Gemini Omni: Gemini app users in India can now edit videos using Omni AI model

Gemini Omni: Gemini app users in India can now edit videos using Omni AI model


Google has announced that users in India can now upload videos to Gemini and edit them using Gemini Omni, the company’s latest multimodal AI model. The new model allows users to transform videos using natural language prompts, making it possible to modify scenes, change visual styles, add new elements, or refine clips without relying on traditional video editing software. 

 


Gemini Omni was first introduced at Google’s annual developer conference, Google I/O 2026, alongside the Gemini 3.5 family of AI models. Google described Omni as a cinematic video generation and editing model capable of understanding text, images, audio, and video inputs along with real-world physics concepts. According to the company, users can upload existing videos, photos, drawings, or reference material and then use conversational prompts to create or edit videos while maintaining visual consistency across multiple changes. 

 
 


How to edit videos with Gemini Omni

 


Google has shared the following steps for users in India:


  • Upload a video from your device or saved files

  • Tell Gemini what changes you want to make

  • Let Gemini Omni edit and transform the video

  • Review and save the final version


The feature is available through the Gemini app and the web version of Gemini.


Gemini Omni: What’s new


Google said Gemini Omni combines the company’s AI reasoning capabilities with advanced content creation tools. The model can work with multiple input formats—including text prompts, images, videos, and voice references to generate a single video output. According to Google, the system uses its understanding of the real world to create more coherent and visually consistent results.

 


One of Gemini Omni’s key features is conversational video editing. Instead of using traditional editing tools, users can simply describe the changes they want through natural language prompts. Google also noted that the model remembers previous instructions, helping maintain consistency in scenes, characters, and visual elements across multiple edits.

 

Google highlighted several video editing capabilities coming with Gemini Omni: 


  • Edit videos using natural language: Users can edit videos simply by typing prompts, while Gemini Omni keeps scenes, characters, and physics consistent across changes.

  • Transform scenes and environments: Users can modify specific parts of a video or completely change the overall setting and visual style.

  • Change actions inside videos: Gemini Omni can alter what is happening in a scene, add new characters or objects, and reimagine moments differently.

  • Refine videos over multiple edits: Users can continue making changes across multiple prompts without losing continuity from the original scene.

According to Google, Omni is also designed to better understand real-world concepts like Physics’s gravity, movement, and fluid dynamics, helping generate scenes that appear more realistic. The company said the model combines Gemini’s knowledge of science, history, and culture with visual generation tools to create more context-aware and meaningful storytelling experiences rather than simply generating clips based on pattern matching.

 


Users can also upload drawings, reference photos, or existing footage and use them as the foundation for AI-generated scenes. According to the company, Omni can also apply visual styles, motion effects, and scene transitions based on either uploaded references or written prompts.

 

Additionally, Google said that users will be able to create a digital version of themselves using their own voice and appearance. The company added that voice references will be supported initially for audio-based inputs, while broader audio support will arrive later. 

 


Transparency

 


As part of its responsible AI push, Google confirmed that all videos created using Gemini Omni will include its invisible SynthID digital watermark. According to the company, users will also be able to verify whether a video was generated using Gemini Omni through tools integrated into the Gemini app, Chrome and Google Search.

 


Google has also open-sourced SynthID text watermarking technology so developers can integrate it into their own AI models and tools. The company also announced a partnership with NVIDIA earlier this year to expand the use of SynthID beyond Google products. OpenAI and ElevenLabs will also use SynthID watermarking, helping identify more AI-generated content across the web.

 



Source link

Intel debuts Arc G-series chips for Windows gaming handhelds, takes on AMD

Intel debuts Arc G-series chips for Windows gaming handhelds, takes on AMD



Intel has announced its new Arc G-Series processors for handheld gaming PCs to take on AMD’s Ryzen Z-series chips used in devices like the Asus ROG Ally, Lenovo Legion Go, and several other portable gaming systems. Intel’s first dedicated handheld gaming chip lineup includes the Arc G3 and Arc G3 Extreme processors based on the Panther Lake architecture, with the company claiming improved gaming performance, power efficiency, and battery life for portable gaming devices.

