Scitke

Tag: AI-driven

  • Jeff Bezos Is Launching a Multi-Billion-Dollar Company In AI-Driven Manufacturing

    Jeff Bezos Is Launching a Multi-Billion-Dollar Company In AI-Driven Manufacturing

    Jeff Bezos is back in the CEO seat after stepping down from Amazon in 2021. According to Cade Metz of The New York Times, the billionaire founder of Blue Origin is now co-CEO of Project Prometheus, a startup focused on creating “AI for the physical economy.”
    Image Credits: Bezos’ new startup is said to be working on AI for manufacturing technical products and vehicles
    Image created with Google AI

    Jeff Bezos is back in the CEO seat after stepping down from Amazon in 2021. According to Cade Metz of The New York Times, the billionaire founder of Blue Origin is now co-CEO of Project Prometheus, a startup focused on creating “AI for the physical economy.”

    Project Prometheus Targets AI in Engineering and Manufacturing

    Anonymous sources say Project Prometheus is developing AI to assist in engineering and manufacturing across sectors such as computers, aerospace, and automobiles. The company has already raised $6.2 billion and recruited over 100 employees from Meta, OpenAI, and DeepMind. Several founding members’ LinkedIn profiles also highlight prior experience at Microsoft, DeepMind, and Nvidia, with expertise in AI and robotics research.

    Details on Project Prometheus’ specific products or location remain unclear. Bezos shares CEO duties with physicist Vik Bajaj, who has an impressive history working on ambitious ventures, including Google X—a “moonshot factory” behind projects like Taara’s light-beamed internet, Waymo robotaxis, Wing drone delivery, and exoskeleton-equipped pants.

    Bajaj also co-founded the life sciences research company Verily and led Foresite Labs, an incubator for AI startups.

    Image Credits: Project Prometheus co-CEO Vik Bajaj has been involved in a number of moonshot projects with Google over the years
    Vik Bajaj / LinkedIn (upscaled using AI)

    Project Prometheus Explores Advanced Robotics and Autonomous AI

    There are additional hints about Project Prometheus’ potential goals. Last year, Bezos took part in a $400 million funding round for Physical Intelligence, a San Francisco startup that uses AI to improve robots’ abilities—tasks like making espresso without spilling or assembling boxes.

    Metz also reports that the company plans to develop robots capable of conducting scientific experiments autonomously and at scale. While some speculate this could eventually support Bezos’ broader space ambitions, the connection to Project Prometheus remains unclear. More immediately, it appears likely that the startup’s technology could enhance humanoid robots in the workplace—or even assist in building them.

    How AI and Robotics Are Revolutionizing Manufacturing

    AI and robotics are increasingly finding applications in manufacturing. Agility Robotics says its Digit robot becomes the first humanoid to operate commercially on factory floors, while Texas-based Apptronik works to equip its Apollo humanoid with the ability to build additional units of itself. AI also optimizes chemical processes for ammonia production, creates stronger nanomaterials, and improves processor designs to boost artificial intelligence capabilities.

    Bezos, who has played a major role in Blue Origin’s work on suborbital tourism, orbital launch systems, and space infrastructure, now takes on a new challenge with Project Prometheus. Unlike his previous ventures, this startup is still in its early stages, offering him a fresh arena to build and innovate.

    Read the original article on: New Atlas

    Read more: Outage at Cloudflare Disrupts ChatGPT, X, and Other Web Services

  • Adobe Introduces new AI-driven Image Editing Tools in Photoshop

    Adobe Introduces new AI-driven Image Editing Tools in Photoshop

    Adobe is expanding Photoshop’s suite of AI-powered image editing tools with new enhancements powered by its Firefly models. The company announced the rollout of features including image upscaling, object blending, and an improved object removal tool.
    Image Credits: Techcrunch

    Adobe is expanding Photoshop’s suite of AI-powered image editing tools with new enhancements powered by its Firefly models. The company announced the rollout of features including image upscaling, object blending, and an improved object removal tool.

    One of the highlights, Generative Upscale, can enhance images up to 8 megapixels—ideal for restoring old photos, preparing high-quality prints, or adapting visuals for different platforms.

