Highlights

• Alibaba’s Qwen models are nearing 1 billion downloads, dominating open-source AI globally
• Qwen holds over 50% of total downloads, far ahead of global competitors
• Growth is driven by aggressive pricing, strong performance, and an open-source strategy

Alibaba’s Qwen AI models are rapidly emerging as the global leader in open-source artificial intelligence. As of March 2026, the Qwen model family has reached around 942 million total downloads, with a massive 153.6 million downloads recorded in February alone. This puts Qwen significantly ahead of competitors, with more than double the combined downloads of the next eight players.

A major boost came from the launch of Qwen 3.5 in February 2026. The latest version delivers major improvements in both speed and efficiency. It is reported to be up to eight times faster and around 60% cheaper than its predecessor, making it highly attractive for developers and businesses.

Qwen’s rise also reflects a broader shift in the global AI landscape. Since mid-2024, Chinese open-source models have overtaken their US counterparts in total downloads. Earlier releases like Qwen 2.5 played a key role in this transition by accelerating adoption worldwide.

The company’s strategy focuses on offering powerful yet low-cost models to drive mass adoption. This approach helps build a large developer ecosystem, which is later monetized through cloud services. As a result, Qwen is not just gaining popularity but also strengthening its position in the global AI race.

Competitors: Still Far Behind Qwen

Qwen’s closest rival, Meta’s Llama, is now significantly behind, with estimates suggesting a gap of hundreds of millions of downloads. Fast-growing players like DeepSeek have seen sharp spikes in popularity, but still lag far in total usage. Meanwhile, other models from Nvidia and smaller AI firms show steady but limited adoption. Overall, no competitor currently matches Qwen’s scale, making its lead both clear and rapidly widening.

(via)

The post Alibaba’s Qwen Dominates Open-Source AI: Nears 1 Billion Downloads appeared first on Gizmochina.

Amazon has launched Alexa+, a generative AI assistant that works with over 600 million existing Alexa devices. The system holds conversations naturally and completes tasks independently.

Alexa+ functions through “experts” – systems that perform specific operations. These enable Alexa+ to control smart home devices, make reservations, play music across services, order groceries, and monitor home security.

The assistant can navigate the internet autonomously to complete tasks. It can locate service providers, make arrangements, and report back without user intervention.

Alexa+ learns user preferences from past purchases and allows users to add information like dietary restrictions or recipes. It applies this knowledge when making suggestions.

The service works across Echo devices, a new mobile app, and Alexa.com. Conversations transfer between devices while maintaining context.

Users can upload documents, emails, photos, and messages for Alexa+ to summarize or use. Amazon built the system with privacy protections, with settings managed in the Alexa Privacy dashboard.

Alexa+ costs $19.99 monthly, but is free for Prime members alongside existing benefits like delivery and entertainment services. The US rollout begins with early access for Echo Show owners, followed by a phased expansion. Users can request early access at amazon.com/newalexa.

In other news, Amazon is shutting down its Appstore for Android devices on August 20, 2025, with new app submissions already halted. Users will lose access after the closure, leaving Google Play as the default marketplace. The Appstore will remain available on Fire TV and Fire Tablets, where Amazon sees the most engagement.

(Source)

The post Amazon Unveils Alexa+ Generative AI: Free for Prime Members appeared first on Gizmochina.

Oppo is reportedly planning to push its AI capabilities to 50 million devices by the end of 2024. It’s also revealed that India is regarded as one of the most important markets when it comes to the company’s latest AI rollout plan.

The company is expected to offer over 100 AI features to its products this year

The company is expected to launch more than 100 AI features this year and become a leader in the platform shift driven by the new technology. Peter Dohyung, the head of product strategy at Oppo, has already revealed the company’s roadmap for the integration of AI in its devices.

Notably, one of the key steps in the roadmap is to push generative AI features to the products. A company official said in an interview that OPPO hopes that all phones will have GenAI capabilities in the near future. He also revealed that the company aims to start with its AI strategy by rolling out generative AI features to 50 million devices by 2024.

Reportedly, the camera system will benefit from the integration of this new technology. How? Well, a company official said the company will “continue to integrate AI to enhance the imaging experience.” However, we don’t know much about how exactly the addition is going to enrich the camera system. There is also a summary feature that utilizes generative AI.

It is also claimed that the AI features will be rolled out with strict data privacy measures. This is especially important because we have already seen server-side implementations of similar AI features from multiple brands, where user data leaves their device for more advanced processing.

That being said, Oppo is committed to uplifting the user experience with the use of generative AI. The company is also making a significant contribution to the research and development of the technology. The company has been focusing on improving its data centers and building Large Language Models that power the AI features for the past couple of years.

(Via)

The post Oppo plans to push AI to 50 million devices by the end of 2024 appeared first on Gizmochina.

Getty Images and Picsart, a popular photo editing platform, have joined forces to create a novel AI image generator. This unique tool aims to address copyright concerns plaguing current AI image generation models.

The training data will comprise exclusively of licensed stock images

The key difference lies in the training data. Unlike other models that might use a mix of sources, the Picsart and Getty generator will be trained exclusively on licensed stock images from Getty Images. This ensures all the content used to create the AI model is authorized for commercial use. The companies are touting this approach as a “responsible, commercially-safe” alternative. Images generated by this model will come with full commercial rights, allowing users to freely incorporate them into their projects without copyright worries.

Similar to Adobe’s recent attempt with their Firefly AI model, the Picsart and Getty tool will be a paid service accessible only to Picsart’s subscribing users. This approach avoids the controversy surrounding Adobe’s plan to potentially use user-generated content for training their AI model.

Getty seems committed to ethical practices. They’ve assured users that photographers whose work contributes to the training data will be “compensated creators included in the dataset on an annual basis.”

While a specific launch date remains unknown, the Picsart and Getty Images AI image generator is expected to be released later this year. It will be accessible through Picsart’s API services, allowing integration with various creative workflows.

(Via)

The post Getty and Picsart Team Up for “Commercially Safe” AI Image Generation appeared first on Gizmochina.

If you are struggling to practice spoken English without face-to-face interactions, Google Search Labs might have the solution for you. As per the source tweet, their new experimental feature, Speaking Practice, utilizes generative AI to help you improve your English speaking skills through conversation practice.

Users will first have to enable Search Labs within their Google Search app settings

To access Speaking Practice, users will first need to enable Search Labs within their Google Search app settings. Once enabled, you can find Speaking Practice by clicking “Try an example” within the feature itself. Alternatively, you can access it on your Android device by translating to or from English using Google Search.

Currently, Speaking Practice is being tested in Argentina, Colombia, India, Indonesia, Mexico, and Venezuela. Stay tuned for a possible expansion to your region! Google mentioned thatIt’s important to remember that generative AI is still under development, so the quality and availability of Speaking Practice may vary.

RELATED:

• Google Pixel 8a Leaked Official Promo Video Reveals Its AI Features
• Could Google Launch Pixel Watch 2a? FCC Listing Hints At A Mysterious Smartwatch
• Lenovo Legion Y700 2024: Latest Gaming Tablet with Enhanced Display now available at Giztop
• Xiaomi 13 Ultra Premium Camera Phone is now only $799
• 10 Best Free Android Games in 2024

(Via)

The post Google Might Become Your New Spoken English Teacher, The Experimental Feature is Called “Speaking Practice” appeared first on Gizmochina.

Generative AI can be used for a lot of things, but it’s hard to say exactly where the upper limits are. As of now, it seems to be vehicles. Recent news state that Japanese car companies are embracing generative AI to streamline the development process for new models. Utilizing the capability of generative AI to create new variations based on existing data, companies are expecting to significantly improve efficiency and potentially cut the planning and design time by 50%.

Honda is also leveraging Generative AI through its joint venture with Sony

Toyota is leading the charge by utilizing image-generating AI at its US research arm. This AI analyzes car design drawings and calculates air resistance, allowing for adjustments that optimize both design and functionality. Honda is also on board, using generative AI through its joint venture with Sony to develop advanced driver-assistance systems (ADAS) for their upcoming AFEELA car.

Subaru and Mazda are taking a slightly comprehensive approach, focusing on training their workforce in AI principles. Subaru has launched a program to equip over 4,000 engineers with the skills needed to utilize generative AI, while Mazda is making AI training mandatory for all indirect employees in design and experimental departments by 2025. Mazda’s ultimate goal is to double overall production efficiency by 2030. That’s a big goal, but a realistic timeframe and rapidly developing AI technology can propel the company to its forecasted numbers. We can just wait and see what happens.

As car companies race to develop electric vehicles and autonomous driving features, one thing is clear – generative AI will be undoubtedly a powerful tool to speed up innovation and potentially reshape transportation. It will be interesting to see if AI also plays a role in churning out special car designs or functionalities that we haven’t even imagined yet. Imagine being able to say, “My car was designed by Artificial Intelligence.” It could be a flex, or not. That’s up to you!

RELATED:

• Nvidia Unveils Eos Supercomputer to Push the Boundaries of Artificial Intelligence
• China is Suffering from Lack of AI Experts Amidst Growing Artificial Intelligence Industries
• Get the Realme GT5 Pro phone on Giztop for $599
• Get $100 OFF on Xiaomi 14 Pro at Giztop (1TB Variant)
• How to hide WhatsApp chats without archive

(Via)

The post Using Generative AI to Design New Cars? Toyota, Honda, and Subaru are Already on it appeared first on Gizmochina.

Adobe has unveiled a game-changing tool called Project Music GenAI Control. This experimental AI system allows you to not only generate music but also edit it directly within the same interface, all starting with simple text descriptions. Imagine needing an epic electronic intro – simply type “epic electronic” and let the AI work its magic. Yes, generative AI is becoming mainstream for real.

You can also edit your existing music files using the AI

But Project Music GenAI Control goes beyond just generating music. It allows you to refine your masterpiece by adjusting tempo, structure, repetition, and intensity to perfectly match your vision. You can even expand your canvas by extending the length, remixing sections, or creating seamless loops, all within the same tool. And if you have an existing piece of audio, you can even transform it based on a reference melody, unlocking even more creative possibilities.