 


Intel said handheld systems powered by these chips will begin arriving from OEM partners starting June 2026, beginning with Acer’s Predator Atlas 8, MSI Claw 8 EX AI+, and a OneXPlayer device.

 


Intel focusing on handheld gaming performance and battery life


According to Intel, the Arc G-Series processors have been designed specifically for handheld gaming devices, with optimised core counts, power management systems, and software tuning aimed at balancing gaming performance with battery efficiency.

 


The company said the processors are intended to deliver smooth, immersive gameplay without compromising battery life.


Arc G-Series to support ray tracing and XeSS 3


Intel further said that the Arc G-Series processors can be configured with up to Arc B390 graphics based on the company’s latest Xe3 graphics architecture. The chips will support real-time ray tracing alongside Intel’s XeSS 3 technology suite.

 


According to Intel, XeSS 3 combines AI-powered upscaling, multi-frame generation, and low-latency technologies to improve gaming performance, frame smoothness, and responsiveness. XeSS Super Resolution uses AI upscaling for higher frame rates, while XeSS Multi-Frame Generation inserts additional interpolated frames for smoother gameplay.

 


The company added that Xe Low Latency is designed to reduce input lag by integrating directly with game engines.


New gaming-focused features announced


Intel has also introduced several gaming-focused software and connectivity features with the Arc G-Series lineup. One of the additions is Xbox Mode, which is a controller-optimised full-screen Windows 11 interface that unifies installed game libraries into a console-like experience. The processors will also support Intel Precompiled Shaders, which download prebuilt shader files from Intel’s cloud servers for supported games to reduce loading and compilation times.

 


On the hardware side, Intel said the processors feature 2 performance cores, 8 efficiency cores, and 4 low-power efficiency cores built using the company’s Intel 18A process technology.


Connectivity features include Wi-Fi 7 and Thunderbolt 4


Intel confirmed that the Arc G-Series processors will support integrated Wi-Fi 7, dual Bluetooth 6 connectivity, and Thunderbolt 4. According to the company, Thunderbolt 4 will support transfer speeds of up to 40Gbps for external storage, peripherals, and rapid game library transfers.

 


The company noted that some Wi-Fi 7 features may depend on OEM configurations, operating system support, and compatible routers.


How Intel’s new chips compare with AMD Ryzen Z-series


Intel’s Arc G-Series processors will directly compete with AMD’s Ryzen Z-series processors, which currently power several popular handheld gaming PCs. According to AMD, the Ryzen Z2 lineup focuses on combining Zen CPU architectures with RDNA graphics technologies for handheld gaming devices.

 


AMD’s current Ryzen Z2 Extreme processors feature up to 8 CPU cores, 16 threads, RDNA 3.5 graphics, and configurable TDP ranges between 15W and 35W. AMD also positions its Ryzen Z-series around handheld gaming performance, battery efficiency, and support for technologies like FidelityFX Super Resolution (FSR) and HYPR-RX.

 


While both Intel and AMD are targeting similar handheld gaming workloads, Intel appears to be differentiating its platform through features like XeSS 3 and Intel Precompiled Shaders.



Source link

Spotify updates app with bulk playlist actions, mobile folders, and more

Spotify updates app with bulk playlist actions, mobile folders, and more



Spotify has rolled out a set of new features designed to simplify playlist management and improve the overall listening experience on its platform. The updates bring playlist folders to mobile devices, allowing users to better organise their music libraries, while new bulk editing tools make it easier to manage playlists and queues. Alongside this, Spotify has also introduced background downloads on iOS for more reliable offline listening and a reshuffle button that lets Premium users instantly generate a fresh playback order. Together, these additions are aimed at giving users more control over how they organise content, manage playback, and access music and podcasts on the go. Here is everything that is new:

 
 


Playlist folders are coming to mobile

 


Spotify is bringing playlist folders to mobile devices, allowing users to organise playlists directly from their smartphones. The feature lets users group playlists into folders based on categories such as mood, activity, genre, or any other preference.

