Leonardo can create detailed AI images, but lacks the wow factor. Leonardo AIis no Leonardo da Vinci or DiCaprio, but it’s still an image generator that falls into the artistic Leonardo category. Originally designed to help create gaming assets, it’s now a full-fledged AI content creation service that offers AI video creation and editing services in addition to image tools.
Overall, Leonardo is a good choice compared to many of itsAIcompetitors. It’s on par with Adobe Firefly and much better than Google’s ImageFX or Canva. Leonardo follows prompts better than Midjourney, but the lack of extensive editing tools makes it hard to choose between the two. OpenAI’s Dall-E 3 is still CNET’s top-ranked choice, but you’ll need to pay $20 for ChatGPT Plus, while Leonardo has a comprehensive free plan. I’ve used Leonardo to generate more than 90 images, ranging from stock images to sci-fi and fantasy renderings. Here’s the full process:
How CNET tests AI image generators
CNET takes a hands-on approach to reviewing AI image generators. Our goal is to determine how it compares to the competition and what applications it’s best suited for. To do this, we provide AIprompts based on real-world use cases, such as rendering in a specific style, combining elements into a single image, and handling long descriptions. Image generators are rated on a 10-point scale, taking into account factors like how well the image matches the prompt, the creativity of the result, and responsiveness. Learn more about how we test AI
Leonardo’s images are so attractive that we encourage you to try Leonardo’s other AIcreation tools, such as Canvas Editor and Live Generate. However, we recommend that you don’t use it. These programs are less user-friendly and produce lower-quality content that’s blurry, off-center, or has strange quirks. Now that better image editing software is available, Meta AI’s “Imagine” feature is a more accurate live image generation tool.
Leonardo’s paid version, Alchemy Refiner, promises “improvements and enhancements” to images that AI image generation struggles with, especially faces and hands. Since I’m a free user, I couldn’t test it myself, but I was impressed by the clarity and accuracy of human hands and teeth compared to other AI generators. How long does it take to receive an image?
Images are generated in 10-20 seconds, making Leonardo one of the fastest AI image generation tools. Image generation time varies depending on the model used. For example, the new Phoenix model takes longer. But with Phoenix, you don’t have to scroll your phone or check email while waiting for the image to load like you do with other generators.
Leonardo is great, but I’m not surprised.
For AI creators, Leonardo checks a lot of important boxes. It’s fast, has a free plan, and the images it creates look completely normal. However, there are a few reasons not to recommend it to everyone. The paid post-editing tools are cumbersome and will quickly drain your tokens to get what you need. Important parts of the privacy policy are hidden in the terms of service and leave a lot to be desired. From a quantitative perspective, I wasn’t surprised by the results. It felt average. Of course, there’s nothing wrong with that. It’s more than an alternative to the current top competitors. For non-professional creators and AI creative enthusiasts, Leonardo is great for making usable (if not perfect) AI images quickly and easily.
Category: news
Microsoft releases Windows 11 artificial intelligence roadmap: smart search, upgrades, and more
What’s next for Windows? Microsoft may have just released the Windows 11 2024 Update (24H2), but the company has already revealed its plans for the next generation of Windows apps — and there are some very interesting AIfeatures coming before the holidays.
Microsoft has revealed that it is working on several AI features for Windows and Windows apps: improved Windows Search using natural description language, super resolution in Photos, generative fill and erase in Paint, and the debut of Recall. All features (except Recall) will appear as part of the Windows Insider program in October, with an expected launch in November All of these features will rely on the NPU inside Copilot+ PCs, which will now include PCs with Qualcomm Snapdragon X Elite processors as well as AMD’s Ryzen AI 300 and Intel’s Lunar Lake. Microsoft is also planning to launch more Copilot features that will run in the cloud, including Copilot Voice and Copilot Vision, similar to innovations used in rival AI services. The timing of these new features rolling out will vary by platform, though, as Snapdragon X PCs have been shipping for a few months; Microsoft will bring support to AMD and Intel Copilot+ PCs with its own updates. Microsoft has revealed more details about the improvements to Recall, and the company now says the feature can be bypassed when setting up a new PC or removed later. Windows Recall takes your screenshots from time to time, extracts the data, and then stores it in case you need it later. The feature has come under fire for violating user privacy and being unsafe. Now, Microsoft says it stores the screenshots and extracted data used in Recall in an encrypted area. Security researchers previously said the data was stored unencrypted. New AI features coming to Copilot+ PCsMicrosoft says it plans to improve search on PCs by using more natural language when searching for files on the PC. You may have seen this feature in apps like Microsoft Photos or Google Photos; for example, if you search for “beach,” the apps will use artificial intelligence to identify beach scenes. Microsoft will bring the same technology to File Explorer, but it’s not clear what folders or files they’ll apply to.