Unlike other AI music tools, Project Music GenAI Control empowers you with precise editing capabilities, offering you a level of control akin to working with pixels in Photoshop, but applied to the world of sound. This seamless workflow eliminates the need to jump between different applications, streamlining your creative process.

Adobe emphasizes responsible AI practices by ensuring the generated music is safe for commercial use. They also implement Content Credentials, acting like “nutrition labels” that disclose the AI-assisted nature of your creation.

Developed in collaboration with esteemed universities, Project Music GenAI Control is still in its experimental stage. While a user-friendly interface is still under development, this glimpse into the future of audio creation suggests a powerful and intuitive tool is on the horizon.

RELATED:

• Figma informs employees that the deal with Adobe to buy the UX design outfit has collapsed
• Unwrap the Magic of Christmas: Exclusive Deals on Microsoft and Adobe Software at Msckey!
• Big Discount: AOOSTAR R1 N100 NAS Mini PC Only For $159
• Lenovo Legion Y700 2023: Save $100 on this 8-inch gaming Android tablet
• Xiaomi Band 8 Genshin Impact custom edion get a huge discount.
• Best of MWC 2024: AI Phone, Transparent Laptop, 3D Tablet & More

(Via)

The post Adobe Unveiled a “Text-to-Music” Tool for AI-powered Music Generation appeared first on Gizmochina.

Microsoft is stepping up its game in the photo editing arena, introducing an innovative feature called Generative erase to Windows PCs. This nifty tool, built into the Windows Photos app, offers a simple way to clean up your photos by removing things you’d rather not have in them—think of a person who photobombed your perfect shot, or a helicopter that showed up out of nowhere in your perfectly framed picture. Yeah, Generative erase will fix that for you.

Although not as good as Magic Eraser, the tool does the job quite well

Microsoft is making sure of that by rolling out Generative erase not just to those on Windows 11 but also to folks using Windows 10 and Windows 11 for Arm64 devices. This move is all about bringing more powerful editing capabilities to a wider audience, right on their existing hardware.

For the moment, this feature is in the hands of Windows Insiders, a group of users who get to try out new stuff before it’s widely released. They’re the lucky ones testing out Generative erase along with other AI-driven editing features, provided they’ve updated their app to the latest version specified by Microsoft. There’s also a background removal tool in the mix, adding even more options for users to play with.

As we move forward, integrating AI into our daily tech tools is becoming a norm. Microsoft’s addition of Generative erase to its Photos app is a big step toward making sophisticated photo editing tools more user-friendly and widely available. It’s a clear signal that the future of photo editing is here, and it’s more inclusive, allowing more people to enhance their digital memories with just a few clicks.

RELATED:

• Google’s Latest Exciting Pixel Feature Drop, All Pixel Devices Get Magic Eraser
• Google Maps Adds Generative AI for Personalized Place Discovery
• Big Discount: AOOSTAR R1 N100 NAS Mini PC Only For $159
• How to add/remove someone in a WhatsApp group (with screenshots)

(Via)

The post Generative Erase is Microsoft’s Version of Magic Eraser on Windows 10 & 11 appeared first on Gizmochina.

Google‘s latest Chrome update introduces a groundbreaking feature aimed at making digital writing a breeze for its users. Dubbed “Help me write,” this AI-powered tool is more than just a fancy add-on; it’s a looking-glass into the future, how artificial intelligence can enhance our daily tasks, specifically writing, by understanding the context of our needs and offering customized suggestions.

Users now don’t have to ponder or think too much while compiling something

Think about the countless times you’ve struggled to articulate your thoughts in an email or wanted to sell an item online but couldn’t quite capture its essence in words. Google’s new feature aims to eliminate such hurdles. For instance, if you’re trying to sell a barely used gadget because you’re upgrading, “Help me write” could craft a compelling narrative that highlights the gadget’s features, its condition, and why it’s a steal at your asking price.

Or consider the scenario where you’ve received a product that unfortunately didn’t meet your expectations. Communicating your concerns and requesting a refund or exchange can sometimes be awkward. This tool could step in to help draft a message that’s polite yet assertive, ensuring your feedback is heard and addressed without sounding confrontational.

Users need to navigate through Chrome’s settings to the Experimental AI section. While it’s currently available only in the U.S. for English language prompts and exclusive to Mac and Windows PC users, its potential for wider application is immense. Beyond personal use, imagine its impact on small businesses, streamlining customer communication, or enhancing product descriptions without the need for a professional copywriter.

RELATED:

• Google Unveils the Android Classic Collectible Figurine, Labelled “The Bot”
• Google Unveils Gemma, its Next-Gen Public AI Models for Ethical Development
• Xiaomi 13 Ultra Premium Camera Phone is now only $799
• How to add custom GIFs and stickers to WhatsApp

(Via)

The post This New Google Chrome Feature will Make Your Write-Ups Way Better appeared first on Gizmochina.

This might be OpenAI‘s biggest announcement since the launch of ChatGPT. An AI-powered video generator that creates videos from merely a couple of sentences? Meet Sora, OpenAI’s latest brainchild, which is bound to change the way videos will be created on the internet forever. And yes, it would not be wrong to say that Sora is crazy good from the examples shown so far.

Sora holds immense potential for filmmakers, designers and artists.

Sora can make videos up to a minute long, focusing on keeping the video close to what the user asks for. Right now, Sora is being tested by experts looking for any potential issues, and it’s also in the hands of visual artists, designers, and filmmakers. Their feedback will help improve Sora, especially for creative work.

Sora stands out because it can handle complex scenes with many characters, different movements, and detailed backgrounds. It gets what the user wants and how those things should look and act in the real world. The tool is smart with language, which lets it create videos with characters that show strong emotions. It can even keep the same characters and style across different parts of a video.

However, Sora isn’t perfect. It might get physics wrong in complicated scenes or miss the mark on cause and effect. It might also mix up details like left and right or have trouble with exact descriptions of events happening over time.

OpenAI is sharing Sora’s development early to get feedback and give a peek at what AI might do next. A lot of security protocols will be put into place it seems, because technology like this is prone to misuse in the lightest sense of the word. Also, keep in mind that it won’t be cheap to produce longer videos, since this will require A LOT of processing power from OpenAI’s servers. Regardless, one thing is for sure -the future of generative AI is more than just exciting – it is inviting!

By the way, if you are wondering how to use Sora AI, know that the access is still limited and it will take some time for the rollout.

RELATED:

• OpenAI passes $2 Billion revenue milestone spurs ambitious Chip-making venture
• OpenAI CEO Sam Altman seeks Trillions in funding to boost Chip technology
• Big Discount: AOOSTAR R1 N100 NAS Mini PC Only For $159
• OnePlus partners with Pixelworks to Elevate Mobile Gaming Experience on the OnePlus 12

(Via)

The post Behold OpenAI’s New Launch – Sora, which can Generate Stunning Videos from Text Prompts appeared first on Gizmochina.

As OpenAI rolls out its latest feature for the DALL-E 3 image generator, it’s clear that the future of digital content is on a fast track to becoming more transparent and trustworthy. The addition of new watermarks to images created by DALL-E 3 is a significant step towards distinguishing AI-generated images from those created by humans, in an era where digital content’s origins are increasingly scrutinized.

Now it will be easier to trace the origins of digital images generated from OpenAI

This move involves embedding watermarks into the image metadata, a strategy backed by the Coalition for Content Provenance and Authenticity (C2PA). The implementation of these watermarks aims to make it easier for everyone to trace the origins of digital images, ensuring users can verify whether an image was generated by AI. The watermark comes in two forms: an invisible metadata component and a visible CR symbol, positioned discreetly in the image’s top left corner.

Starting from the ChatGPT website and extending to the DALL-E 3 API, this feature will soon be available for mobile users, promising a seamless integration that maintains the high quality of generated images. Despite concerns about potential increases in image sizes and processing times, OpenAI assures users that these changes will have minimal impact.

Behind the initiative is the C2PA, a consortium of tech giants like Adobe and Microsoft, advocating for digital content authenticity through the Content Credentials watermark. This effort is not just about adding a layer of transparency; it’s about fostering a digital ecosystem where the line between human and AI-created content is clear, enhancing the trustworthiness of online content.

However, challenges remain, such as the ease with which metadata can be stripped away, deliberately or accidentally, by social media platforms or through simple actions like taking a screenshot. This vulnerability underscores the ongoing battle against misinformation and the complex landscape of digital content verification.

RELATED:

• Apple Vision Pro gets OpenAI’s ChatGPT app
• Microsoft employee discovers a vulnerability in OpenAI’s DALL-E 3 that allows it to generate inappropriate content
• Xiaomi Band 8 Genshin Impact custom edion get a huge discount.
• Best VR / AR Headsets of 2023 – Gizmochina

(Via)

The post OpenAI Introduces Watermarks for DALL-E 3 Images to Boost Authenticity appeared first on Gizmochina.

Google is leaving no stone unturned to bring AI utilization to its products and services. Since the launch of Gemini, the search giant has been actively pushing artificial intelligence-based features into compatible products, including Chrome, Android, and others. In addition to several UI-based and performance updates, Google is rolling out several new generative AI features for Chrome users. Google Chrome is rolling out three experimental generative AI features in its latest release (M121). Users on Macs and Windows PCs in the U.S. can access these features.

Generative AI Features on Chrome

AI Tab Groups

Along with enhancing the seamless browsing experience, the new update enhances the browser’s AI prowess. The first feature, Tab Organizer, automatically suggests and creates tab groups based on open tabs, streamlining the browsing experience. If you’re searching for a new smartphone and, in the process, have opened multiple tabs, then the feature can become handy. It will help to group all the smartphone search-related tabs into a tab group. Since the inception of the Tab Groups feature last year, there haven’t been any major changes to the feature. The new generative AI features seem to elevate the present Tab Group feature to a new level.

Custom Themes with AI

The second feature allows users to create custom themes with AI using a text-to-image diffusion model. From generating wallpapers and now to custom themes, Google has been working diligently to enhance Chrome’s AI prowess. The same can be accessed in the US by heading to the “Customize Chrome” side panel, clicking “Change theme” and then “Create with AI.” The users can add additional details like animation style and choose between different scene moods, like serene. The users can also use the uploaded photos to create a new theme.