    Adobe Debuts Harmonize for Seamless Object Integration

    Previously previewed in 2024 as Project Perfect Blend, Adobe has now introduced this feature as Harmonize. It allows users to insert an object into a scene, remove its background, and have Photoshop automatically adjust its color, lighting, shadows, and tones to match the surrounding environment using Firefly’s AI capabilities.

    Additionally, Adobe has refined its object removal tool. Earlier versions using generative fill sometimes created unwanted artifacts by analyzing the entire scene. The updated algorithm now focuses specifically on removing the object and seamlessly filling in the background for cleaner, more accurate edits.

    Improved Object Removal Delivers Cleaner, More Accurate Results

    In a demonstration viewed by TechCrunch, Adobe’s Director of Product Management for Photoshop, Joel Baer, attempted to remove a small bowl of pico de gallo using an earlier version of the object removal tool. Instead of cleanly erasing the item, the tool left behind a green smudge. The updated version, however, successfully removed the object and filled in the area with a seamless background.

    Adobe is also introducing a few additional enhancements to Photoshop, such as improved asset management tools for organizing and sharing files. Users can now also select which version of the Firefly model they want to use when working with AI-powered features.

    “We’ve spoken with photographers, designers, and content creators to understand the bottlenecks and pain points in their workflows,” said Shambhavi Kadam, Senior Director of Product Management for Photoshop, during a briefing. “These new capabilities are designed to streamline their processes, reduce friction, and handle some of the more time-consuming tasks.”

    The Generative Upscale, Harmonize, and upgraded object removal tools are currently available in beta for both desktop and web versions of Photoshop. The Harmonize feature is also accessible through early access on the new Photoshop iOS app.

    Read the original article on: Techcrunch

    Read more:Amazon Acquires Bee, the AI Wearable that Records Conversations

  • Apple’s iOS 19 is Expected To Offer AI-Driven Battery Improvements

    Apple’s iOS 19 is Expected To Offer AI-Driven Battery Improvements

    With just about a month remaining until Apple's highly anticipated WWDC 2025 event, the rumor mill is already buzzing with speculation about the next major update for iOS. The most recent leak comes from Bloomberg's trusted reporter Mark Gurman, who reveals that Apple is working on an innovative, AI-powered battery management feature for iOS 19.
    Credit: imageBROKER/Valentin Wolf via Getty Images

    With just about a month remaining until Apple‘s highly anticipated WWDC 2025 event, the rumor mill is already buzzing with speculation about the next major update for iOS. The most recent leak comes from Bloomberg’s trusted reporter Mark Gurman, who reveals that Apple is working on an innovative, AI-powered battery management feature for iOS 19.

    This new technology would harness artificial intelligence to optimize battery life by dynamically adjusting power usage based on users’ habits and app preferences. If true, this could mark a significant step in Apple’s ongoing efforts to enhance the user experience through smarter, more efficient device performance. As the WWDC event draws closer, expectations are building around what else iOS 19 might bring to the table, and how these features could shape the future of the iPhone.

    AI-Powered Battery Management to Optimize iPhone Power Usage

    According to Gurman’s sources, Apple’s AI-powered battery management system will analyze iPhone usage patterns, including app activity and environmental factors, to optimize energy consumption. By adjusting power draw based on app usage, it will conserve battery while keeping essential apps running efficiently. This could greatly extend battery life, especially for power users.

    Additionally, iOS 19 will introduce a new lock screen battery indicator that shows the estimated time until a full charge, helping users better plan their charging. This feature aims to reduce battery anxiety and improve device management, highlighting Apple’s focus on smarter, more efficient energy use.

    A more tailored approach to battery life boosts speculation that Apple may reveal an iPhone Air this year. Similar to its laptop version, a thinner design would mean a smaller battery, so finding ways to maximize performance will be crucial to appealing to potential buyers.

    This new battery management feature will be part of Apple’s Apple Intelligence suite, which has been slowly rolled out since last fall, receiving mixed feedback. We likely won’t know until WWDC whether iOS 19 will include the features we’re most excited about.

