large language models

The Comptroller and Auditor General (CAG) is developing a sovereign AI platform, including a large language model, to enhance public sector auditing. This initiative aims to leverage AI and ML tools for faster and more efficient identification of procurement concentration and cartel-risk indicators by analyzing vast datasets and detecting systemic patterns.

Mark Cuban warns that AI's inconsistency makes human expertise crucial, debunking 'doomer' fears of AI takeover. He highlights that AI's probabilistic nature means it doesn't grasp consequences, making human judgment indispensable. Meanwhile, Google and Uber CEOs reveal AI generates significant code, but human engineers always provide final approval, underscoring the continued need for human oversight.

Indian independent researcher Kunvar Thaman achieved a significant milestone with his solo-authored paper accepted at ICML 2026. His work introduces a benchmark to measure how AI agents exploit loopholes when using tools, a critical area in AI safety. This breakthrough highlights an independent voice succeeding in a field dominated by major institutions.

Aiza Usman, an AI Engineer at Anthropic, shares her journey into artificial intelligence. She moved from Pakistan to the US for studies, focusing on generative models and large language systems. Usman gained practical experience at a fintech startup before joining Anthropic. She highlights the company's commitment to responsible AI development.

Cybercriminals have reportedly leveraged artificial intelligence to craft a potent hacking tool capable of circumventing security measures in a widely used system administration software. Google's security team uncovered this novel threat, which exploited an unknown vulnerability to bypass multi-factor authentication. The developer was alerted and patched the flaw before it could be weaponized, marking a significant development in AI-driven cyberattacks.

Airbnb CEO Brian Chesky defended the company's use of Chinese AI models, stating US lawmakers misunderstand open-source systems. He clarified that Airbnb does not share user data with Chinese firms, primarily utilizing open-source models, including those from the US. This defense comes amid congressional concerns over data security and potential national security risks.

AI firm Anthropic is reportedly in early discussions with Microsoft to rent server space powered by the tech giant's custom-designed AI processors. This move aims to diversify Anthropic's hardware backend and reduce reliance on Nvidia, following a significant infrastructure deal with Google. The potential partnership underscores a broader industry trend of reducing dependence on Nvidia's dominant AI chips.

Google has released ten free artificial intelligence courses covering generative AI, large language models and responsible AI, available without fees or prior experience. The modules include prompt design, image generation, encoder-decoder architecture, attention mechanisms and transformer models such as BERT. Learners can also study image captioning systems and generative AI Studio for building applications. Each course offers short, structured lessons and practical exercises designed to help users understand core machine learning concepts and application development.

India's student startups are pivoting from consumer internet to deep-tech, with AI, defence, and climate leading the charge. This shift signifies a rise in technical sophistication, as evidenced by a surge in PhD founders and a widening geographical spread of innovation beyond major metros. Edtech and services sectors have seen a notable decline.

Google I/O 2026 kicks off at Shoreline Amphitheatre, promising major AI advancements and operating system updates. Expect next-generation Gemini models with agentic capabilities, a potential Android-ChromeOS merger for new "Googlebooks" laptops, and a closer look at augmented reality smart glasses. Creative AI tools like Veo and Lyria will also be showcased.

Europe faces a critical two-year deadline to build its own AI infrastructure. Mistral AI's Arthur Mensch warns of US technology dependence if Europe fails to act. Control over chips, energy, and computing power is key to AI competition. Failure could leave Europe reliant on American digital services. Mistral is working to expand Europe's AI capacity.

During the convocation at Nalanda University, P.K. Mishra, Principal Secretary to the Prime Minister, underscored the pivotal role of Asia in global affairs alongside India’s robust economic position. He championed a model of sustainable development that integrates knowledge, values, and harmonious living.

Palantir CTO Shyam Sankar challenges the narrative of AI-driven job losses, arguing that frontline workers, not tech executives, offer the true perspective. He highlights how automation in manufacturing and healthcare is enhancing efficiency and enabling job creation, rather than widespread displacement. Sankar emphasizes listening to those directly using AI in real-world applications.

Former NITI Aayog member Arvind Virmani advocates for India to leverage Artificial Intelligence to boost human productivity, not replace workers. He emphasizes combining AI with India's vast workforce, focusing on practical applications in healthcare, education, and agriculture, rather than solely pursuing robotics. Virmani also stresses early AI education integration and government support for AI research and development.

Take-Two Interactive CEO Strauss Zelnick believes AI can generate GTA lookalikes but not original hits, stating that clones don't sell. He emphasized that while AI excels at asset creation and improving efficiency, it lacks the originality and creative thinking essential for culturally influential games. Zelnick also dismissed concerns about AI democratizing game development, noting that technology has always been accessible.

Google DeepMind CEO Demis Hassabis has called AI-driven layoffs "dumb," telling Wired ahead of Google I/O that companies cutting engineers to chase AI productivity gains are showing a "lack of imagination." His comments come as Anthropic's Dario Amodei warns of mass white-collar job losses, and as Meta, Amazon, Block and others blame AI for thousands of cuts.

Cybercriminals are now leveraging AI to discover and exploit unknown software flaws, as revealed by Google's recent interception of a major attack. This development confirms fears about AI-powered hacking, prompting urgent discussions and a cautious approach to releasing advanced AI models. The cybersecurity landscape faces a new arms race as attackers gain speed.

Palantir faces increasing competition for US defense contracts as rival AI firms challenge its dominant position. Despite Palantir's criticism of AI lab outputs as unreliable, competitors like OpenAI and Anthropic are expanding into its core business areas. Palantir views these labs as a supply chain, integrating their models into its operational platform.