Help Me Write

Lastly, an AI-powered writing assistance feature, “Help me write,” assists users in drafting content on the web with suggested text. It is set to launch in the next month’s release. Users can right-click on a text box or field on any site and select “Help me write.” It will be quite helpful by providing suggested text and making tasks like leaving reviews, RSVPs, or inquiries more streamlined and user-friendly.

These experiments aim to enhance browsing efficiency while maintaining personalization. AI and ML integration, including the new AI model Gemini, is expected to bring further improvements to Chrome in the coming year. The new generative AI features are not only enhancing the functionality of the Chrome browser but also enhancing the customization ability.

To enable these features, the users should head to “Settings” and navigate to the “Experimental AI” page after signing in.

Related:

• Google Chrome will Now Let Users Easily See How Much RAM are Tabs Consuming
• Google is working on a new “Performance Panel” feature for Chrome
• Get the Realme GT5 Pro phone on Giztop for $599
• Xiaomi 13 Ultra Premium Camera Phone is now only $799
• Best of CES 2024 – Products that stood out this year!
• Best Smartphones Awards 2023

The post Chrome Unveils Generative AI Features for a Smarter Browsing Experience appeared first on Gizmochina.

In a groundbreaking move, a Chinese court recently recognized the copyright of an AI-generated image, setting a precedent that could significantly impact the burgeoning field of artificial intelligence and creativity. This decision, made by the Beijing Internet Court, revolves around a picture created using Stable Diffusion, a text-to-image software. The court’s recognition of the image as a protected artwork underlines the increasing significance of AI in the realm of artistic creation.

Acknowledging the human element in AI is a big deal for the future

The core of this ruling lies in the court’s understanding that the human creator’s originality and intellectual input played a crucial role in the creation of the AI-generated image. This perspective shifts the focus from the AI itself to the creative process and decision-making of the human behind the AI. By acknowledging the human element in AI-generated content, the court opens new doors for artists and creators who are exploring this novel domain.

This decision is particularly noteworthy as it diverges from the stance in other countries, like the United States, where the debate over copyright protection for AI-generated content remains unresolved. It emphasizes the need to view such creations on a case-by-case basis, considering the unique blend of AI and human input.

The dispute that led to this ruling began when an individual, identified as Li, used Stable Diffusion to create an image of a young Asian woman. After Li’s work was allegedly used without permission on Baijiahao, a Chinese content-sharing platform, he initiated an intellectual property infringement lawsuit. The court’s ruling in Li’s favor reinforces the idea that AI-generated images, when infused with human creativity and judgment, can be seen as personal expressions deserving of copyright protection.

This landmark ruling is a significant step for China’s growing interest in generative AI. It not only acknowledges the potential of AI in creative industries but also sets a legal precedent that could encourage more innovation and exploration in this field. As AI continues to evolve and find its place in various sectors, decisions like this will be crucial in shaping its role and the rights of those who harness its power for creative expression.

RELATED:

• TP-LINK BE3600/5100/6500 Wi-Fi 7 routers now available in China, starting at 229 Yuan ($32)
• Apple’s Limited-Time iPhone Discount for Lunar New Year in China, Starts Soon!
• Get $100 Off on Vivo X100 Pro at Giztop
• Best VR / AR Headsets of 2023 – Gizmochina

(Via)

The post Chinese Court Upholds Copyright in Landmark AI-Generated Image Case appeared first on Gizmochina.

Baidu, a Chinese leader in AI technology, has achieved a significant breakthrough with its ERNIE Bot, a state-of-the-art generative AI tool. This bot, which is a product of the ERNIE Foundation Model, has now reached over 100 million users. What’s more impressive is the ERNIE Foundation Model 4.0, which has seen a 32% boost in performance in just two months.

Baidu claims that the bot has produced 3.7 billion words in workplaces

Since its debut in late August 2023, ERNIE Bot has rapidly become a go-to solution for a wide range of tasks. Its versatility is evident in its usage statistics: it has produced 3.7 billion words in workplace settings, aided over 2 million professional users, penned 300 million lines of code, and processed 400 million words in contracts. It’s not just about work, though. ERNIE Bot has also created 5 million travel plans and helped write 10.83 million supportive messages, earning 20 million likes from users.

These numbers are more than just statistics. They reflect a growing reliance on AI tools in everyday life and work. From drafting codes to planning trips, ERNIE Bot is reshaping how we approach tasks that were once time-consuming and complex. It’s a testament to the evolving relationship between humans and technology, where AI is no longer just an assistant but a collaborator.

Baidu’s commitment to advancing AI doesn’t stop with ERNIE Bot. The Wave Summit highlighted the company’s PaddlePaddle platform, which has become a critical resource for developers and enterprises. With over 10.7 million developers and 235,000 enterprises using it to create 860,000 models, PaddlePaddle is at the forefront of AI innovation.

The rise in ERNIE Bot’s popularity, combined with the continuous improvements in the ERNIE Foundation Model, signals a significant shift in AI’s role in our lives. As these tools become more integrated into various sectors, we can expect to see more groundbreaking changes, simplifying tasks and opening new possibilities for innovation and efficiency.

RELATED:

• Geely and Baidu unveil the Jiyue 07 EV in China with a Red Dot Award 2023-winning design
• Baidu’s Strategic Approach to Navigate AI Development amidst US Export Restrictions
• Get $100 Off on Vivo X100 Pro at Giztop
• Best Messaging Apps for Android in 2023

The post Baidu’s ERNIE AI Bot Crosses the 100-Million-User Milestone appeared first on Gizmochina.

Apple is currently in talks with major news and publishing giants, including Condé Nast, NBC News, and IAC, in a bid to secure permission to utilize their content in the development and training of generative AI systems. The tech giant’s proposed multiyear deals, valued at a minimum of $50 million, aim to license the extensive archives of news articles from these renowned organizations.

The crux of Apple’s technological endeavor lies in the application of neural networks, a sophisticated form of artificial intelligence that learns intricate patterns through vast datasets, such as digital text and photos. This strategic move aligns with the broader industry trend, as Microsoft, OpenAI, Google, Meta, and other tech behemoths have already introduced products leveraging similar neural network technology, including innovative chatbots.

However, the negotiations have not been without their fair share of challenges. Executives from the publishing world have expressed reservations about the expansive terms put forth by Apple. The initial proposal included broad licensing of publishers’ archives, raising concerns about potential legal liabilities stemming from Apple’s utilization of their content. The lack of clarity regarding how generative AI would be applied to the news industry has also left some stakeholders uneasy.

Sources close to the discussions, speaking on the condition of anonymity, revealed that Apple’s intentions were met with skepticism. Some publishers questioned the competitive risks associated with Apple’s significant news audience on its devices. Apple seems to have encountered difficulties in assuaging these concerns, leading to an air of uncertainty surrounding the negotiations.

Despite these challenges, the potential impact on the news industry is substantial. The integration of generative AI has the power to reshape traditional workflows, offering the promise of significant revenue generation. Apple’s move aligns with a broader trend within the tech industry, where major players are aggressively investing in AI technologies.

In response to media inquiries, Apple has remained tight-lipped, refraining from immediate comments on the reported negotiations. The company’s recent product releases, including the unveiling of new MacBook Pro and iMac computers with specialized chips designed to enhance AI applications, underscore its commitment to advancing artificial intelligence research. Apple asserts that these chips, by addressing limitations on data storage in computer memory, hold particular promise for AI researchers, especially those working on cutting-edge creations like chatbots.

Related:

• Apple won’t repair out-of-warranty Watch in the US due to Import Ban
• Apple Vision Pro launch could be set for February 2024
• Humane’s AI Pin, a screenless wearable projector, will start shipping in March 2024
• India Embraces AI for Enhanced Weather Forecasting Amid Rising Extreme Events

(via)

The post Apple’s $50 Million AI Push: Negotiations with News Giants raise stakes in Tech’s Neural Network Race appeared first on Gizmochina.

At the Artificial Intelligence Forum, Samsung introduced “Samsung Gauss,” a powerful generative AI model named after mathematician Carl Friedrich Gauss. This AI model marks a significant advancement in Samsung’s technology, designed to improve user experiences and productivity. Now the company has presented this AI model again at the ongoing two-day Samsung Developer Conference (SDC) Korea 2023, which began on November 14th in Seoul.

Samsung Gauss will support multiple languages

The Samsung Gauss AI suite consists of three specialized components: Samsung Gauss Language for linguistic tasks like email composition and translation, Samsung Gauss Code for coding assistance in software development, and Samsung Gauss Image for image generation and editing, reflecting Samsung’s dedication to generative AI in various domains.

Lee Joo-hyung from Samsung Research’s Global AI Center highlighted that Samsung Gauss goes beyond language understanding and generation. It aims to enhance user-device interaction, providing a more intuitive and natural experience. This technology is expected to offer users convenience, personalization, and the ability to engage in creative activities with Samsung devices.

Additionally, the AI operates on-device, ensuring that consumer data is processed and stored locally, addressing common concerns about data security in AI applications.

Now according to tipster @Tech_Reve, Samsung’s Gauss language model supports various languages, including Korean, English, French, Spanish, Chinese, and Japanese, indicating its availability for Chinese users in the future.

Meanwhile, Samsung also announced plans to launch Galaxy AI in their upcoming Galaxy flagship phones, expected to be the Galaxy S24 series, enabling real-time audio and text translations during phone calls. Renowned tipster, Ice Universe confirmed through a conversation that Galaxy AI will indeed be usable on domestic Galaxy S24 series phones.

RELATED:

• Google paid Samsung $8 billion to ensure its position as the default search engine
• Samsung Galaxy A25 5G Bags TUV Rheinland Certification
• Samsung’s Android Impact: More Than Half of Google Play Revenue Stems from Samsung Devices
• Honor Magic Vs 2 vs OPPO Find N3: Specs Comparison
• Download the Best GCam APK for Motorola Edge 40 Neo
• Samsung Galaxy Z Fold5 vs OnePlus Open: Specs Comparison

(Source, Via)

The post Samsung Gauss AI will bring Chinese language support to the domestic Galaxy S24 series appeared first on Gizmochina.