    Read the original article on: Engadget

    Read more: Apple’s New Feature May Soon Let You Control Your IPhone With Your Mind

  • All Trails Rolls Out A New $80-Per-Year Membership That Includes AI-Driven Smart Route Capabilities

    All Trails Rolls Out A New $80-Per-Year Membership That Includes AI-Driven Smart Route Capabilities

    AllTrails, the popular hiking and biking app crowned  2023’s iPhone App of the Year, is launching a new premium tier called “Peak.” Priced at $80 annually, the subscription adds advanced features like AI-generated custom routes, real-time trail condition updates, trail traffic heatmaps, and a tool to identify trees and plants using your phone’s camera.
    Credit: Pixabay

    AllTrails, the popular hiking and biking app crowned  2023’s iPhone App of the Year, is launching a new premium tier called “Peak.” Priced at $80 annually, the subscription adds advanced features like AI-generated custom routes, real-time trail condition updates, trail traffic heatmaps, and a tool to identify trees and plants using your phone’s camera.

    AllTrails to Roll Out New Subscription Globally by June, Early Access Now Open

    The new subscription will first launch over the next few weeks to a limited group of users, with global availability expected by June. Those interested can register for early access on the AllTrails website.

    A longtime favorite among outdoor lovers, AllTrails round of $150 million in growth funding in 2021, during a pandemic-driven surge in interest for nature and outdoor activities.

    Unfortunately for AllTrails, Apple entered its market just before the app was named App of the Year,  adding support for topographic maps with trailheads, contour lines, hill shading, elevation gains, and more to Apple Maps in the U.S. The following year, Apple expanded further with features like offline maps, custom routes, and tools for exploring national parks and trails beyond the U.S.

    In response, AllTrails’ latest update appears to be a strategic move to appeal to its most dedicated and active users, offering a set of features designed to strengthen its position.

    Credit: AllTrails

    AI Smart Routing Customizes Trails by Distance, Difficulty & Scenery

    The AI-driven feature in the Peak membership, for example, allows users to create their perfect trail, either by designing one from the ground up or modifying an existing route. The smart-routing function leverages AI to adjust the trail, making it shorter, less steep, or even more scenic, among other options.

    The new trail conditions forecast will provide details on temperature, precipitation, snow depth, and more, essentially integrating a weather app within AllTrails. Users will also receive alerts about air quality and mosquito activity. Future updates will include terrain and ground conditions, such as pavement, gravel, or mud.

    Credit: AllTrails

    New Heatmap Feature Helps Users Choose Trails Based on Crowd Levels and Safety Preferences

    The new heatmap feature is designed to assist hikers and bikers who want to avoid crowded trails, as well as those who feel safer with more people around, such as when hiking alone. By leveraging data from millions of users’ activities, the app can estimate trail traffic, much like how Google Maps predicts peak times for stores, restaurants, bars, and other locations.

    Credit: AllTrails

    This summer, Peak members will be able to use the Outdoor Lens feature, which allows them to quickly identify trees, plants, or flowers and record them in the app. The company also mentions that additional identification features will be added to the Lens over time.

    Credit: AllTrails

    Alongside the launch of Peak, AllTrails is also updating its Base and Plus membership tiers. Base, the free plan, includes trail maps, community reviews, and navigation.

    Plus offers additional perks like offline access, wrong-turn alerts, 3D trail previews, and live trail activity sharing. Altogether, AllTrails serves a community of over 80 million registered users across all plans.

    Both Base and Plus members will now have access to a new feature called Points of Interest, which adds over 500,000 mapped landmarks, viewpoints, and trail amenities. Plus members will also get a new Offline Area tool, allowing them to download custom regions for offline use with high-resolution maps and detailed trail data.

    Read the original article on: TechCrunch

    Read more: Amazon Reveals a Glimpse of Future Human Roles in an AI-Driven Workplace

  • Dropbox Introduces New Capabilities To Dash, Its AI-Driven Search Tool

    Dropbox Introduces New Capabilities To Dash, Its AI-Driven Search Tool

    Credit: Depositphotos

    Tech giants like  Google  and  Microsofthave integrated AI tools and assistants into their productivity platforms, while startups like ClickUp  and ReadAI have concentrated on developing robust AI integrations and search functions.