The improved Windows Search seems to be more context-aware than before: “BBQ party” is listed as an example search term in the demo below. “You no longer have to remember file names, settings locations, or even worry about spelling — just type what’s in your head to find it on your Copilot+ PC,” Microsoft says. However, it seems unlikely that you’ll be able to find a specific .ini file in your user folder as easily as you can find your aunt’s wedding photos.
The improved search feature will start in File Explorer and then expand to Windows Search and Settings in the “coming months.” “Super Resolution” in Photos is probably my favorite potential application for a few reasons: a.) I have a lot of old photos taken with old, low-quality digital cameras; and b.) Journalists often receive low-resolution photos that need to be enlarged or blown up before they can be published. Regardless, the new “Super Resolution” feature will hopefully solve these problems.
Microsoft announced the Auto Super Resolution feature to improve its gaming capabilities, but Photo Super Resolution seems more practical. Many websites and apps promise to offer upgrades, and it’s unclear whether this new app will surpass them. Photo Super Resolution will be free, though. Microsoft says that using Copilot+ PC’s AITOPS, you’ll be able to increase resolution by eight times. Super resolution will be part of the photo, which can already automatically adjust lighting and tones, remove backgrounds, add generative elements, and more.
Shop Laptop Battery the Online Prices in UK – Large Selection of Batteries and Adapters from batteryforpc.co.uk With Best Prices.
Train your brain for creative work with Gen AI
There are countless articles on how to use generative artificial intelligence (gen AI) to improve work, automate repetitive tasks, summarize meetings and client interactions, and synthesize information. There are also vast virtual libraries filled with tips and guides that can help us achieve more effective or even superior output with gen AI tools. Many common digital tools already come with integrated AIco-pilots that automatically enhance and complete writing, coding, designing, creating, and whatever you’re working on. But generative AI does more than just enhance or accelerate what we already do. With the right mindset shift, we can train our brains to creatively rethink how to use these tools to unlock entirely new value and achieve exponential results in an AI-first world.
Generative AI relies on natural language processing (NLP) to understand requests and generate relevant results. It is basically pattern recognition and pattern assembly based on instructions to provide output that accomplishes the task at hand. This approach fits with our brain’s default mode: pattern recognition and the pursuit of efficiency, which favors short, direct prompts for immediate, predictable results.
If most people use AI in this way, no matter how powerful these tools are, we will inadvertently create a new status quo in the way we work and create. Training our brains to challenge our thinking, our assumptions about AI’s capabilities, and our expectations for predictable results starts with a mindset shift to recognize that AI is not just a tool, but a partner in innovation and exploration of unknown territory.
Rethinking Collaboration with AI for More Creative and Innovative OutcomesChanging your mindset to collaborate with AI in a more creative and open way means being willing to explore unknown territory and having the ability to learn, unlearn, and experiment. Plus, it’s fun.
Insight Center SeriesCollaboration with AIHow humans and machines can best work together.
I often say that I maximize the potential of AI and achieve the best results when I put aside my cognitive biases. With a smile on my face, I ask myself, “WWAID?” or “What would AI do?” I acknowledge that the way I unconsciously use AI tools may default to predictable inputs and outputs. But by asking WWAID, I open myself up to new interactions and experiences that may yield unexpected results.
Tapping into AI’s creative and transformative potential, and training your brain for an AI-first world, requires us to shift our prompting approach to thinking of AI as a partner, not just a tool.
12 Exercises to Train Your Brain to Work More Creatively with AI
Here are a dozen ways to train our brains to achieve broader, more innovative outcomes with AI:
1. Practice “exploratory prompts” every day
Start each day with an open-ended prompt that pushes you to think boldly. Try asking yourself, “What trends or opportunities are there in my industry that I don’t see coming?” or “How can I completely redefine my approach to key challenges?”
2. Create prompts around “what if” and “how can we” questionsInstead of asking direct questions, ask open-ended possibilities. For example, instead of asking “How can I be more efficient?”, try asking “If I could be more efficient in an unconventional way, what would that look like?”
3. Embrace ambiguity and curiosity in promptsBy training ourselves to prompt without a clear endpoint, AI can generate answers that may surprise us. Prompts like “What might I have overlooked in approaching X?” can open doors to insights we never considered.
4. Use prompts to explore rather than solve problemsMany prompts focus on solutions. Shifting to exploration can yield deeper insights. For example, “Let’s explore what the future of leadership would look like if AI had a seat at the board or C-suite — how would our jobs, roles, and corporate culture change?”
5. Chain prompts to develop ideas iterativelyDon’t stop at the first answer, ask follow-up questions that make the answer more complex and visionary. If the AI comes up with an idea, build on it with questions like “What will it look like in 5 years?” or “How could this approach change the way the company operates in the future?”
