Estonia has launched a new initiative aimed at preparing students and teachers for the age of AI. The ‘AI Leap’ programme will provide access to popular AI chatbots, including an educational version of ChatGPT, to help build digital skills. Starting in September 2025, the programme will involve 20,000 high school students and 3,000 teachers, with plans to expand to vocational schools and an additional 38,000 students and 3,000 teachers in 2026.
Education Minister Kristina Kallas emphasised that Estonia’s economic competitiveness depends on how well the country adapts to AI, ensuring young people are equipped for the future. As part of the initiative, Estonia will also invest in teacher training to support the integration of AI in classrooms.
The programme is a public-private partnership, with negotiations underway with major AI companies, including OpenAI and Anthropic. OpenAI has expressed its pride in collaborating with Estonia to bring ChatGPT Edu to the education system, aiming to better prepare students for the workforce. Estonia’s use of AI in education is seen as a model that other countries may follow as the EU pushes to increase digital skills across Europe by 2030.
For more information on these topics, visit diplomacy.edu.
British universities have been urged to reassess their assessment methods after new research revealed a significant rise in students using genAI for their projects. A survey of 1,000 undergraduates found that 88% of students used AI tools like ChatGPT for assessments in 2025, up from 53% last year. Overall, 92% of students now use some form of AI, marking a substantial shift in academic behaviours in just a year.
The report, by the Higher Education Policy Institute and Kortext, highlights how AI is being used for tasks such as summarising articles, explaining concepts, and suggesting research ideas. While AI can enhance the quality of work and save time, some students admitted to directly including AI-generated content in their assignments, raising concerns about academic misconduct.
The research also found that concerns over AI’s potential impact on academic integrity vary across demographics. Women, wealthier students, and those studying STEM subjects were more likely to embrace AI, while others expressed fears about getting caught or receiving biased results. Despite these concerns, students generally feel that universities are addressing the issue of academic integrity, with many believing their institutions have clear policies on AI use.
Experts argue that universities need to adapt quickly to the changing landscape, with some suggesting that AI should be integrated into teaching rather than being seen solely as a threat to academic integrity. As AI tools become an essential part of education, institutions must find a balance between leveraging the technology and maintaining academic standards.
For more information on these topics, visit diplomacy.edu.
Nvidia has forecast strong growth for the first quarter, reinforcing confidence in the booming demand for AI chips. Orders for the company’s new Blackwell semiconductors were described as ‘amazing’, with CEO Jensen Huang stating that AI is advancing rapidly.
Investors had raised concerns last month after Chinese startup DeepSeek claimed to have developed cost-efficient AI models, but Nvidia’s results eased doubts.
Shares initially rose before fluctuating in extended trading, continuing a trend that has seen Nvidia’s stock surge over 400% in two years. The company generated $11 billion in revenue from Blackwell-related products in the fourth quarter, accounting for around half of its total data centre revenue.
Analysts had expressed scepticism about the transition to Blackwell and competition from DeepSeek, but the latest figures reassured investors.
Margins remain under pressure, with Nvidia forecasting a slight drop to 71% in the first quarter, below Wall Street estimates. Chief Financial Officer Colette Kress expects a recovery later in the year as production scales up and costs decline.
Despite some concerns about oversupply in AI infrastructure, Chinese firms have reportedly increased orders for Nvidia’s H20 AI chip, further supporting demand.
Nvidia’s fourth-quarter revenue rose 78% to $39.3 billion, surpassing expectations. Data centre sales surged 93% to $35.6 billion, continuing strong growth from the previous quarter.
Investors remain focused on Nvidia’s ability to maintain momentum in a competitive AI landscape, particularly as tech giants like Microsoft and Meta commit substantial investments to AI expansion.
For more information on these topics, visit diplomacy.edu.
Intuit has projected stronger-than-expected third-quarter revenue, driven by growing demand for its AI-powered financial services and a busy US tax season.
The company’s optimistic forecast sent its shares up nearly 5% in extended trading. Chief Financial Officer Sandeep Aujla expressed confidence in the firm’s performance, highlighting a strong start to the tax season.
The company, known for TurboTax, QuickBooks, and Credit Karma, has been expanding its use of AI to enhance financial management and automation.
Intuit Assist, its AI-powered assistant, is integrated across its products to provide personalised financial recommendations and streamline tasks like bookkeeping. The firm recently launched an AI tool for QuickBooks to help small businesses manage tax and financial operations more efficiently.
Revenue for the second quarter reached $3.96 billion, surpassing Wall Street estimates, while adjusted earnings per share also exceeded expectations.
For the third quarter, Intuit expects revenue between $7.55 billion and $7.60 billion, above analysts’ projections. However, its forecasted adjusted profit per share of $10.89 to $10.95 fell short of estimates. Despite this, the company reiterated its confidence in its full-year outlook.
For more information on these topics, visit diplomacy.edu.
Anthropic’s Claude 3.7 Sonnet, the latest in AI development, is being put to the test by playing Pokémon Red on Twitch. The livestream has drawn attention as viewers watch the AI slowly navigate through the game, reasoning its way through each step. While Pokémon is a game designed for children, it serves as a useful benchmark for evaluating the AI’s problem-solving skills, demonstrating how far AI models have come in reasoning through complex puzzles.
Despite its progress, Claude 3.7 Sonnet’s gameplay is far from perfect. Early on, the AI struggled with basic tasks, like leaving the starting town, but later managed to win several gym leader badges. Yet, its performance is often slow, with moments of confusion that highlight the challenges AI faces in navigating human-designed games. For example, Claude became fixated on a rock wall, unable to move past it until it reasoned a way around.