    Following this broader movement to enhance digital workspaces with AI, Dropbox on Thursday announced enhancements to Dash, its AI-powered search tool originally launched in 2023.

    Dropbox is enhancing Dash with AI that can comprehend various content types, allowing users to search through audio, video, and images—not just text. Additionally, it’s introducing a people search feature, enabling users to find individuals involved in specific projects or identify subject-matter experts.

    Last year, Dropbox launched Dash for Business, enabling enterprises to leverage AI-powered search. This year, it’s enhancing Dash’s enterprise features by allowing IT administrators to exclude certain sensitive documents from appearing in search results.

    Credit: Dropbox

    Dropbox Expands AI Capabilities with New Writing Tools in Dash

    Dropbox already offers AI-powered document summarization. With this latest update, the company is rolling out new writing tools within Dash that use summaries from various data sources to generate fresh documents and presentations.

    According to Dropbox, these tools can gather content from emails, meeting notes, and existing files to create items like project plans, memos, or briefs. The main goal is to streamline the workflow, eliminating the need to switch between multiple apps to collect and incorporate information into a document.

    Additionally, Dropbox is expanding Dash with new integrations, including communication tools such as Slack, Zoom, and Microsoft Teams, as well as project management and creative platforms like Figma, Canva, and Jira. According to Dropbox, these additions will make it easier for users to search for information across various platforms and projects.

    As AI providers release more advanced models, companies focused on productivity and workforce tools are finding it more practical to leverage AI for tasks like analyzing vast amounts of data, generating summaries, and producing fresh content. The key challenge they now face is quickly building new features and integrating smoothly with other platforms to keep users happy.

    Read the original article on: TechCrunch

    Read more: Find Out Why Having a Small Hard Drive is No Longer an Issue

  • Microsoft Integrates AI-Driven Deep Research Tools Into Copilot

    Microsoft Integrates AI-Driven Deep Research Tools Into Copilot

    Microsoft is rolling out a “deep research” AI tool within Microsoft 365 Copilot, its AI-powered chatbot app. Recently, several deep research agents have emerged across AI chatbots, including OpenAI’s ChatGPT, Google’s Gemini, and xAI’s Grok. These tools rely on reasoning AI models, which can analyze problems and verify facts—key abilities for conducting in-depth research.
    Image Credits:Rafael Henrique/SOPA Images/LightRocket / Getty Images

    Microsoft is rolling out a “deep research” AI tool within Microsoft 365 Copilot, its AI-powered chatbot app. Recently, several deep research agents have emerged across AI chatbots, including OpenAI’s ChatGPT, Google’s Gemini, and xAI’s Grok. These tools rely on reasoning AI models, which can analyze problems and verify facts—key abilities for conducting in-depth research.

    Microsoft Introduces Researcher and Analyst for Advanced AI-Powered Research

    Microsoft’s versions, Researcher and Analyst, bring unique capabilities. Researcher integrates OpenAI’s deep research model—also used in ChatGPT’s research tool—with “advanced orchestration” and “deep search” features. According to Microsoft, it can handle tasks like developing go-to-market strategies and generating quarterly client reports.

    Analyst, built on OpenAI’s o3-mini reasoning model, is designed for advanced data analysis, Microsoft said. It works iteratively, refining its approach step by step to deliver precise answers. Analyst can also execute Python code to handle complex data queries and provides transparency by making its process visible for review.

    Microsoft’s Deep Research Tools Stand Out with Access to Internal and Web Data

    What sets Microsoft’s deep research tools apart is their ability to access both internal work data and the web. For instance, Researcher can pull information from third-party data connectors, integrating insights from AI agents, tools, and apps like Confluence, ServiceNow, and Salesforce.

    However, the real challenge remains preventing hallucinations or misinformation. Models like o3-mini and deep research are not infallible—they can misattribute sources, draw incorrect conclusions, or rely on unreliable public data.

    Microsoft is introducing the Frontier program, allowing Microsoft 365 Copilot users to access Researcher and Analyst. Participants in Frontier, which will receive early access to experimental Copilot features, will be able to use Researcher and Analyst starting in April.

