Google is rolling out new AI-driven features for Chrome’s desktop version, powered by its Gemini technology. These updates include the introduction of Google Lens for desktop, accessible from the address bar and menu, allowing users to select parts of a web page or objects in images to ask questions and receive search results. The feature aims to enhance the user experience by enabling more interactive and visual searches.
Another addition is the Tab Compare feature, which is designed to simplify online shopping. It provides an AI-generated summary of similar items from different tabs, presenting product specifications, features, prices, and ratings in one consolidated view. This tool will help users make more informed purchasing decisions by comparing items across multiple pages.
Google also improves search functionality by integrating natural language queries for browsing history. Users can search their history using everyday language, making it easier to find previously visited pages. For example, users can ask, ‘What was that ice cream shop I looked at last week?’ and get relevant results based on the URL, title, and content of the pages.
These updates reflect Google’s ongoing efforts to integrate AI into its services, following previous enhancements such as a writing assistant, tab organiser, and theme creator introduced earlier this year. By leveraging AI, Google aims to provide users with a more intuitive and efficient browsing experience.
An Israeli start-up, Run, is under scrutiny as the US Justice Department investigates its acquisition by Nvidia on antitrust grounds. The $700 million deal, announced in April, raises concerns over Nvidia’s growing dominance in the AI chip market. Run’s technology, which optimises graphics processing units (GPUs), has become highly valuable amid soaring demand for these chips.
Nvidia, which has seen profits surge due to its AI-adapted chips, is now under investigation for potentially anti-competitive practices. Regulators fear that the company’s control over high-end AI chips and related software could stifle competition and innovation in the AI industry. Nvidia spokesperson Mylene Mangalindan defended the company, emphasising its commitment to competition and innovation.
Why does this matter?
The Justice Department’s investigation is part of a broader regulatory effort to address potential monopolistic behaviours in the AI sector. To ensure fair competition, the DoJ and the Federal Trade Commission are examining the business practices of leading AI companies, including Nvidia and Microsoft.
The investigation into Nvidia’s acquisition strategy and business practices reflects broader concerns about market concentration in the tech industry. As regulators continue to scrutinise AI investments and mergers, the outcome of these investigations could significantly impact the future landscape of the AI market.
OpenAI, previously a close partner of Microsoft, is now officially recognised as a competitor. Microsoft’s recent SEC filing marks the first time the company has publicly acknowledged this shift. OpenAI is now listed alongside tech giants like Google and Amazon as a competitor in both AI and search technologies.
The relationship between the two companies has been under scrutiny, with antitrust concerns arising from the FTC. Microsoft’s decision to relinquish its board observer seat at OpenAI follows a series of significant events, including the brief dismissal of OpenAI’s CEO Sam Altman. The filing may reflect a strategic move to alter public perception amid these investigations.
Silicon Valley has a history of companies navigating complex relationships, balancing roles as both partners and competitors. The dynamic between Yahoo and Google in the early 2000s serves as a notable example. Microsoft and OpenAI might be experiencing a similar evolution, with both entities maintaining competitive and cooperative elements.
Meanwhile, Microsoft continues to expand its own AI initiatives. The hiring of Inflection AI co-founders to lead a new AI division and the development of Microsoft Copilot highlight the company’s broader strategy. The diversification suggests a strategic approach to AI that goes beyond its ties with OpenAI.
Following a recent lawsuit by the Recording Industry Association of America (RIAA) against music generation startups Udio and Suno, Suno admitted in a court filing that it trained its AI model using copyrighted songs. Suno claimed this was legal under the fair-use doctrine.
The RIAA’s lawsuit, filed on 24 June, alleges that both startups used copyrighted music without permission to train their models. Suno’s admission is the first direct acknowledgement of this practice. Suno CEO Mikey Shulman defended the use of copyrighted material on the open internet, comparing it to a kid learning to write rock songs after listening to the genre.
The RIAA responded by calling Suno’s actions ‘industrial scale infringement’ that does not qualify as fair use. They argued that such practices harm artists by repackaging their work and competing directly with the originals. The outcome of this case, still in its early stages, could set a significant precedent for AI model training and copyright law.
As per a report by the World Bank and International Labour Organization, between 2% to 5% of the jobs in Latin America and the Carribean are at risk of being redundant because of automation. The highest risk of automation confronts women and young workers in the formal sector. That is a huge challenge for one of the world’s most unequal regions, where most low-paying jobs are concentrated in the informal economy.
Despite the growth in AI, the report also highlighted how ‘gaps in digital infrastructure and other inequalities could hinder the potential impacts of generative AI in the region’. The figure hovers between 26% and 38% of the total employment in Latin America and the Caribbean regarding jobs being exposed by generative AI and being impacted by it somehow.
On the flip side, researchers also draw attention to the fact that AI could bring many benefits by enhancing workers’ productivity between 8% and 14% of jobs, especially for high-income earners working in the urban formal sector.
The Isle of Man, a self-governing British Crown dependency in the Irish Sea between England and Ireland, has launched a programme to upskill its residents on the use of AI. By tapping on the ‘enhanced productivity’ of the workers, the government intends to improve its GDP by 10%. Chief executive of Digital Isle of Man, Lyle Wraxell, explained how the program and a public roadshow accompanying it will help ensure that people ‘keep up with the rapidly evolving technology’.