 


Until now, playlist folders were primarily a desktop feature, requiring users to switch devices to organise their libraries. According to Spotify, the update is designed to make it easier for users to navigate large music collections and quickly find the right playlist for a specific moment. The feature is available globally for all Spotify users.

 


More control over playlists and queues

 


Spotify is also introducing new bulk editing actions for playlists. Users can now select and manage multiple tracks, audiobooks, or podcast episodes at the same time, making it easier to reorganise content without editing items one by one. The update is intended for users who frequently curate playlists and want a faster way to clean up, rearrange, or update their collections. The feature is rolling out globally.

 

In addition, Spotify Premium subscribers are regaining the ability to select and manage multiple songs in their play queue simultaneously, providing greater control over upcoming tracks. 

 


Offline downloads become more reliable on iPhone

 


Spotify is improving offline listening for iOS users with the introduction of background downloads. Previously, downloads could be interrupted if the app was not actively open. With the new update, music and podcast downloads can continue in the background, allowing content to finish downloading even when users switch apps or lock their devices.

 

 


New reshuffle button for Premium users

 


Spotify is also introducing a dedicated reshuffle button for mobile users on Premium plans. The feature allows listeners to instantly generate a new shuffle order for a playlist without having to turn shuffle off and back on again. Users can repeatedly reshuffle playlists until they find a sequence that better matches their mood or listening preferences.

 


According to Spotify, the feature is designed to help users rediscover familiar playlists and hear their favourite tracks in a different order. The reshuffle button is available globally for Premium subscribers on mobile devices.

 



Source link

Apple may rely on Google Cloud, Nvidia compute to power some AI features

Apple may rely on Google Cloud, Nvidia compute to power some AI features


Apple may rely on Google Cloud and Nvidia infrastructure for some of its upcoming AI features, even as the company continues positioning on-device processing and privacy as key pillars of its Apple Intelligence strategy. According to a report by 9To5Mac, citing The Information, Apple is preparing a new wave of AI features powered by a mix of local processing and cloud-based infrastructure, with some Siri queries running on Gemini models.


Apple using Gemini to train smaller AI models


According to The Information, Apple is using a version of Google’s Gemini large language model to train smaller AI models that can run directly on Apple devices through a process called distillation. The approach would allow Apple to offer AI-powered features locally on iPhones, iPads, and Macs without relying entirely on cloud processing. 

 


The report noted that Apple continues prioritising on-device AI as part of its privacy-focused strategy. However, the company is also reportedly exploring acquisitions to strengthen its local AI capabilities. One of the firms Apple has allegedly considered acquiring is Liquid AI, a startup focused on running AI models efficiently on-device.


Cloud processing still expected for complex AI tasks


Despite its local AI efforts, the report suggests that Apple still faces limitations when handling larger AI workloads internally. According to The Information, the full Gemini model contains trillions of parameters and requires significantly more computing power than Apple’s current Private Cloud Compute infrastructure can efficiently handle. 


As a result, Apple is reportedly expected to use Google Cloud infrastructure alongside Nvidia AI hardware for certain AI-related requests, particularly those involving a newer version of Siri. 


The report claims that some Siri queries may run through Google Cloud using a licensed Gemini model from Google. Apple has also reportedly approved Nvidia’s confidential computing technology for use in order to strengthen security and privacy protections for cloud-based AI processing.


Nvidia’s compute system may help Apple privacy push


According to the report, Nvidia’s confidential computing technology encrypts data and AI models while they are being processed inside graphics processing units (GPUs). Although this reportedly introduces a slight performance slowdown, it could help Apple maintain its privacy commitments while still handling more advanced AI tasks in the cloud.\ 


The Information added that Apple continues exploring ways to balance cloud-based AI processing with stronger privacy protections, particularly as AI features become more computationally demanding.


Private Cloud Compute branding expected to remain


The report also stated that Apple is likely to continue using its “Private Cloud Compute” branding for future Apple Intelligence features, even if some workloads no longer run exclusively on Apple-owned server infrastructure. 