Oppo is ramping up its investment in artificial intelligence (AI) in an ambitious bid to bring the power of large language models (LLMs) directly to your smartphone. The game plan? Integrate its own homegrown language model, AndesGPT, into its devices, and make AI chat features more responsive and reliable by running them locally on the phone itself.

Oppo wants their phones to understand their customers’ worlds

Jason Liao, the head honcho at Oppo Research Institute, is all-in on the idea. At a recent media briefing in Singapore, he confirmed the company’s enthusiasm for AI and how it could reshape our smartphone experiences. But there’s a catch. Localizing AI services on a phone isn’t an easy task. Oppo has the significant challenge of figuring out how to give the phone enough computing muscle to run LLMs without relying on the cloud.

This move to localized AI could be a pivotal moment for Oppo, setting it apart in a market that’s chock-full of fierce competitors like Vivo, Huawei, and Xiaomi—all of whom are on their own AI journeys. Oppo’s vision involves not just AI chat but also delves into computer vision tech that could understand and interpret images and videos in real-time. Imagine your phone not just hearing you, but “seeing” and understanding your world as you do.

So why should we care? Oppo is already a major player, ranked as the world’s fourth-largest smartphone vendor. Its efforts to create an in-house, locally-run AI could help it carve out a unique niche. While other companies look for cloud-based solutions, Oppo’s local approach could mean faster, more reliable AI services for users, all without needing an internet connection.

In addition, Oppo’s active collaboration with tech startups through grants and competitions shows it’s keen on crowd-sourcing innovation. From 3D-tracking tech to contactless health monitoring, Oppo is clearly willing to bet on fresh ideas that align with its vision.

RELATED:

• Oppo Find N3 press renders & specs revealed
• OPPO Find N3 camera samples revealed ahead of October 19 launch
• Best Pixel 8 Pro accessories 2023

(Via)

The post Oppo is Committing to Generative AI to Gain an Advantage in the Smartphone Market appeared first on Gizmochina.

MediaTek has collaborated with OPPO and ColorOS to create a lightweight large model edge deployment solution, with the goal of gradually enabling large model capabilities on edge devices. Additionally, OPPO has introduced the beta version of Xiao Bu 1.0, which is based on the AndesGPT large model, for early testing.

MediaTek and OPPO’s Collaborative Edge Deployment Solution

Edge AI refers to using artificial intelligence on local devices rather than relying solely on centralized cloud processing. It offers advantages like reduced latency, improved privacy, and offline functionality. Common Edge AI applications include autonomous vehicles, industrial automation, smart cameras, healthcare, retail, smart cities, agriculture, energy management, IoT devices, NLP, robotics, and gaming/AR/VR. It empowers devices to operate independently while maintaining data privacy and responsiveness, making it crucial in various industries.

The collaboration between MediaTek and OPPO is centered around MediaTek’s AI processor, APU, and the AI development platform, NeuroPilot, which together create an ecosystem for terminal-side AI and generative AI computation. This partnership accelerates the development of edge AI applications, with a particular focus on improving large language and generative AI models.

One notable outcome of this collaboration is OPPO’s AndesGPT, a substantial generative language model. AndesGPT employs 4-bit quantization technology to maintain high accuracy when deployed on edge devices, and it has demonstrated enhanced performance.

In testing with OPPO’s digital assistant, Xiao Bu, AndesGPT achieved an impressive score of 98.33 in the “Knowledge and Encyclopedia” category of the SuperCLUE benchmark, positioning it as a leading domestic language model, second only to GPT-4. This achievement has gained significant recognition in the tech community.

OPPO intends to extend the use of AndesGPT beyond its digital assistant, Xiao Bu, and incorporate it into a wider range of the company’s core products.

RELATED:

• Mediatek Dimensity 9300 Leaked to Feature Only 8 Performance Cores, and No Efficiency Cores
• MediaTek Has Historic-First Longer Software Support than Snapdragon Flagships in 2023
• MediaTek continues to lead smartphone AP processor market for 12th consecutive quarter
• iPhone 15 Pro vs Google Pixel 8 Pro: Specs Comparison
• iPhone 15 Pro vs Google Pixel 8 Pro: Specs Comparison
• How To Launch a Mobile App in The Right Way?

(Via)

The post MediaTek and OPPO collaborate to bring large-model AI to the edge appeared first on Gizmochina.

In recent news, Meta Platforms Inc. introduced 28 AI chatbots that wear the faces of celebrities, ranging from Kendall Jenner to Tom Brady. These are not your average chatbots; they offer more than just mechanical responses. Kendall Jenner’s chatbot, named Billie, serves as a virtual big sister, dishing out advice to users, while Tom Brady’s counterpart, Bru, engages in sports debates.

Brand safety is a huge concern for generative AI

While the industry is buzzing about the innovation, it raises eyebrows on several fronts. For one, Meta is pulling out all the stops financially. According to The Information, one top-tier creator has been offered as much as $5 million for just six hours of studio time to use their likeness. The heavy price tag underlines the tech giant’s commitment to making AI relatable and personalized. But is it too much too soon?

Meta’s CEO Mark Zuckerberg is optimistic, believing there’s a “huge need” for AI versions of celebrities. However, he has also acknowledged that rolling out such technology might take longer due to brand safety concerns. Celebrities are wary about their image being manipulated into making inappropriate or contentious remarks.

It’s not just the AI celebrity figures that have raised questions. Meta also launched AI stickers that quickly came under fire. Users managed to create disturbing images, including a child soldier and a lewd portrayal of Justin Trudeau. These instances pose a concern about the ethical use of AI technology and content moderation.

Given the setbacks, it’s safe to say that Meta is treading on complex terrain. The company’s aspiration to revolutionize the chatbot experience could either usher in a new era of interactive AI or open Pandora’s box of ethical dilemmas and brand safety nightmares. It’s a gamble that has the tech world watching closely.

RELATED:

• Meta launches innovative Generative AI Tools for advertisers
• Meta Faces Obstacles with Metaverse, Plans to Lay Off Employees from Chip Division
• Best Apple Watch Cases in 2023: Spigen, Otterbox, Casetify & More

(Via)

The post Meta is Reportedly Paying Celebrities Millions of Dollars to Use their Personality as AI Chatbots appeared first on Gizmochina.

Privacy Concerns have always remained part of the Generative AI services since its short lifetime. These concerns came into the limelight when OpenAI introduced ChatGPT and Google introduced its Bard. The learning methods and data for these tools have always remained a concern for the users and lawmakers. However, Meta has detailed its protective measures to protect User Privacy and how it collects data for the Generative AI tools that it launched recently along with 28 AI personas chatbots.

Must See: Meta launches new Ray-Ban Smart Glasses with live streaming, 12MP camera, and longer battery life

There are some privacy checks that Meta has established to safeguard user privacy in the new Generative AI tools. Following the previous check from the development language models like Llama 2 and others, Meta has further strengthened the “privacy safeguards and integrity mitigations” while building new tools and services based on AI.

From Where Meta Generative AI Tools Fetch Data?

To better train Generative AI tools, different sets of enormous data sets are required for better information and reliability of such tools. However, feeding such large systems with by eliminating privacy concerns has always remained a task. But the Beta has achieved it by using only public data from Meta services and products. It has excluded some websites where the user shares their personal information. For Eg: On Facebook, the company has processed only the data that is labeled as public, while it has not even touched the content under the private tag.

Even the data on messages are not touched while training their AI generative Tools. However, if the user asks anything from Meta AI by adding @Meta AI in their message, then only that message is processed by these tools. Otherwise, all the messages remain private. In addition, the company uses data from AI Stickers in your chat to improve these stickers.

Even, the businesses interacting with Generative AI tools such as Meta AI are part of the data sharing pacts. The company fetches the data as presided over by their new Privacy Guide.

Why does Sometimes AI Retain Information?

While using the Generative AI tools, these may retain some of your personal information to personalize the responses. It may be helpful for the users as it may help in generating more precise answers to your queries. However, there are always safeguards to protect your private information appearing in other’s responses. While talking about such issues, Meta details that “We use automated technology and people to review interactions with our AI so we can, among other things, reduce the likelihood that models’ outputs would include someone’s personal information as well as improve model performance.”

Built-in commands to Delete Private Data in Meta AI

To further equip users with better tools to delete their private information, Meta introduced some built-in commands to delete the information.

• /reset-ai – Deletes AI messages.

Further, the generative AI tools keep the account information on different Meta platforms – Facebook, Instagram, and other Apps separate.

Meta aims to protect user privacy and even has dedicated work members to abide by these privacy policies. Further, the Meta has signed the White House’s voluntary commitments for AI and is a participant in AI’s Synthetic Media Framework.

Related:

• Meta introduces 28 AI Personas for Facebook, Instagram and WhatsApp
• Meta launches new mixed-reality headset ‘Quest 3’ starting at $499.99
• Meta’s Threads will Impose Strict Guidelines around Potentially Sensitive Topics
• Meta & LG team up for a future mixed-reality headset could launch in 2025

The post Here are Meta’s Protective Measures to Enhance User Privacy while Interacting with its Generative AI Services appeared first on Gizmochina.

As Hong Kong strives to solidify its role as a global financial hub, the widespread adoption of artificial intelligence (AI) tools like ChatGPT faces roadblocks, largely due to copyright and legal concerns. Despite the buzz, businesses are taking cautious steps in integrating these tools into their workflow.

Understanding Complex Language Models is an uphill task

Herbert Chia, who helms the data governance steering committee at the Hong Kong Science and Technology Parks Corporation, pointed out at an IAB Hong Kong event that it’s not just about jumping on the AI bandwagon. “Understanding and adopting these large language models is a complex task,” he said. Chia’s perspective comes with weight, given his past role at Alibaba Group between 2010 and 2016.