The initiative, ‘Activate AI,’ will unfold in three stages: the first will focus on education, the second will enable the island to establish a global positioning, and the final stage will be acceleration. A new platform will launch by the end of August, providing 200 hours of complimentary training for businesses and individuals.
The programme’s applied AI service will help companies pair up with experienced firms to explore AI-driven solutions for their business challenges. Given that this facility will be provided free of cost, board member Jason Bissell highlighted how this approach will enable the creation of a partner ecosystem, allowing local businesses to experiment and understand the technology’s potential value.
Teachers have long worried about ChatGPT enabling students to cheat, with its ability to produce essays and solve problems quickly. According to Chegg, 40% of undergraduates worldwide have used generative AI, with many relying on it daily. However, education experts at the Fortune Brainstorm AI Singapore conference believe that generative AI can now be a valuable tool for educators.
Initially, educators feared widespread plagiarism with the introduction of ChatGPT, said Sonita Jeyapathy from the National University of Singapore. Yet, this concern has shifted as teachers began to see the potential of AI in enhancing their teaching methods. Khairul Anwar, founder of Malaysian edtech startup Pandai, noted that teachers seek AI assistance for lesson planning, student motivation, and professional development.
Pandai’s chatbot, designed to assist with homework, exemplifies this shift. Instead of providing direct answers, it guides students through problems step-by-step, encouraging more profound understanding. Tim Baldwin of the Mohamed bin Zayed University of Artificial Intelligence highlighted further AI advancements, such as personalised tutoring tailored to individual learning styles, making education more accessible.
Panelists acknowledged that AI-enabled cheating isn’t new and emphasised the importance of student motivation and institutional values in deterring dishonesty. Anwar suggested that framing education as a transactional path to material success can encourage cheating, advocating instead for highlighting the intrinsic value of learning.
Meta has informed investors that its heavy investments in generative AI won’t yield revenue this year but are expected to pay off in the long run. Founder Mark Zuckerberg explained that Meta is preparing for future computing needs, particularly its next AI model, Llama 4, which will require significantly more computing power than its predecessor. Despite the high costs, with capital expenditures projected to reach up to $40 billion this year, Meta aims to position Llama 4 as the industry’s most advanced model by next year.
CFO Susan Li acknowledged during the earnings call that generative AI would not contribute to this year’s revenue. However, she emphasised that Meta’s AI infrastructure is designed to be flexible and multifunctional, capable of supporting various AI needs, including content ranking and recommendations. Meta’s existing AI tools have already enhanced user engagement on platforms like Facebook and Instagram, and new video recommendation tools have further boosted interaction with features like Facebook Reels.
Zuckerberg also discussed AI’s potential to transform Meta’s advertising business. He predicted that AI would eventually handle tasks like writing personalised ad copy and managing campaign objectives, simplifying the process for advertisers. This vision aligns with Meta’s broader strategy of leveraging AI to improve user experience and business outcomes.
Why does this matter?
Despite significant losses in its Reality Labs division, which focuses on metaverse-related technology, Meta reported solid financial performance. The company showed robust growth with quarterly revenues of $39 billion and net income of $13.5 billion. Additionally, Meta’s daily user base reached 3.2 billion, and its new social media app, Threads, is nearing 200 million monthly active users, contributing to investor confidence and a rise in Meta’s share price.
Telford and Wrekin Council, the local authority of Telford and Wrekin in the ceremonial county of Shropshire, England, have introduced an upgraded AI phone assistant named Ask Tom, designed to improve customer service by handling queries without the need for human staff.
Initially launched in 2021 as a digital assistant, Ask Tom now has the capability to answer, listen, speak, and send information directly to customers’ devices. The enhancement aims to free up customer service staff to focus on more complex issues, although human advisors will still be available if needed.
The council highlighted that when Ask Tom was first introduced, it significantly reduced customer waiting times by 50%. With the new upgrade, the council expects the system to assist even more people by addressing multiple questions in a single call and preventing the need to contact different numbers for specific information.
Councillor Zona Hannington emphasised that the automated assistant is built on a well-tested, extensive knowledge base of council information. The council can allocate its staff to more critical tasks by automating simple queries, thus improving efficiency and effectiveness. Residents will benefit from 24-hour access to a broader range of council services, enhancing the overall service experience.
Meta’s AI ambitions have received a significant boost from WhatsApp’s 500 million users in India. During Meta’s second-quarter earnings call, CFO Susan Li highlighted that India has become the largest market for Meta AI usage, a remarkable achievement given the product’s recent launch. Users in India have generated billions of queries, showing high engagement and retention, particularly on WhatsApp.
Meta AI was initially launched in the US and expanded to India and over a dozen other countries with the upgraded Llama 3 model. Despite initial cultural challenges, such as generating images and jokes that were culturally insensitive, Meta has adjusted its algorithm to suit the Indian market better. Additionally, the AI now supports multiple languages beyond English, including Hindi and Spanish, broadening its accessibility.
CEO Mark Zuckerberg emphasised the increasing computational demands for future AI models, stating that training the upcoming Llama 4 model would require nearly ten times the computing power used for Llama 3 and so demonstrating Meta’s commitment to advancing its AI capabilities.
Apart from AI, Meta’s social networks also show promise. The Threads platform is nearing 200 million users, and Facebook usage among young adults in the US is rising, indicating continued growth and engagement across Meta’s suite of products.