Apple is expected to announce its next set of AI features during WWDC, which begins June 8.



Source link

Why Musk's  billion Cursor bet is really about AI's 'last mile'

Why Musk's $60 billion Cursor bet is really about AI's 'last mile'



Elon Musk’s reported move to secure a $60 billion acquisition option for Cursor is being widely viewed as another eye-popping AI valuation story. But industry experts argue the real battle is not over revenue or even AI models, it is over controlling the interface through which future software is written.

 


At the heart of the reported deal is Anysphere, the company behind Cursor, an AI-native coding environment increasingly used by developers to write, edit, and deploy software with artificial intelligence assistance.

 


The structure itself is unusual. SpaceX reportedly secured a $10 billion partnership arrangement tied to an option to acquire Cursor at a $60 billion valuation later this year. The move has sparked debate over whether the valuation is justified. But experts tracking the AI ecosystem say the bigger question is what Musk is really buying.

 


Why is Musk’s Cursor deal being seen as a battle for AI’s interface layer?


“Most headlines focus on the impressive $60 billion number, but Elon Musk’s interest in Cursor seems to be about something much bigger than just revenue,” said Mr Dinesh Jotwani, Co-Managing Partner at Jotwani Associates.

 


“Cursor is not just another fast-growing AI startup with strong yearly earnings,” Jotwani said, adding, “What Musk may really want is control over a key part of the future AI ecosystem: The developer interface.”

 


The core argument emerging from industry observers is that the AI race is shifting away from raw model development towards ownership of the “last mile”, the place where AI is actually used.

 


“In technology, the biggest winners are often not the companies that build the best infrastructure. Instead, it’s the ones that control how users interact with that infrastructure,” Jotwani said, adding, “Microsoft understood this with Windows, Apple with iOS, and Google with Search. In AI, developer tools could become that same key control layer.”

 


That interface layer matters because developers increasingly interact with AI through coding assistants embedded inside their daily workflow. Whoever controls that environment gains influence over which models are used, how software is built, and where valuable developer data flows.

 


Jyoti Singh, co-founder at Plus91Labs, said the industry is witnessing a broader shift in where value is being created in AI.

 


“As foundational models become more accessible, the real advantage is moving to the layer where AI is actually used, the developer interface,” Singh said. “This ‘last mile’ is where AI turns into real applications, workflows and business outcomes.”

 


She added that ownership of this layer is “not about today’s revenue” but about controlling “how developers build, integrate, and scale AI in the real world.”


What is the missing piece in Musk’s broader AI strategy?


The logic behind the move becomes clearer when placed against Musk’s broader AI ambitions.

 


Through xAI, Musk already competes against OpenAI, Anthropic and Google in frontier AI model development.

 


He also has access to massive compute infrastructure, including the Colossus supercomputer cluster, which has become central to xAI’s aggressive scaling efforts.

 


But experts argue that infrastructure alone no longer guarantees dominance.

 


“The companies that tend to succeed are the ones that control distribution, which is where users actually engage with technology,” Jotwani said.

 


Platforms such as X provide Musk consumer reach and chatbot exposure. Yet analysts say he still lacks deep integration into professional developer workflows, an area increasingly dominated by tools such as GitHub Copilot and Anthropic’s Claude Code.

 


“Musk has two key ingredients, frontier model development through xAI and massive computing power via Colossus,” Jotwani said, adding, “However, history shows that having infrastructure alone rarely ensures victory.”

 


Nikhar Arora, director and builder at BOTS.AI by HR Anexi, described the gap more directly.

 


“Musk already has models,” Arora said, adding, “What he lacked was proximity. Cursor gives him the developer’s seat: the screen that is open when the first decision of the working day is made.”

 


Arora described the deal not as a coding-tool acquisition, but as “an interface acquisition”.


Why are developer tools becoming central to the AI economy?


The growing importance of developer interfaces reflects a wider transition underway in AI.