Hong Kong’s banking giant, HSBC, is one of those treading carefully. The bank dipped its toes into generative AI by launching an ad campaign featuring an AI-generated avatar of Hong Kong comedian Dayo Wong. The move sparked new challenges, primarily concerning ownership of the generated image. “It’s uncharted territory. Traditionally, the celebrity-owned the image rights, but now we’re generating something new,” said Cheuk Shum, head of marketing for Wealth and Personal Banking at HSBC Hong Kong.

Interestingly, Hong Kong’s internet users have been quicker to adapt. A chatbot aggregator, Poe, developed by Q&A platform Quora, claims nearly 49% of its global user base from the city. This suggests that while the general public is eager to embrace AI, businesses remain skeptical due to murky legal waters.

RELATED:

• China’s First Suspended Monorail Opens for Operation, Offering a Futuristic Ride
• China’s Huge $13.7 Billion Investment Push in AI, Chipsets Points Towards a Strategic Shift
• Download the Best GCam APK for Redmi 12 5G

(Via)

The post Hong Kong Struggles with Generative AI Amidst Copyrights & Legal Concerns appeared first on Gizmochina.

During its 20th-anniversary event, TAL Education Group launched the public beta testing of its innovative mathematical large model, MathGPT. This LLM model is designed primarily for global mathematics enthusiasts and research institutions, marking a significant milestone as China’s first large model tailored for mathematics.

Simplify Your Math: TAL’s MathGPT Now Available for Public Testing

MathGPT boasts features like problem-solving algorithms and explanations. It allows users to input math problems, either as text or images and get interactive solutions. A unique feature enables users to request random math questions. Currently, it is bilingual, supporting both Chinese and English and is compatible with both PC and mobile devices.

A considerable advantage of MathGPT is its foundational data, sourced from years of TAL Education Group’s extensive educational and research content in the domain of mathematics. This foundation, combined with continuous training and fine-tuning, positions MathGPT as a top performer in its class. Its superior performance metrics encompass problem-solving accuracy, stability, and overall user experience. Additionally, human feedback mechanisms are set to bolster the model’s efficacy even further.

The platform caters to a wide academic range, covering elementary to high school mathematical problems. This includes various math segments like computation, algebra, and application, among others. The platform’s interactivity extends to follow-up questions about math problems, but it currently doesn’t support questions outside the math realm.

With MathGPT, TAL envisions a transformative educational landscape where AI complements traditional teaching methods. This fusion promises reduced costs, increased personalization, and a revolutionary approach to learning, emphasizing AI-assisted Q&A tailored for individual students.

During MathGPT’s ongoing public beta, its capabilities are expected to improve. The TAL Education Group is actively developing applications centered around MathGPT and is on track for an upcoming product launch.

RELATED:

• Microsoft May Bring AI Capabilities to Paint, Photos, and Other Windows 11 Apps
• YouTube is Taking Steps to Earn Revenue From AI-Generated Music
• Adobe Express brings AI-powered tools with Adobe Firefly
• OnePlus Ace 2 Pro vs Redmi K60 Ultra: Specs Comparison
• Realme 11 4G vs Realme 11 5G: Specs Comparison
• How to Turn Off Chrome browser’s new Download Tray

(Source, Via)

The post TAL’s MathGPT launches public beta: your next-gen math assistant appeared first on Gizmochina.

In a significant technological advancement, MediaTek and Qualcomm have recently revealed their collaborative efforts with Meta. Both semiconductor giants announced plans to implement Meta’s Llama 2 large language model (LLM) on mobile devices equipped with their upcoming flagship processors.

More manufacturers are expected to follow suit and design chipsets for AI-based applications

MediaTek’s next-generation flagship chipset, rumored to be the Dimensity 9300, is scheduled to launch later this year. It will feature a software stack optimized to run Llama 2 and upgraded hardware to enhance the performance of generative artificial intelligence applications. Qualcomm previously announced similar plans last month, indicating a trend towards on-device generative AI in the industry.

One key benefit of these innovations is increased privacy. Running generative AI tasks locally means users’ data doesn’t need to be sent to the cloud for processing. This can lead to enhanced security measures and greater control over personal information.

Details of the Dimensity 9300 chipset suggest a radical design, with specifications including Arm’s Cortex-X4 and A720 CPU cores and the Immortalis-G720 GPU. This technical composition suggests a strong commitment to delivering a robust and efficient AI experience directly on mobile devices.

These developments represent a significant step towards more powerful, privacy-conscious smartphones. As both MediaTek and Qualcomm move towards on-device AI capabilities, consumers can look forward to more responsive and secure mobile technology by the end of 2023.

In the context of growing concerns about privacy and data security, these announcements are a welcome move. The collaboration with Meta in the field of on-device generative AI is likely to set a new standard in the mobile industry and may drive further innovation in the coming years.

RELATED:

• Qualcomm to use TSMC’s N3E process next year alongside MediaTek, Won’t switch to Samsung
• Blackview A200 Pro with MediaTek Helio G99 chipset and 108MP main camera unveiled
• Download the Best GCam APK for Realme 11 Pro

(Via)

The post Dimensity 9300 Processor will Run Generative AI on Smartphones, Says Mediatek appeared first on Gizmochina.

Adobe is finally bringing AI-powered tools to Adobe Express, which is the company’s cloud-based designing platform. The new AI generative tool will be offered through the Adobe Firefly model, giving it an edge over similar cloud-based design platforms like Canva.

Adobe Express now supports generative AI tools

With the addition of this new technology, the brand is promising that users will gain access to tools for their convenience. So now, Adobe Express will let make quick and easy templates to create social graphics, posters, edit videos, and much more. This feature is being rolled out with the latest version of Express on desktop PCs for free. Although, the company plans on releasing this feature to mobile users soon.

The Adobe Express Premium plan, which is a 9.99 US Dollar monthly subscription, is also included for no additional cost. Furthermore, Adobe has also released an Express enterprise tier, which is also available for larger teams. Adobe Express had originally released back in 2021, but it has been working on the new generative AI tool since June of this year.

There are a number of new features that let’s use to generate custom images, text effects, and more. Users can add these to their projects by using the text prompts in over 100 languages. Other functionality includes automatic background revival and simple character animations. Adobe guarantees that Firefly is commercially safe for companies to use since the model is trained on Adobe’s proprietary content.

RELATED:

• Adobe and Nvidia Introduce AI Imagery Systems to Address Copyright Concerns
• Microsoft rolling out a Windows 10 update to remove Adobe Flash Player
• Adobe officially ends support for Flash, recommends uninstallation
• Photoshop to get an advanced version of MIUI Gallery’ Sky Replacement feature
• Redmi K60 Ultra vs Redmi K60 Pro: Specs Comparison
• Xiaomi Mix Fold 3 vs Honor Magic V2: Specs Comparison

(Via, Source)

The post Adobe Express brings AI-powered tools with Adobe Firefly appeared first on Gizmochina.

The year 2023 has ushered in a tidal wave of opportunities in the generative AI industry. With job openings quadrupling since the beginning of the year, this surge represents more than just an increase in demand; it reflects a seismic shift in the technology landscape.

Multiple platforms are starting to implement AI, increasing demands for relevant roles

Leading tech giants like Amazon, Netflix, Meta, and even unconventional players like the dating site Hinge, are competing in a fierce battle for AI expertise. Salaries in the seven-figure range are no longer a rarity. Netflix’s whopping $900,000 offer for an artificial intelligence product manager position exemplifies this trend, as does Capital One’s list of top salaries for generative AI engineers.

While these figures are eye-popping, they reveal a critical shortage of skilled personnel in the field. They also highlight a determined push to innovate and excel in areas that were previously untapped. Walmart‘s hunt for a senior manager for its conversational AI platform stands as evidence of this broader application of AI technology.

These enormous salaries are not just numbers; they symbolize the value and urgency that the industry places on advancing this cutting-edge technology. They invite aspiring professionals to join the AI revolution and challenge current experts to push the boundaries.

RELATED:

• The Godfather of AI, Geoffrey Hinton, Reveals the Future of Artificial Intelligence
• X.AI Corp is Elon Musk’s New Artificial Intelligence Company
• Best Docks for Mac 2023 – Anker, Belkin, Baseus & More

(Via)

The post The Generative AI Talent Gold Rush is More Than Just Numbers appeared first on Gizmochina.

Xiaomi’s mega launch event has concluded, leaving us with several new products. The spotlight is on the Xiaomi Mix Fold 3—the thinnest horizontal foldable smartphone. The new waterproof Redmi K60 Ultra also receives some fanfare. Among these exciting announcements, Xiaomi also slipped in the news of its voice assistant, XiaoAI, going all generative AI. Here’s what you should know.

The new XiaoAI could go against ChatGPT

Xiaomi’s CEO, Lei Jun, announced during his annual speech today that XiaoAi will be upgraded to a generative large language model. For the unaware, XiaoAI is Xiaomi’s equivalent of Google Assistant. If you didn’t hear of it before, it’s because the voice assistant is currently limited to the China ROM.

The company also showcased a video detailing what the generative AI-equipped XiaoAI can do. It will now intelligently answer your questions and will be able to hold a conversation similar to mainstream models like ChatGPT and Bard. The voice assistant will be able to do copywriting and even picture content according to the user’s text requirements.

The new XiaoAI can also help with summarizing and translating in real time. For Xiaomi products, the fresh XiaoAI can be a helpful in guiding users on how to use them.

The large language model is currently under testing and Xiaomi has started test recruitment. The new version of XiaoAI will first be supported on the Xiaomi Mix Fold 3, Xiaomi 13 series, Xiaomi 12 series, Redmi K60 series, and Redmi K50 series smartphones. Xiaomi Sound speakers will also be able to participate in the first round of testing.

Users whose devices meet the requirements can apply for the test in the Xiaomi Community App or say “Experience the Large Model” to XiaoAI to go directly to the registration page.

