NerdNews

Google is using AI to improve its advertising products, while OpenAI is testing ads in its ChatGPT service, sparking concerns about credibility and safety. Google's approach focuses on ad quality, advertiser tools, and new AI user experiences, whereas OpenAI's ads will be limited to free and low-tier users, with safeguards in place to maintain trust.

xAI held a public all-hands meeting where they discussed their product roadmap, organizational structure, and interplanetary ambitions, including a moon-based factory for AI satellites and a lunar mass driver. The company also shared usage and revenue figures, with X exceeding $1 billion in annual recurring revenue and Imagine generating 50 million videos a day.

OpenAI has disbanded its mission alignment team, which aimed to ensure artificial general intelligence benefits humanity. The team's leader, Josh Achiam, is now the company's chief futurist, studying the impact of AI on the world.

A Microsoft VP discusses how AI changes the math for startups, making it easier and cheaper to launch and maintain ventures, and how agentic AI can reduce the cost of software operations, leading to higher-valuation startups with fewer people.

Apple has delayed its planned overhaul of Siri, which was set to introduce more personalized features. The update, initially planned for iOS 26.4, has been pushed back to later versions of iOS, including iOS 26.5 and iOS 27. The company is also working on an AI chatbot version of Siri, set to launch with iOS 27.

OpenAI researcher Zoë Hitzig resigned over concerns that ChatGPT ads could manipulate users, drawing parallels to Facebook's history of prioritizing profits over user privacy. Hitzig warned that OpenAI's advertising strategy risks repeating the same mistakes, and proposed alternative structures such as cross-subsidies and independent oversight boards.

Microsoft has warned customers of a technique called 'AI Recommendation Poisoning' where attackers manipulate AI models to produce biased advice. This is done by adding hidden instructions to 'Summarize with AI' buttons and links on websites, which can include manipulative prompt text. The technique can erode trust in AI services and make users vulnerable to subtle biases.

A researcher at OpenAI has quit over concerns that the company's introduction of advertisements to ChatGPT could compromise user data and privacy. The researcher warned that the company's 'archive of human candor' is unprecedented and potentially dangerous, as it contains sensitive information shared by users without realizing how it could be used.

As AI becomes ubiquitous, individuals and groups are pushing back against its influence, citing concerns over data governance, privacy, and job displacement. From 'tarpitting' AI bots to refusing AI-powered upgrades, citizens are taking matters into their own hands, but it's unclear if these efforts will be enough to effect lasting change.

This article provides a playbook for developing agentic AI applications, focusing on identifying agentic components, implementing, deploying, testing, and tracing agents. It covers architecture and design patterns, development workflows, versioning, and automated testing.

Google has entered a power purchase agreement with TotalEnergies to build two solar farms in Texas, providing 1GW of solar power to fuel its AI ambitions. The deal will supply Google's datacenters with an estimated 28 terawatt hours of renewable electricity over 15 years, supporting its $185 billion AI spending spree.

Humanoid robots are being developed to perform various tasks, from factory work to household chores. However, they still face significant challenges, including limited battery life, processing power, and AI capabilities. Researchers are working to improve these robots, with some predicting that they will become ubiquitous in the next few decades. Companies like Tesla and Boston Dynamics are already making progress in this field, but there is still a long way to go before humanoid robots can truly replicate human abilities.

Apple's new Creator Studio subscription offers premium features for productivity apps like Keynote, Pages, and Numbers, but users are unhappy with the introduction of ads and potential macOS bloatware strategy. The subscription costs $12.99/month or $129/year and includes intelligent features and premium content.

A report by BCS found that only 20% of European datacenters are AI-ready, due to skills shortages, land and labor issues, and grid bottlenecks. Despite growing demand, available supply is forecast to shrink, with 95% of respondents expecting the shortage of skilled professionals to worsen.

OpenAI's Atlas browser is an agentic browser that uses AI to handle tasks such as downloading forms and navigating complex websites. The browser is designed to balance the need for AI assistance with the need for human control and decision-making. The Atlas team, including Ben Goodger and Darin Fisher, discuss the development of the browser and the potential impact on the future of the web.

A Harvard Business Review study found that employees who used generative AI worked at a faster pace, took on more tasks, and extended their work hours, often without being asked to do so. However, this led to feelings of being stretched too thin, cognitive fatigue, and burnout. The researchers suggest implementing standards to prevent burnout, such as intentional pauses and regulating the tempo of development.

Microsoft has rolled out fixes for security vulnerabilities in Windows and Office, which are being actively exploited by hackers. The vulnerabilities, known as zero-days, can be exploited by tricking someone into clicking a malicious link or opening a malicious Office file, allowing hackers to plant malware or gain access to a victim's computer.

SpaceX and others plan to launch orbital AI data centers, but high costs, technological hurdles, and harsh space environment pose significant challenges. Experts estimate that launching a 1 GW orbital data center could cost $42.4 billion, almost 3x its ground-bound equivalent. To make space data centers viable, costs need to come down, and technological advancements are required.

Google has released the first beta of Android 17, which brings performance improvements, new camera capabilities, and a continuous developer release plan. The company is adopting a Canary channel to push updates for developers, allowing for over-the-air updates and more integrated workflows.

Meridian AI, a startup aiming to remake the agentic spreadsheet, has raised $17 million in seed funding. The company's goal is to make financial modeling and spreadsheets more predictable and auditable using AI. Meridian operates as a stand-alone workspace, integrating data sources and minimizing hallucinations that slow down enterprise deployments.