 


“This is why tools like GitHub Copilot, Cursor, Claude Code, and other AI-native developer environments are so significant,” Jotwani said.

 


He argued that developers are becoming the most important power users in the AI economy because they influence enterprise software adoption, create recurring usage patterns and generate valuable feedback loops for model improvement.

 


“Developers aren’t just using Cursor for productivity; they are forming habits, workflows, and dependencies within its environment,” Jotwani said, adding, “Whoever owns that interface can influence which models are used, where the data goes, and how enterprise software is created.”

 


Arora pointed to evidence that distribution may now matter more than pure model quality. Citing JetBrains’ AI Pulse survey, he noted that GitHub Copilot continues to lead adoption despite rivals posting higher satisfaction scores.

 


“Once an interface embeds itself into a team’s daily workflow, improving the model beneath a rival interface is a slower race to win than it appears,” he said.


Why is the reported deal structure drawing attention?


The reported deal structure has also drawn attention among industry observers.

 


“A $10 billion partnership with a $60 billion acquisition option is a classic real options strategy in corporate finance,” Jotwani said, adding, “It allows an acquirer to secure the right of first refusal and deep technical integration without the immediate integration cost of a full merger.”

 


Jotwani added that the structure effectively reduces risk in a rapidly evolving market where today’s AI architectures could quickly become outdated.

 


“If the integration leads to a 10x increase in developer productivity, the $60 billion price tag — which seems huge now — might look like a bargain in a future where AI-driven software development becomes a multi-trillion-dollar industry,” he said.

 


Arora argued the arrangement is less about immediate ownership and more about preventing rivals from controlling a critical AI workflow layer.

 


“Musk bought an option,” Arora said, adding, “He is not committed to sixty billion. He is committed to ensuring nobody else buys the interface before he decides.”


Valuation concerns and the AI bubble debate


The reported $60 billion valuation has also raised questions over whether the deal reflects long-term strategic value or excessive optimism in the AI market, particularly given Musk’s mixed acquisition track record following X (formerly Twitter).

 


Industry experts say the comparison is relevant, but caution that AI infrastructure assets are being evaluated differently from traditional technology businesses.

 


“Musk has a well-documented pattern of paying a significant premium for control rather than for conventional business value — Twitter being the clearest example,” said Dr Srinivas Padmanabhuni, CTO at AiEnsured. “The logic with Cursor appears similar: it is about controlling the interface through which developers interact with AI.”

 


Padmanabhuni said the valuation remains a “high-stakes” strategic bet rather than a conventional investment case.

 


“If the integration with xAI or the developer ecosystem does not materialise at the scale anticipated, the valuation becomes very difficult to justify on fundamentals alone,” he said. “The question investors should be asking is not whether the vision is coherent, but whether the execution risk is priced in.”

 


Ankush Sabharwal, founder and chief executive officer at CoRover, argued that AI platforms with deep developer adoption cannot be judged purely through short-term revenue metrics.

 


“Valuation in AI today should not be judged only through short-term revenue lenses,” Sabharwal said. “Investors also need to look at long-term strategic value such as developer adoption, ecosystem strength, workflow integration, and the platform’s ability to shape how software is built in the future.”

 


He added that platforms embedded into developer workflows could eventually become foundational infrastructure for future AI applications.

 


The debate has also revived concerns that investor enthusiasm around AI infrastructure and tooling companies may be entering bubble territory.

 


“I would not dismiss this as pure hype,” Padmanabhuni said, adding that AI coding tools have already demonstrated measurable productivity gains and strong adoption among developers.

 


However, he cautioned that the valuation is “clearly pricing in a dominant future outcome, not the current revenue base”.

 


Sabharwal similarly said signs of speculative pricing are emerging across parts of the AI ecosystem.

 


“We are beginning to see signs of a traditional market bubble, as infrastructure and tooling companies around AI are being priced on the basis of future potential rather than current fundamentals,” he said.

 


Still, experts argue the companies most likely to justify such valuations will be those that become indispensable to everyday AI workflows.