Related:

• Xiaomi Mix Fold 3 launched with more durable hinge, periscope camera, overclocked SD8 Gen 2 chip, and more
• Xiaomi Mix Fold3 is on pre-sale on Giztop
• Xiaomi Smart Band 8 Pro With 60Hz Display, Somatosensory Support Launched In China
• Redmi K60 Ultra launched with IP68 rating, 24GB RAM, Sony IMX800 camera, and more
• Hohem iSteady MT2 4-in-1 Gimbal Kit with AI tracking launched in India

(Via)

The post Xiaomi’s XiaoAI Voice Assistant Gets a Generative AI Upgrade appeared first on Gizmochina.

AI now has the capability to generate music samples from text prompts using neural networks. It’s actually great for producing compositions based on provided descriptions like mood or style. These AI models can be used to generate video game soundtracks, or simply to break creative block while relying on quality data and ongoing AI advancements. The innovative tech team Samplab has unveiled TextToSample, a VST3 (Virtual Studio Technology) plugin that allows users to convert text prompts into generated audio samples. The tool, available for free, runs both independently and within a DAW (Digital Audio Workstation), and could be the answer to generating a unique “melody” straight from your thoughts. Imagine the possibilities.

The plugin does use a lot of resources while generating even a medium-sized clip

Drawing upon the potential of Meta‘s open-source MusicGen, an AI-based sound generation suite, TextToSample was developed using the data fed by this advanced algorithm. Adding to its capabilities, the plugin can accept uploaded sounds, analyze them, and generate related audio samples, besides processing text instructions.

Sporting a very minimalist interface, TextToSample does, however, come with a few challenges. The responses are not always accurate, and you have to grapple and wrestle back and forth with multiple prompts to actually get where you need to be. So, perfection may not be immediate, but there’s a chance you can get there.

Despite these limitations, TextToSample can be a lot of fun. It’s a great example of generative AI‘s capabilities, especially with regard to music production in 2023, although it’s not without its share of bugs. Some users experienced unwanted additions, such as drums when they specifically opted out of percussion. I personally used the application, and it used up a huge amount of resources to generate a 10-second audio clip that was of mediocre quality at best, slowing my machine down in the process.

RELATED:

• The Godfather of AI, Geoffrey Hinton, Reveals the Future of Artificial Intelligence
• Microsoft is Now Offering Free Generative AI Skills & Training Courses
• Best Foldables of H1 2023 (Fold & Flip)

The post This Free-to-Use AI Model can Generate Audio Samples based on Text Prompts appeared first on Gizmochina.

Posting images online carries the risk of manipulation due to AI-driven technologies that can create deepfakes, alter photos, spread misinformation, and facilitate identity theft. To protect against this, a lot of people usually consider watermarking, sharing lower-resolution images, being cautious about the information shared, and using reliable sources. Raising awareness about image manipulation can also help combat the spread of deceptive content. However, with AI advancing rapidly, even this might not be enough in the long run. This is where the minds at MIT CSAIL come in. They have taken up the responsbility, showcasing their very own “PhotoGuard” technique. Unlike conventional watermarking, it tweaks certain pixels in an image, and it manages to confuse AI algorithms while remaining completely invisible to the human eye. Think of it like a puzzle that leaves Artificial Intelligence scratching its virtual head, clueless about what’s happening.

The more teamwork there is between corporations and developers, the better PhotoGuard can work

The “encoder” attack tricks AI models into seeing the image as something else entirely, while the more advanced “diffusion” attack disguises the image as a different target, making any generative AI edits pretty much a waste of time.

Of course, no solution is foolproof (for now), and PhotoGuard has its limitations as well. For instance, someone who is really desperate might try to reverse engineer the protected images by adding digital noise or using other tricks. That’s where teamwork comes in! The more model developers, social media platforms, and policymakers work collaboratively, the more it can strengthen PhotoGuard’s defenses against sneaky attempts.

As AI technology charges forward, it’s essential for AI developers to invest in robust defenses to safeguard digital art. Only then can we maintain a world where AI and human creativity thrive together. The future of artistic expression is bright, and with the guardians of PhotoGuard on watch, we can rest assured that the virtual canvas remains a safe space for creative minds.

RELATED:

• X.AI Corp is Elon Musk’s New Artificial Intelligence Company
• boAt Airdopes 161 Pro TWS earbuds go on sale in India for just Rs 999
• Best Mid-Range Smartphones of 2023 – OPPO, Moto, Samsung & More

(Via)

The post This Anti-AI tool will Confuse AI Algorithms if Your Picture gets Manipulated appeared first on Gizmochina.

It is hard to believe that chatbots are causing so much mayhem in the tech industry, and Apple – a trillion-dollar global tech giant is staying away from this battle. To be fair, it was only a matter of time before Apple did something, and the company has now broken its silence with the introduction of its own AI chatbot, internally named “AppleGPT” but actually called Ajax, a large language model (LLM) built using Google’s JAX framework. With multiple teams working on the project, including a dedicated group to address privacy concerns, Apple seems like it has a game plan to make it work.

The AI chatbot has been internally named AppleGPT, but will be referred to as Ajax

Despite the exciting news, Apple’s specific plans for the chatbot remain shrouded in mystery. There are no confirmed use case scenarios, launch dates, or platforms for Ajax. An unnamed source has hinted that Apple might make an official announcement with more details next year, and there’s hope that the upcoming annual earnings call might shed some light on the matter.

Apple’s interest in generative AI has been evident. Tim Cook, the company’s CEO, previously mentioned that generative AI was something they were “looking at closely.” Spearheading the Ajax project are John Giannandrea and Craig Federighi, with Giannandrea’s expertise in overseeing Siri and machine learning capabilities potentially leading to significant improvements for the digital assistant.

The generative AI landscape has experienced remarkable growth, with several big tech companies introducing their AI chatbots. Just recently, Meta and Microsoft unveiled Llama 2, an AI chatbot tailored for commercial use. The competition is fierce as companies vie for dominance in this burgeoning field.

As Apple enters the chatbot race, the industry and consumers are eagerly anticipating updates and clarifications from the tech giant. In the coming months, more information will undoubtedly emerge, revealing Apple’s strategy to revolutionize its AI services and compete in the fast-paced AI wars.

RELATED:

• Apple Watch Users In India Diagnosed With AFib Can Now Track Their AFib History
• Apple iPhone 16 Pro Max’s telephoto lens might surpass Samsung Galaxy S23 Ultra
• Best Feature Phones 2023

(Via)

The post Apple’s Contender to ChatGPT Joins the Race with “Ajax” AI Chatbot appeared first on Gizmochina.

In a major development in the world of artificial intelligence (AI), Inflection AI, a startup founded just a year ago, announced on June 30th that it has successfully raised $1.3 billion in financing. The funding round included contributions from industry giants such as Microsoft and Nvidia, among others. This significant investment underscores the growing interest and confidence in Inflection AI’s potential to revolutionize the AI landscape.

Inflection AI creates personalized AI Products that assist the users

Co-founded by Mustafa Suleyman, a co-founder of Google DeepMind, and Reid Hoffman, co-founder of LinkedIn, Inflection AI aims to develop consumer-oriented AI products, positioning itself as a direct competitor to OpenAI. The company has already made a splash in the market with the release of its chatbot, Pi, last month. Powered by generative AI technology similar to ChatGPT, Pi engages users in interactive dialogue, enabling them to ask questions and provide feedback. Inflection AI envisions creating personalized AI products that assist individuals with tasks such as planning, scheduling, and gathering information.

With its headquarters based in Palo Alto, California, Inflection AI currently employs approximately 35 professionals. Earlier this year, the company raised $225 million in its first funding round, attracting investments from renowned venture capital firms Greylock and Microsoft, as well as Reid Hoffman himself.

In a recent report, Inflection AI unveiled its model, Inflection-1, which forms the foundation for the Pi chatbot. The company asserts that its performance surpasses that of most existing language models, positioning it as a leader in the field.

Inflection AI CEO Mustafa Suleyman outlined the company’s plans for the raised capital, stating that a significant portion would be dedicated to building computational capabilities to develop more advanced base models. Suleyman revealed during the Collision Conference that the company aims to construct a computing cluster featuring around 22,000 Nvidia H100 chips, boasting computing power estimated to be three times greater than that used to train GPT4. This enhanced infrastructure will allow Inflection AI to create truly differentiated AI products.

The recent success of OpenAI‘s ChatGPT has generated considerable excitement within the AI industry, solidifying AI as the next frontier of technology. Consequently, the sector has witnessed a surge in investor interest over the past few months. Inflection AI’s substantial funding round and its ambitious plans for AI development demonstrate the growing momentum and competition within the industry. As the company continues to expand its capabilities and offerings, the future of AI holds great promise for both businesses and consumers alike.

RELATED:

• Baidu’s Ernie Bot 3.5 AI outperforms ChatGPT in some key metrics
• Mercedes-Benz’s new Test Program will introduce ChatGPT to U.S. Luxury Models
• Best Smartphones of 2023 – Samsung, Xiaomi, Apple & More

(Via)

The post Inflection AI, ChatGPT Competitor Secures $1.3 Billion in Funding, Microsoft & Nvidia Among Investors appeared first on Gizmochina.

In a move that reflects the growing impact of generative AI technology, Microsoft has unveiled a comprehensive training program aimed at equipping workers with the necessary skills to harness the power of artificial intelligence. Announced as part of Microsoft‘s “Skills for the Jobs” initiative, the program offers free online courses and certificates to help individuals master the fundamental concepts and applications of AI.

The Program is currently only available in English, but will soon expand to more languages

LinkedIn, a subsidiary of Microsoft, has developed a series of online courses that cover introductory knowledge of AI and provide a “framework for responsible AI.” These courses when completed will grant a professional certificate, which is the first of its kind in generative AI available on the online learning scenario. Although currently only offered in English, Microsoft has plans to expand the program to include additional languages such as Spanish, Portuguese, French, German, Simplified Chinese, and Japanese in the near future.

Recognizing the importance of educators in shaping the next generation of AI professionals, Microsoft has also introduced an educator toolkit. This resource offers teachers and trainers downloadable content designed to facilitate AI education in classrooms and training environments. Furthermore, a specialized AI course has been developed specifically for teachers and trainers to enhance their understanding and instructional abilities in the field of AI.

