Direct input from businesses gives policymakers a stronger basis for decisions on support, regulation, skills, and investment, especially at a time when the UK is trying to turn AI ambition into measurable economic capacity.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
The paper links AI adoption in digital banking to customer experience, risk management, process automation, financial inclusion and regulatory compliance, arguing that these factors are increasingly important as Bangladesh’s financial sector becomes more digital.
A study that uses a narrative literature review of recent research from 2024 and 2025 and builds its conceptual model on the UTAUT2 framework, which is commonly used to explain technology adoption.
The authors extend the model by adding ethical trust and fraud prevention as mediating mechanisms, arguing that consumers are more likely to use AI-enabled banking services when they see them as useful, secure, transparent and fair.
Ethical trust is treated as a central part of adoption. The paper identifies transparency, algorithmic fairness, data privacy, reliability, accountability and digital inclusion as key factors shaping how users respond to AI in banking.
It also notes that explainable AI tools and localised interfaces, including Bengali-language systems, could help reduce uncertainty for users with lower digital literacy.
Fraud prevention is presented as a critical enabler of consumer confidence. The authors point to real-time monitoring, anomaly detection, secure authentication, biometric e-KYC and explainable fraud alerts as tools that can reduce perceived risk.
Additionally, they argue that AI systems should not only detect fraud effectively, but also explain decisions clearly enough for users to trust them.
The paper also highlights Bangladesh-specific issues, including Islamic banking, Shariah-compliant AI models, rural and urban digital access gaps, and the need for inclusive design. However, the study remains conceptual and has not yet been peer reviewed.
The authors recommend future empirical research with Bangladeshi banking users to test the model across income levels, regions, generations and gender groups.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
Gemini can now generate downloadable and ready-to-share files directly in chat across a wide range of formats, including PDF, Microsoft Word, Excel, Google Docs, Sheets, and Slides.
The new feature is meant to remove the extra steps that often follow AI-assisted brainstorming, such as copying content into other applications and reformatting it manually. Instead, users can ask Gemini to create a structured file that is already formatted and ready to download or export to Google Drive.
Supported formats include Google Workspace files, PDF, DOCX, XLSX, CSV, LaTeX, TXT, RTF, and Markdown. The company says the feature is now available globally to all Gemini app users.
Possible uses include turning budget plans into spreadsheets, organising rough ideas into structured documents, converting long discussions into concise reports, and generating PDF study guides from uploaded lecture notes.
Why does it matter?
What changes here is not simply that Gemini can create more file types, but that it moves AI one step closer to replacing part of the software workflow itself. Instead of using AI to generate rough text and then finishing the task manually in Word, Excel, or Google Docs, users can now get output in a format that is already structured for immediate use.
That may reduce friction between prompting and execution, making AI more useful in everyday work, study, and administration. In practical terms, the update pushes Gemini further from being just a conversational assistant towards becoming a tool that can produce finished digital outputs people can actually work with.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
The United Nations has warned that the rapid adoption of AI in advertising could deepen a global information integrity crisis. With worldwide advertising spending now exceeding $1 trillion annually, concerns are growing over how automated systems influence what users see, trust, and engage with online.
A briefing by the Department of Global Communications and the Conscious Advertising Network places advertising at the centre of the digital information ecosystem. It argues that advertising helps fund and shape the systems that influence what people see and believe, while AI-driven tools are increasingly being used in media buying and content generation in ways that can amplify disinformation, hate speech, and opaque decision-making.
Transparency gaps in AI advertising systems are also raising concerns over fraud, inefficiency, and declining trust in digital platforms. The report warns that these pressures can weaken independent journalism and reduce advertising effectiveness as confidence in online environments continues to erode.
UN officials and industry representatives are calling for stronger governance, clearer oversight of AI supply chains, and closer cooperation between regulators, advertisers, and civil society. The core message is that without stronger guardrails, AI may accelerate the breakdown of information ecosystem integrity rather than simply improve commercial performance.
Why does it matter?
AI is becoming embedded in systems that shape the online flow of information, which means advertising is no longer only a commercial mechanism but also a force affecting public perception and trust. As automation expands without clear oversight, risks can spread beyond brand safety into disinformation, media sustainability, and democratic discourse.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
Kazakhstan has introduced an AI-powered assistant designed to simplify the process of starting a business, according to Zhaslan Madiyev. Developed in cooperation with the Ministry of Finance, the platform aims to provide data-driven guidance to early-stage entrepreneurs.
Built around a digital mapping system, the assistant evaluates factors such as nearby businesses, customer flow, and competition. Its recommendations aim to help users choose more viable locations and avoid oversaturated sectors, thereby reducing the risk of duplicating businesses in the same area.
Officials say the tool could reduce startup operating costs by up to half while improving long-term business sustainability. Alongside it, a second AI assistant already provides continuous guidance on tax reporting and regulatory compliance, translating complex requirements into clearer, more practical steps for users. According to Kazakhstani reporting, the tax assistant has already processed more than 5,000 requests.
The development forms part of Kazakhstan’s wider digital transformation agenda, which aims to modernise public services and strengthen the country’s digital economy through practical AI deployment. The government says more than 50 AI-powered services are now being developed to support citizens and businesses.
Why does it matter?
Kazakhstan’s AI assistant points to a shift from basic digital services towards more active, real-time decision support for entrepreneurs. Data-driven recommendations can help reduce startup risks, limit market oversaturation, and support more efficient resource allocation across local economies.
Simplified tax and compliance guidance also targets one of the main barriers facing early-stage businesses: administrative complexity. Placed within Kazakhstan’s broader AI-first digital strategy, the initiative signals a wider move towards a more competitive and operationally AI-driven digital economy.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
China has introduced new online marketing rules for financial products, further tightening its long-standing restrictions on cryptocurrency-related activity. The new framework limits the promotion of financial products to licensed entities and treats digital currency trading and issuance as illegal financial activity.
Issued by the People’s Bank of China and seven other regulators, the Administrative Measures for Online Marketing of Financial Products will take effect on 30 September 2026. The rules extend responsibility to platforms, intermediaries, and content creators who promote or facilitate financial products online.
Any assistance in promoting or facilitating prohibited financial activity may now be treated as participation in illegal finance, expanding enforcement beyond direct trading bans. In practice, that broadens the focus from financial products themselves to the wider digital promotion layer, including online displays, traffic generation, and other forms of internet-based marketing support.
Authorities say the measures are intended to protect consumers by limiting misleading or aggressive online promotion, including livestream marketing and viral investment content. In that sense, the rules are not only about crypto, but about tighter control over how financial products are marketed in digital environments.
The policy also reinforces China’s existing position, dating back to 2021, when regulators declared all cryptocurrency transactions illegal, while pushing enforcement deeper into the digital advertising and distribution layers of financial markets.
Why does it matter?
Stronger oversight of online financial promotion shows that crypto-related advertising is increasingly being treated as a regulatory risk category, not just a marketing issue. The Chinese move also points to a broader trend in which regulators are extending scrutiny beyond financial products themselves to the digital channels, influencers, and platforms that help distribute them.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