 


“The winners will be the companies that provide daily operating layers that are critical to developers and enterprises,” Sabharwal said, “not just AI utility companies with large R&D budgets.”


What does the Cursor deal reveal about the future of AI competition?


For experts, the Cursor story ultimately signals a larger transformation unfolding across the AI industry.

 


“Yes. The AI market is shifting from just competing on models to focusing on distribution, workflow integration, and user habits,” Jotwani said, adding, “As the quality of models becomes more similar among major players, the main advantage lies in embedding AI into everyday tasks.”

 


That shift could reshape how power is distributed across the AI economy.

 


“If Cursor becomes the Windows of AI coding, its influence would be unmatched,” Jotwani said, adding, “The AI-IDE is the gateway to the digital economy.”

 


For Musk, the strategic logic may therefore extend well beyond buying a fast-growing AI company.

 


As Singh put it, “Cursor is not just a product, it’s a strategic gateway to the future of software development.”



Source link

Samsung says Galaxy Watch can predict fainting episodes before they happen

Samsung says Galaxy Watch can predict fainting episodes before they happen


Samsung has announced research suggesting that its Galaxy Watch could help predict fainting episodes before they happen. According to the company, a joint clinical study conducted with Chung-Ang University Gwangmyeong Hospital in South Korea found that the Galaxy Watch6 was able to detect warning signs linked to vasovagal syncope (VVS), a common condition that causes sudden fainting.

 


Samsung said the smartwatch used heart rate variability data and AI analysis to identify possible fainting episodes several minutes in advance, potentially giving users enough time to sit down, seek help, or move to a safer position before losing consciousness.


Samsung predicts fainting: How it works


According to Samsung, vasovagal syncope occurs when a person’s heart rate and blood pressure suddenly drop, often triggered by stress, pain, exhaustion, or standing for long periods. While fainting itself is usually temporary, sudden falls can lead to serious injuries such as fractures or head trauma.

 
 


Samsung said researchers tested 132 patients with suspected VVS symptoms during medically induced fainting tests. Using the Galaxy Watch6’s photoplethysmography (PPG) sensor, researchers monitored heart rate variability (HRV) data and analysed it using an AI-based algorithm.

 


The study reportedly found that the system could predict a possible fainting episode up to five minutes in advance with an accuracy of 84.6 per cent.

 


Junhwan Cho, professor in the Department of Cardiology at Chung-Ang University Gwangmyeong Hospital, said early warning signs could help people move into a safer position or call for assistance before losing consciousness.


Other companies


Wearable brands have increasingly been adding health and safety-focused features in recent years, though most existing systems focus on detecting emergencies after they happen rather than predicting them beforehand.

 


Apple Watch models already include fall detection, irregular heart rhythm notifications, ECG support, and crash detection.

 


Google’s Pixel Watch lineup also offers irregular heart rhythm alerts, emergency SOS, and pulse-related health tracking features. Google’s Pixel Watches include a “Loss of Pulse Detection” feature that can detect when a user may have experienced cardiac arrest or sudden loss of pulse and automatically contact emergency services if the user is unresponsive. Google described it as a first-of-its-kind feature after receiving FDA clearance earlier this year.

 


Samsung said the research reflects a broader shift from reactive healthcare towards preventive care using wearable devices and AI-based monitoring systems.


The company added that it plans to continue expanding health monitoring capabilities across its wearable portfolio through collaborations with medical institutions and further research in digital health technologies.


Samsung Galaxy Watch6


Samsung launched the Galaxy Watch6 in India in 2023 as part of its premium smartwatch lineup focused on health tracking, fitness, and Wear OS features.

 


The Galaxy Watch6 series runs on Wear OS with Samsung’s One UI Watch interface and includes features such as heart rate monitoring, ECG, blood pressure tracking, sleep coaching, body composition analysis, fall detection, and irregular heart rhythm notifications in supported regions.

 


The smartwatch is powered by Samsung’s Exynos W930 chipset and features AMOLED displays with sapphire crystal protection.

 



Source link

YouTube
Instagram
WhatsApp