This experiment draws comparisons to the earlier phenomenon “Twitch Plays Pokémon,” where thousands of people worked together to guide a character through the game. Now, AI plays solo, and while it’s a fascinating display of technological progress, some viewers lament the shift from collaborative human gameplay to watching a machine take on the same challenges. Still, the experiment highlights the growing sophistication of AI in handling tasks once thought too complex for machines.
For more information on these topics, visit diplomacy.edu.
Quora has unveiled a new feature on its Poe platform that allows users to create custom AI-powered applications. Named Poe Apps, the tool enables users to describe the app they want to build, and it generates the necessary code, powered by Anthropic’s Claude 3.7 Sonnet. The apps can integrate various AI models, such as OpenAI’s o3-mini or Google’s Veo 2, allowing for personalised functionality.
Once created, these apps can be run alongside Poe’s chatbot window or function as standalone web apps, with the option to adjust the underlying code manually. Poe Apps can currently be shared with other users on the web, with iOS and Android versions expected soon. While free users get a daily points allotment to use the models, premium subscribers have access to more flexible packages.
Quora has also hinted at future monetisation options for app creators, expanding the potential of Poe Apps. With this new feature, Quora is not only enabling the creation of unique apps but also setting the stage for users to earn from their AI-driven innovations.
For more information on these topics, visit diplomacy.edu.
Meta Platforms is reportedly in talks to build a new data centre campus for its AI projects, potentially costing over $200 billion, according to sources familiar with the matter. The company is considering locations in states like Louisiana, Wyoming, and Texas, with senior executives visiting potential sites this month.
This comes as the AI sector sees a surge in investment, especially following the launch of Microsoft-backed OpenAI’s ChatGPT in 2022. Companies are eager to incorporate AI into their products, leading to significant spending on AI infrastructure.
Despite the report, a Meta spokesperson denied the claims, stating that its data centre plans and capital expenditures have already been disclosed and calling the rest ‘pure speculation’. Meta’s CEO, Mark Zuckerberg, had previously mentioned that the company plans to invest up to $65 billion this year to expand its AI infrastructure.
In comparison, Microsoft has pledged around $80 billion in data centre investments for fiscal 2025, while Amazon has indicated its 2025 spending could exceed $75 billion.
For more information on these topics, visit diplomacy.edu.
European AI-related stocks declined for a second consecutive day on Tuesday, as concerns over Microsoft‘s data centre expansion dampened investor sentiment.
A recent analyst note suggested Microsoft had cancelled major US data centre leases, raising questions about its AI and cloud investment strategy. The uncertainty comes ahead of Nvidia’s upcoming earnings report, which is expected to provide insight into the strength of AI demand.
Companies exposed to data centres and infrastructure saw significant losses, with Germany‘s Siemens Energy and France‘s Schneider Electric continuing their declines from Monday.
Italian cable manufacturer Prysmian and Swiss engineering firm ABB also suffered losses as analysts debated whether Microsoft’s actions signalled a broader trend or a temporary reassessment. Microsoft has maintained that its AI and cloud expansion plans remain unchanged.
Market analysts remain divided, with some viewing the selloff as an overreaction and a potential buying opportunity. The volatility follows last month’s global tech downturn triggered by China‘s low-cost AI model, DeepSeek.
Nvidia’s earnings on Wednesday will be closely watched as investors assess whether AI-related stocks can sustain their high valuations in the face of shifting market dynamics.
For more information on these topics, visit diplomacy.edu.
Dassault Systèmes has collaborated with Apple to integrate its industrial 3D software with the Vision Pro headset. Widely used across manufacturing, life sciences and urban planning, the software allows engineers and designers to create precise digital twins of real-world objects.
The French company serves hundreds of thousands of customers designing cars, aircraft and factories.
A new app called 3DLive is set for release this summer, enabling real-time collaboration on 3D models. Engineers and designers will be able to work remotely on industrial projects, detecting potential issues early in the design process.
The technology aims to streamline workflows, particularly for manufacturing, where maintenance challenges often arise too late.
Deep integration with Apple’s Vision Pro will allow 3DLive to leverage its advanced sensors and computing power. The software’s high level of scientific accuracy ensures it can be used for complex applications such as aircraft design, where precision is crucial.
Apple executives highlighted its potential to reduce costs and rework by facilitating earlier design decisions.
For more information on these topics, visit diplomacy.edu.
Alibaba has made its AI model Wan 2.1 publicly available, enhancing competition in video and image generation. The move aligns with growing interest in open-source AI, following the recent rise of cost-effective models from firms such as DeepSeek.
The company introduced four variants of Wan 2.1, capable of generating images and videos from text and image input. Each model supports varying levels of complexity, with the most advanced handling 14 billion parameters for improved accuracy.
In January, the company unveiled an upgraded version of its AI model, later rebranded as Wan, with a focus on generating highly realistic visuals. The model has achieved top rankings on VBench, a leaderboard assessing video generative AI capabilities.
A preview of Alibaba’s reasoning model, QwQ-Max, was also released, with plans for an open-source launch.
The firm announced a $52 billion investment in cloud computing and AI infrastructure over the next three years. The commitment highlights its long-term focus on advancing AI and maintaining a competitive edge in the rapidly evolving sector.
For more information on these topics, visit diplomacy.edu.