In a bid to promote inclusivity and diversity in the AI sector, Microsoft has launched the AI Open Funding Program for Historically Marginalized Groups. This collaborative effort between data.org, Microsoft’s AI for Good Lab, and GitHub aims to empower nonprofit organizations, social enterprises, research institutions, and academic organizations to train and enable workers to utilize generative AI. The grant program provides financial support, collective learning opportunities, data training and mentoring, as well as access to Microsoft events and cloud computing resources. Interested parties can submit their applications until August 15th.

Microsoft’s new AI training program and initiatives are clearly setting a precedent of the company’s commitment to fostering a skilled workforce that can navigate the evolving landscape of AI technology.

RELATED:

• Microsoft Doesn’t Care If You Buy an Xbox Series X or S
• New Vulnerability Discovered in Microsoft Teams Puts Users at Risk
• Best Sony Over-Ear Headphones 2023

(Via)

The post Microsoft is Now Offering Free Generative AI Skills & Training Courses appeared first on Gizmochina.

Artificial Intelligence (AI) continues to revolutionize various industries, and the motorcycle industry is no exception. Recent developments in AI have opened up new possibilities for innovation and design, particularly in the realm of electric motorcycles. Lightning Motorcycles, a renowned performance electric motorcycle manufacturer based in California, has harnessed the power of advanced AI to enhance its engineering processes.

The AI-generated design also accommodated a 20 percent weight reduction

Partnering with Autodesk, a global leader in design and engineering technology, Lightning Motorcycles has embraced Generative AI to transform the way their high-performance electric motorcycles are designed. This cutting-edge technology allows engineers to input specifications, after which the Generative Design program generates hundreds of thousands of design studies. From this vast pool of options, engineers can carefully select and optimize a design that will be implemented in the final product.

The results have been remarkable. Lightning Motorcycles’ initial foray into AI-driven engineering led to the development of a lightweight swingarm for their LS-218 model. The AI-generated design not only exceeded expectations in terms of strength but also achieved a weight reduction of over 20% compared to the standard swingarm used in the production version of the bike.

The implementation of AI in motorcycle engineering offers several benefits. By leveraging AI’s capabilities, Lightning Motorcycles aims to enhance performance, reduce weight, meet cost targets, and significantly save time in the design and manufacturing processes. This powerful technology enables the creation of designs that were once considered impossible, pushing the boundaries of what can be engineered in the motorcycle industry.

The integration of AI in motorcycle engineering is part of a broader trend in which AI is becoming increasingly intertwined with everyday life. From smart home appliances to artistic endeavors, AI’s presence is pervasive. While the concept of AI may initially evoke concerns for some, it is crucial to recognize it as a tool that has the potential to bring about remarkable advancements and enhance our daily lives.

Lightning Motorcycles’ partnership with Autodesk serves as a prime example of how AI-driven design can revolutionize the manufacturing process, opening doors to unprecedented levels of innovation and efficiency. With AI as a driving force, the possibilities for the motorcycle industry are boundless, and enthusiasts can look forward to a future of cutting-edge design and exhilarating performance.

RELATED:

• An AI-Designed Drug is now Entering Human Clinical Trials
• Samsung to mass produce HBM chips for AI, competing with SK Hynix
• Best VR Games 2023: The Ultimate List of Virtual Adventures and Experiences

(Via)

The post The Integration of Generative AI in Motorcycle Engineering appeared first on Gizmochina.

In response to the rapid growth of generative AI and its far-reaching implications, the House of Representatives has introduced new regulations surrounding the use of the popular AI chatbot, ChatGPT, by congressional offices. This move highlights Washington’s ongoing efforts to grapple with the challenges posed by this transformative technology both at a legislative and personal level.

Offices can only use ChatGPT Plus for privacy reasons

According to a memo obtained by Axios, Catherine L. Szpindor, the Chief Administrative Officer of the House, outlined the guidelines for ChatGPT usage. Offices are now only authorized to utilize the paid version, ChatGPT Plus, which offers enhanced privacy features essential for safeguarding House data. Additionally, the memo specifies that no other large language models are permitted for use.

The regulations further emphasize that ChatGPT is to be employed solely for research and evaluation purposes, prohibiting its incorporation into regular workflow. Offices are instructed to input non-sensitive data, refraining from sharing unpublished text blocks with the chatbot. Privacy settings, initially disabled, should be activated to ensure interaction history is not stored or used to augment the large language model.

This development aligns with broader efforts in Congress to address the implications of AI. A bipartisan group of senators, led by Senate Majority Leader Chuck Schumer, is actively working on comprehensive legislation. Senators Richard Blumenthal and Josh Hawley have conducted hearings on the subject, with Hawley proposing a framework for corporate accountability. In parallel, individual lawmakers have introduced standalone bills, aiming to contribute to the final regulatory package.

As Axios tech and policy reporter Ashley Gold explains, Congress is not unique in its struggle to integrate generative AI into existing workflows while confronting fundamental questions about the technology itself.

By establishing these new guidelines, Congress aims to strike a balance between leveraging AI’s potential for improving operations and ensuring privacy and data protection. As the conversation around AI regulation evolves, it is crucial to find common ground between innovation and accountability, fostering a responsible and ethically-driven AI landscape.

RELATED:

• The Godfather of AI, Geoffrey Hinton, Reveals the Future of Artificial Intelligence
• Samsung to mass produce HBM chips for AI, competing with SK Hynix
• Best Electric Toothbrush 2023: Elevate Your Dental Care with Cutting-Edge Technology

(Via)

The post The US Congress Implements New Rules for ChatGPT Usage in Offices appeared first on Gizmochina.

In a recent development, LinkedIn, the leading workplace social platform, has implemented another aspect of generative AI. LinkedIn now allows users to directly post content using the power of generative AI. For professionals worldwide, LinkedIn has emerged as a valuable platform for sharing up-to-date resumes and networking. Users often post about their career trajectories, promotions, and achievements to enhance their job prospects. However, crafting these posts with careful planning and writing can be a hard task, especially for those who have never posted on LinkedIn before. To address this, LinkedIn’s product director, Keren Baruch, has announced the integration of generative AI into post writing, offering users a faster alternative in this era.

Generative AI will help speed up the writing process greatly, but at what cost?

To utilize this feature, users need to provide a minimum of 30 words outlining the key message they wish to convey in their post. Leveraging user input, the generative AI generates drafts that can be reviewed, edited, or directly published. Keren Baruch stated that the feature is currently in the testing phase and will eventually be accessible to all users.

Nevertheless, a lot of people have already raised their voices in objection within the LinkedIn community regarding this feature. Some users argue that it compromises the authenticity of posts on the platform. They claim that using generative AI could transform interactions on LinkedIn into exchanges between chatbots rather than real human connections.

LinkedIn’s introduction of generative AI for post writing represents a delicate balance between efficiency and authenticity. While it undeniably offers a streamlined approach to content creation, concerns about the potential loss of human touch and genuine engagement cannot be ignored.

As the feature undergoes testing, it will be crucial for LinkedIn to address user concerns and ensure that the integration of generative AI aligns with the platform’s overarching goals. Striking the right balance will be key towards maintaining the value of LinkedIn as a professional networking platform, fostering genuine interactions, and empowering individuals to advance their careers effectively. Not too long ago, LinkedIn announced AI-generated cover letters, which was something along the same line.

RELATED:

• Streamlining Job Hunts: LinkedIn Tests AI-Generated Cover Letters for Premium Subscribers
• LinkedIn to shut down its China operations due to increasing regulatory hurdles
• Best VR Games 2023: The Ultimate List of Virtual Adventures and Experiences

(Via)

The post LinkedIn Introduces Generative AI for Writing Posts off Mere 30-word Prompts appeared first on Gizmochina.

It seems like everyone has been talking about nothing but generative AI for the past few years. In a promising piece of news for generative AI, the latest data released by the job search website Indeed indicates that the United States experienced a 20% month-on-month increase in the number of jobs related to generative AI in May this year. This surge indicates that an increasing number of American companies are eager to embrace and leverage this technology to drive their business growth.

The number of AI-related job postings doubled in the last two years

The data from May reveals that there are now 204 generative AI-related jobs for every million job postings in the United States, which is more than double the figure recorded in 2021. This exponential growth signifies a significant shift in the demand for talent with expertise in generative AI.

Among the various artificial intelligence positions posted on Indeed, data scientist roles make up 5% of the total. Additionally, positions such as software engineers, machine learning engineers, and data engineers are also highly sought-after by companies looking to capitalize on the potential of generative AI.

Nick Bunker, the Director of Economic Research at Indeed, provided his perspective on this trend, stating, “After the global explosion of ChatGPT, job seekers’ interest in artificial intelligence-related jobs has increased significantly.” The rise of generative AI models like ChatGPT has undoubtedly sparked a surge in curiosity and interest among job seekers in pursuing careers in the field of artificial intelligence. However, despite the growing interest from job seekers, Indeed reports that overall tech job opportunities in the United States have decreased by 43.6% compared to June of the previous year. The rapid increase in the number of available AI jobs has not kept pace with the rising demand from job seekers, suggesting a potential skills gap in the market.

As American companies recognize the transformative potential of generative AI, the need for skilled professionals in this field will continue to grow. Bridging the skills gap and fostering a talent pool capable of meeting the demand for generative AI expertise will be crucial, to say the least.

RELATED:

• Razer Skins aim to give dbrand a Run for Its Money
• Chinese Cyberspace Administration Announces New Categorization and Legal Framework for AI Services
• Best Air Fryers in 2023

(Via)

The post The Rise of Generative AI Jobs Signals Growth for American Companies appeared first on Gizmochina.

In a recent interview with GRAMMY.com, Recording Academy CEO Harvey Mason Jr. shed light on the decision-making process behind the rule amendments for the 2024 GRAMMY Awards. Mason explained how these changes were implemented to ensure that the GRAMMYs adapt to emerging technologies and create a level playing field for artists worldwide.