    Read the original article on: TechCrunch

    Read more: Microsoft is Investigating a Method to Recognize and Credit Contributors to AI Training Data

  • Flora is Developing an AI-driven Infinite Canvas Designed for Creative Professionals

    Flora is Developing an AI-driven Infinite Canvas Designed for Creative Professionals

    With just a few words, AI models can generate stories, images, or even short films. However, Weber Wong argues that these tools are designed by non-creatives for other non-creatives to feel creative—meaning they don’t truly serve professional artists.
    Image Credits: Flora

    With just a few words, AI models can generate stories, images, or even short films. However, Weber Wong argues that these tools are designed by non-creatives for other non-creatives to feel creative—meaning they don’t truly serve professional artists.

    Wong aims to change that with Flora, a startup he founded and now leads as CEO.

    A Bold Vision for AI and Creativity

    Launched this week, Flora introduced itself with a manifesto rejecting AI tools that produce low-quality, automated content. Instead, Wong and his team are committed to developing a powerful tool that will redefine creative work.

    The manifesto sets Flora apart from existing AI tools, which simplify creation but lack artistic control, and from traditional creative software, which offers control but is often complex and time-consuming.

    Image Credits:Flora /

    Redefining Creativity with an Infinite Canvas

    Flora isn’t focused on developing more advanced generative AI models. Wong believes one of the startup’s key insights is that “models are not creative tools.” Instead, Flora provides an “infinite canvas,” a visual workspace that integrates with existing models, allowing users to generate text, images, and video.

    “The model does not matter, the technology does not matter,” Wong explained. “It’s about the interface.”

    For instance, a user might prompt Flora to generate an image of a flower, then request additional details about it. Those details could lead to new prompts and variations, all systematically mapped on the canvas. This flexible workspace can also be shared for collaborative projects with clients.

    Wong envisions Flora as a tool for all artists and creatives, but the company is initially focusing on visual design agencies. To refine the product, Flora is gathering feedback from designers at the renowned agency Pentagram.

    The goal, Wong explained, is to help a designer at Pentagram “produce 100 times more creative work.” For example, a designer could create a logo and instantly generate 100 variations. He likened this shift to the evolution of music composition—where Mozart once required a full orchestra to bring his music to life, today’s musicians can compose and produce entire tracks from a home studio using software like Ableton and share them instantly on SoundCloud.

    From Investor to Innovator

    With a background in both art and technology, Wong previously worked as an investor at Menlo Ventures but left after realizing, “I was not the person I’d back.” Determined to become the kind of founder he’d invest in, he joined NYU’s Interactive Telecommunications Program, a graduate program exploring the intersection of art and technology.

    When Flora launched its alpha version in August, Wong introduced it with an art project that highlighted the platform’s real-time AI capabilities. The Flora homepage featured a live feed from a GoPro mounted on Wong’s head, allowing website visitors to sign up for the waitlist and use AI to stylize the footage.

    Image Credits: Flora /

    Bridging the Gap Between AI and Artists

    With his background, Wong understands that many artists and professionals are skeptical—or even strongly opposed—to AI in art. In fact, Pentagram faced backlash last year for using Midjourney to develop an illustration style for a U.S. government project.

    Wong acknowledged that current AI models have been embraced by “AI natives,” but his goal with Flora is to attract the “AI curious” and, over time, make the tool so valuable that even “AI haters” feel compelled to try it.

    When I brought up concerns about AI models being trained without regard for copyright and intellectual property, Wong pointed out that Flora doesn’t train its own models, as it relies on existing ones. He added, “We will follow societal standards.”

    Passionate about preventing Flora from contributing to the flood of low-quality AI-generated content (“We’re even getting hats that say ‘anti-AI slop’”), Wong emphasized that the platform is designed to help artists explore “new aesthetic and creative possibilities”—similar to how Kodak’s Brownie camera revolutionized photography by making it more accessible and spontaneous.

    Flora has not disclosed its funding details, but its investors include A16Z Games Speedrun, Menlo Ventures, and Long Journey Ventures, along with angel investors from Midjourney, Stability, and Pika. The platform is free for a limited number of projects, with monthly professional plans starting at $16.

    Read the original article on: TechCrunch

    Read more: Apple iPhone 16e Review: A18 Chip and Apple Intelligence for $599