Today in AI: Alexa+ hits 1 million | Google's new internet-less AI | OpenAI vs Microsoft
Storyboard18 brings you the top AI news of the day.
ADVERTISEMENT
The world of Artificial Intelligence has only begun to affect human lives. In times like these, staying up-to-date with the AI world is of utmost importance. Storyboard18 brings you the top AI news of the day.
Amazon’s generative AI-powered Alexa+ hits 1 Million Users in early access
Amazon’s next-generation virtual assistant, Alexa+, has already surpassed one million users—despite not yet being publicly available. According to a report by TechCrunch, the generative AI-powered service is currently in its Early Access phase, with access granted to customers via a waitlist system.
Alexa+ marks a significant leap in Amazon’s AI ambitions, offering natural, personalized conversations, deeper smart home integration, and broader capabilities powered by generative AI. While the service remains in preview, early users have taken to social media in recent weeks to share that they’ve received invites, underscoring growing demand. Once fully launched, Alexa+ will be free for Amazon Prime members, while non-Prime users will be able to subscribe for $19.99 per month. The move signals Amazon’s strategy to strengthen its Prime offering and take on rivals in the increasingly competitive AI assistant market. The company has not yet announced a formal public release date but continues to roll out access gradually. As part of the rollout, Amazon is positioning Alexa+ as more than just a voice assistant—it aims to be a proactive AI companion capable of understanding context, preferences, and handling complex requests with ease.
With over a million users already interacting with the enhanced assistant, Alexa+ could potentially redefine how consumers engage with AI-powered smart devices in their daily lives.
Court Orders OpenAI to Retain All ChatGPT Logs Amid Ongoing Privacy Lawsuit
A U.S. judge has dismissed concerns that an order requiring OpenAI to preserve all ChatGPT chat logs constitutes a “nationwide mass surveillance program,” ruling that the data retention is limited in scope and purpose.
The ruling came in response to objections filed by ChatGPT users after Judge Ona Wang instructed OpenAI to indefinitely retain user chat data—including deleted logs—as part of an ongoing copyright lawsuit.
One objector, a business owner who integrates ChatGPT into company operations, argued that the court order risked exposing confidential materials such as trade secrets and source code. The judge dismissed this filing on procedural grounds, noting the business failed to retain legal representation.
Another user argued the order would harm ChatGPT users by storing "highly sensitive" personal and commercial information, likening the court's directive to a national surveillance effort. Judge Wang rejected that argument as well, stating, “Proposed Intervenor does not explain how a court’s document retention order… is, or could be, a ‘nationwide mass surveillance program.’”
OpenAI Eats Into Microsoft’s Turf? ChatGPT Gains Ground in Corporate AI Race
The partnership between Microsoft and OpenAI, once hailed as one of the most synergistic collaborations in tech, is starting to show cracks—particularly as both companies now find themselves vying for the same enterprise customers, as reported by Bloomberg.
Behind the scenes, OpenAI’s growing presence in the corporate AI space is raising eyebrows at Microsoft, which has invested nearly $14 billion in the startup. Microsoft’s sales teams, already under pressure to drive Copilot adoption, now face the awkward reality of being undercut by their own partner.
While Microsoft has relied on deep-rooted enterprise relationships to promote Copilot, OpenAI is aggressively targeting the same clients with direct business offerings, educational tools, and a rapidly expanding suite of enterprise products. OpenAI’s acquisition of Windsurf, an AI coding assistant competing with Microsoft’s GitHub Copilot, is the latest sign of its independent ambitions.
As the lines between partner and competitor blur, one thing is clear: The battle for AI dominance in the enterprise space is just heating up. And for Microsoft, it may be fighting that battle on two fronts—against rivals like Google and Amazon, and now, against its own billion-dollar partner.
Dwarka Expressway gets India’s first AI-powered traffic management system
In a major step towards smart infrastructure, the Dwarka Expressway has become the first in India to deploy an AI-powered Advanced Traffic Management System (ATMS), aimed at enhancing road safety and streamlining vehicular movement.
The cutting-edge system, developed by the Indian Highways Management Company Ltd. (IHMCL) and implemented by the National Highways Authority of India (NHAI), spans a total project length of 56.46 kilometres. It uses artificial intelligence, sensors, and real-time data analytics to manage traffic flow and reduce accidents. The initiative is being hailed as a national model for AI-driven traffic governance and is expected to set new benchmarks in highway safety and efficiency.
The 29-kilometre Dwarka Expressway, a key connector between Delhi and Gurugram, runs from Shiv Murti on National Highway 8 to the outskirts of Gurugram. It serves as the third direct route between the two cities and offers a faster travel option for commuters heading towards Gurugram and the Indira Gandhi International Airport. With the introduction of ATMS, the expressway aims to improve emergency response, monitor traffic violations, and ensure smoother commutes—marking a significant upgrade in India’s road infrastructure framework.
Google DeepMind unveils AI model that runs without data network on robots
In a breakthrough for robotics and artificial intelligence, Google DeepMind has announced the launch of a new on-device AI model that can run directly on robotic systems—without the need for constant internet connectivity. Called Gemini Robotics On-Device, the AI-powered vision-language-action (VLA) model enables general-purpose dexterity and rapid task adaptation. The announcement was made via a blog post on Tuesday, June 24, by Carolina Parada, Senior Director and Head of Robotics at Google DeepMind.
“Since the model operates independently of a data network, it’s helpful for latency-sensitive applications and ensures robustness in environments with intermittent or zero connectivity,” Parada said.
The system builds on the Gemini Robotics platform introduced in March and is specially designed for bi-arm robotic devices. Unlike earlier models, this version can follow natural language instructions and perform complex, real-world tasks — such as unzipping bags, folding clothes, pouring salad dressing, or assembling products — with high precision.
Importantly, this is Google DeepMind’s first VLA model available for fine-tuning, allowing developers to further customize the AI for specific use cases. According to Parada, the model can adapt to new tasks with as few as 50 to 100 demonstrations, a significant improvement in task generalization and learning efficiency. While many actions will work “out of the box,” the on-device model's ability to generalize foundational knowledge and adapt quickly makes it suitable for a wide range of practical applications — from industrial automation to household robotics.
The launch marks Google’s growing momentum in the race to develop AI-powered generalist robots that can understand and execute human instructions with natural ease. It also reflects the broader shift in Silicon Valley toward multimodal AI systems that combine text, image, and audio processing to enable smarter machines. As tech giants including Google, Tesla, and Amazon race to embed large language models into robotics, Gemini’s offline, fine-tunable architecture could offer Google a key edge in real-world deployments.