AI Elements and Instrumentation can still be considered for Awards

One significant area of focus in the new GRAMMY Awards rules is the inclusion of artificial intelligence (AI). The guidelines explicitly state that only human creators are eligible for consideration, nominations, and wins. Mason emphasized the importance of accounting for AI’s impact on human artistic expression, as it is poised to shape the future of the music industry.

Mason acknowledged the uncertainty surrounding the full extent of AI’s influence and expressed concerns about potential challenges it may pose. To address this, the Recording Academy has taken proactive steps, including hosting a summit with industry leaders, tech entrepreneurs, streaming platforms, and artists, to discuss the role of AI and the future of music.

Regarding the protocols instituted for the 2024 GRAMMYs, Mason clarified that while music made with generative AI and content can be submitted, the awards will only be granted to human creators who have made significant creative contributions in the appropriate categories. AI-generated voice or instrumentation will be considered, but in songwriting-based categories, human involvement must be predominant. Similarly, performance categories will exclusively feature human performers. Mason explained that while AI may contribute to the songwriting or music creation process, the ultimate recognition and GRAMMYs will be reserved for human creators.

These new rules and guidelines demonstrate the Recording Academy’s commitment to embracing change while upholding the integrity of artistic expression. By considering the impact of AI and creating a framework that acknowledges human contributions, the GRAMMYs aim to navigate the evolving landscape of the music industry and ensure a fair and inclusive platform for all artists.

As the 2024 GRAMMY Awards approach, the Recording Academy’s proactive approach and dedication to staying ahead of technological advancements showcase their commitment to recognizing and celebrating the artistry and creativity of human musicians in an ever-changing musical landscape.

RELATED:

• The Godfather of AI, Geoffrey Hinton, Reveals the Future of Artificial Intelligence
• AI Is Playing an Increasingly Important Role in SoftBank’s Business, Says CEO Masayoshi Son
• Best Password Managers of 2023: Top 5 Picks

(Via)

The post Here is the GRAMMY’s Viewpoint on AI winning Awards for AI-Generated Music appeared first on Gizmochina.

As AI slowly becomes a staple in media and entertainment, another interesting piece of information has popped up. Paul McCartney, the legendary member of the Beatles, has announced that the band will be releasing a new single this year with the help of artificial intelligence (AI). Set to mark McCartney’s 81st birthday, this anticipated release has been dubbed the band’s “last song” – a testament to their fantastic legacy.

AI allows users to extract an Artist’s voice to use as a dataset for new Voice Models

Utilizing cutting-edge AI technology, McCartney shared that they have successfully extracted the late John Lennon’s voice and continued work on an unfinished single from the late 1970s. This announcement follows the Beatles’ previous releases of “Free as a Bird” and “Real Love” in the mid-90s, which were also crafted from Lennon’s incomplete compositions.

Advancements in AI have now opened up unprecedented possibilities for artists who are looking to layer new vocals alongside old vocal samples from other artists. With the ability to alter melodies and lyrics after extracting Lennon’s voice, the production team claims to have harnessed the true essence of his artistry, bringing his long-lost work to life.

As speculation mounts among devoted Beatles fans, one particular track has piqued their attention. This track is “Now and Then,” a demo from the late 1970s that was believed to have been sent by John Lennon’s widow, Yoko Ono, to McCartney in 1994. “Now and Then” holds great potential as the new single. However, McCartney has revealed that the song was not included in the compilation due to the band’s guitarist, George Harrison not exactly liking the song.

Reflecting on the band’s iconic yet tragic history, it is sad to note that John Lennon was assassinated in 1980, while George Harrison succumbed to cancer in 2001. Today, only McCartney and 83-year-old drummer Ringo Starr remain as the sole surviving members of the globally known band.

RELATED:

• The Godfather of AI, Geoffrey Hinton, Reveals the Future of Artificial Intelligence
• Meta’s new MusicGen AI can Generate Music from Prompts, just like ChatGPT
• Best Gaming Mouse in 2023 under INR 1000

(Via)

The post The Beatles to Release Last Song with AI, A Tribute to John Lennon appeared first on Gizmochina.

Oracle, the American commercial software giant, released its financial report for the fourth quarter of the 2023 fiscal year, showcasing robust growth. The company’s stock price soared to a record high at the close of trading and continued to rise by 3.5% in after-hours trading. In the midst of an ongoing tech recession in a lot of parts of the world, this is a major achievement.

Oracle’s Cloud Computing business had the biggest growth numbers

According to the financial report, Oracle’s total revenue for the quarter reached $13.8 billion, marking a 17% year-on-year increase. This surpassed analysts’ average forecast of $13.7 billion, demonstrating Oracle’s ability to surpass expectations. The net income, excluding special income, stood at $1.67 per share, outperforming analysts’ projection of $1.58 per share. Oracle’s cloud computing business specifically experienced impressive growth, with fourth-quarter revenue reaching $4.4 billion, a 54% increase compared to the previous year. The cloud infrastructure segment contributed $1.4 billion in revenue, growing by 17%, while cloud software sales rose by 45% to $3 billion.

Oracle CEO Safra Catz expressed confidence in the company’s future, stating that they anticipate that the company will surpass the current growth rate of the previous fiscal year in the upcoming year. Oracle has strategically expanded its cloud computing infrastructure to compete with industry leaders like Amazon and Microsoft. While the growth rates of these companies have slowed down, Oracle’s cloud computing business has thrived. The increasing demand for generative artificial intelligence in the market has significantly boosted Oracle’s cloud computing business. As evidence, generative AI startup Cohere recently secured $270 million in funding, with Oracle being one of the investors.

Following the release of the financial report, Barclays Bank analyst Raimo Lenschow predicted that Oracle’s stock price would continue to rise. The stock closed at a record high of $116.43 per share, representing a 42% increase since the beginning of the year.

RELATED:

• Nokia, Ericsson, Sony, and Oracle pull out of Mobile World Congress 2021
• Meta CEO Mark Zuckerberg Predicts the Arrival of Generative AI to All Their Products
• The Best Canon Cameras in 2023

(Via)

The post Oracle Reports Strong Fourth Quarter Results, Stock Price Hits Record High appeared first on Gizmochina.

AI can create music from prompts using a combination of machine learning techniques and a vast dataset of pre-existing music. When given a text description of the desired music style, the AI algorithm analyzes patterns, structures, and characteristics from the dataset to generate a musical composition that aligns with the provided instructions. It learns the relationships between various musical elements such as melodies, harmonies, rhythms, and instrumentations and uses this knowledge to generate an original piece of music that reflects the desired style, mood, and genre. Through iterative training and fine-tuning, the AI becomes adept at creating compositions that resemble and emulate the musical styles it has learned from the dataset, offering a unique and AI-generated musical experience based on the given prompt. Now, Meta has done something of this sort, creating an AI that generates music based on prompts.

Meta’s MusicGen Library includes 10,000 high-quality Licensed Tracks

Considered the musical rendition of ChatGPT, MusicGen allows users to enter a brief text description of the type of music they want and the AI generates a 12-second track that aligns with their instructions. Although it may not replace the talents of musicians just yet, MusicGen offers a unique experience in the field.

While MusicGen’s instrumental capabilities are impressive, it’s important to note that the AI currently lacks vocal synthesis. MusicGen focuses solely on instrumentals (for now). This, however, doesn’t undermine the tool’s potential for creating decent background audio for videos or presentations. Who knows, a future update to MusicGen can include an option to generate vocals as well.

Meta’s commitment to legality is evident in their use of 10,000 high-quality licensed music tracks, combined with royalty-free songs from platforms like Shutterstock and Pond5. However, the growing popularity of generative AI has raised concerns about copyright infringement. Artists and labels have begun pursuing legal action against developers and platforms for the unauthorized use of their works. Meta may soon find itself in the crosshairs of these frustrated organizations as well. We’ll see how this fares for the company in the near future.

RELATED:

• Meta CEO Mark Zuckerberg Predicts the Arrival of Generative AI to All Their Products
• Apple Vision Pro sets modest sales target of 150,000 units amidst tough competition from Meta Quest
• Best lightweight and durable travel backpacks with smart organization features in 2023

(Via)

The post Meta’s new MusicGen AI can Generate Music from Prompts, just like ChatGPT appeared first on Gizmochina.

Generative AI is a class of artificial intelligence techniques and models that are designed to generate new and original content, such as images, text, music, or even videos. It involves training algorithms on large datasets and enabling them to learn patterns and structures within the data, allowing them to create new instances that resemble the training examples. Generative AI algorithms can produce creative outputs by combining and extrapolating from the learned knowledge, providing valuable tools for tasks like content generation, artistic expression, and data augmentation, among others. Dall-E 2 and Midjourney are great examples of generative AI. CEO of Meta, Mark Zuckerberg recently stated that Generative AI will soon be available in every single one of our devices.

All Meta products, such as Whatsapp, Instagram & Messenger will get this functionality

Recently, at a companywide all-hands meeting, Zuckerberg shed more light on Meta’s plans regarding generative AI. He provided employees with an overview of how the company intends to incorporate generative AI “into every single one of our products.” One of the proposed features involves the introduction of AI “agents” for popular platforms like WhatsApp and Messenger. This concept has been previously mentioned by Zuckerberg himself. Notably, details of the Instagram AI chatbot were discovered recently, offering users a choice of up to 30 distinct “personalities.”

Furthermore, Meta is reportedly developing generative AI photo editing capabilities for Instagram. This feature would enable users to modify their images using text prompts and subsequently share the edited photos with their followers. Zuckerberg has also hinted at the creation of post generation tools for Facebook, as well as for the platform’s advertisers, during the company’s recent earnings call.

The exact timeline for the release of these features remains uncertain. Nevertheless, Zuckerberg seems eager to see them implemented sooner rather than later. In addition to driving internal development, Meta has organized an internal hackathon, encouraging employees to propose innovative ideas for new generative AI features.

RELATED:

• Meta brings paid verification subscription for Facebook & Instagram in India
• Google is Killing the Feature from Instagram That It Stole from Snapchat
• Best Video Doorbells with a wide view for comprehensive coverage in 2023

(Via)

The post Meta CEO Mark Zuckerberg Predicts the Arrival of Generative AI to All Their Products appeared first on Gizmochina.