6. Think in metaphors or analogiesTraining our brains to use metaphors or analogies in prompts can open up creative avenues. For example, instead of asking for a product
OpenAI’s transcription tool can produce hallucinations. Hospitals are still using it
In healthcare, accuracy is important. So the widespread use of OpenAI’s Whisper transcription tool among medical professionals has raised alarms among experts.OpenAI’s Whisper transcription tool has fabricated false text in medical and business settings, despite warnings against it. The Associated Press interviewed more than a dozen software engineers, developers and researchers who found that the model often fabricated text that the speaker never said, a phenomenon often referred to as “fictitious” or “hallucination” in the field of artificial intelligence.
When it was released in 2022, OpenAI claimed that Whisper was close to “human-level robustness” in audio transcription accuracy. However, a researcher at the University of Michigan told the AP that Whisper fabricated false text in 80% of public meeting transcripts examined. Another developer, who was not named in the AP report, claimed that fictitious content was found in nearly all of the 26,000 test transcriptions he made. These fabrications are particularly dangerous in medical settings. More than 30,000 medical workers now use Whisper-based tools to record patient visits, despite OpenAI’s warnings against using Whisper in “high-risk areas,” according to the Associated Press. Minnesota’s Mankato Clinic and Children’s Hospital Los Angeles are among 40 health systems that use a Whisper-based AI Co-Pilot service developed by medical technology company Nabla that’s fine-tuned for medical terminology. Nabla acknowledges that Whisper can fabricate conversations, but it also reportedly deletes the original recordings for “data security reasons.” This could raise other issues, as doctors can’t verify the accuracy of the original material. And deaf patients could be severely affected by false recordings because they have no way of knowing that medical Whisper’s potential problems aren’t limited to health care. Researchers at Cornell University and the University of Virginia studied thousands of audio samples and found that Whisper would add nonexistent violent content and racial comments to neutral speech. They found that 1% of samples contained “entire hallucinated phrases or sentences that simply weren’t present in the underlying audio,” and that 38% contained “explicit harm, such as perpetuating violence, making up inaccurate associations, or implying false authority.”
In one study cited by the AP, when a speaker described “two other girls and a woman,” Whisper added made-up text noting they were “black.” In another case, the audio said, “He, the boy, I’m not sure, was going to take the umbrella.” Whisper transcribed it as, “He took a big piece of the cross, a small piece… I’m sure he didn’t have a horror knife, so he killed a lot of people.”
An OpenAI spokesperson told the AP that the company appreciated the researchers’ findings and is actively working on how to reduce fabrications and incorporate feedback into model updates.
OpenAI Dall-E 3 review: Generative AI for creating fantastical and fun illustrations
One of the earliest text-to-image tools, it’s the best we’ve tested. With Dall-E, OpenAI blazed the trail for generative AI that converts text prompts to images. The competition is stiffer now, but version 3 of the service still holds up.
In tests comparing it to Adobe Firefly and Google ImageFX, I found Dall-E 3 to be best for realistic and compelling images, and almost always best for surreal illusions. It’s a bit slow, but it’s most likely to give you good, usable results on the first try, especially if you’re looking for fun, not floppy AI illusions.
Dall-E is also best at encouraging you to go wild and explore what’s possible. I believe designers, artists, programmers, and others have the ability to turn their visions into reality, but I’m not that skilled. So for me, Dall-E is a wonder.
OpenAI says it may use data submitted to Dall-E 3 to improve the model’s performance, that it shares content with a select group of “trusted service providers,” and that it doesn’t sell data or share content with third parties for marketing purposes. You can also submit a privacy request to have OpenAI stop using your data for training or delete your account. For more information, see OpenAI’s general privacy FAQ and main privacy policy.
Here are my further findings about Dall-E 3.
How CNET tests AI image generators
CNET takes a practical approach to reviewing AI image generators. Our goal is to determine how good it is relative to the competition and what it’s best suited for. To do this, we give the AI prompts based on real-world use cases, such as rendering in a specific style, combining elements into a single image, and handling long descriptions. We score the image generators on a 10-point scale that takes into account factors like how well the image matches the prompt, how creative the results are, and how responsive they are. Check out How We Test AI to learn more.
How good is the image? How well does it match the prompt?
ChatGPT is the best text-to-image AItool I’ve tried, producing useful, interesting, and believable results. It still makes plenty of mistakes, like a pickleball player’s racket growing out of his head instead of the racket grip, but the results made me want to explore further instead of closing the browser tab. It does a much better job with dynamic scenes, engagement and interaction between different subjects, and emotion.
ChatGPT is a big part of Dall-E. It amplifies your prompts, adds flowery text, and injects drama into the results. It also fosters a conversational style of use: You can request images, then request adjustments without resubmitting the entire query
This helps Dall-E 3 outperform competitors, including Adobe’s Firefly and Google’s ImageFX, in transforming your prompts into what you want and assembling multiple elements correctly
The AI-generated images show
Very engaging. Dall-E 3 produces vivid, compelling images over and over again. Even with the issues, I often enjoyed them. They sometimes made me laugh and observe the details.
Still, Dall-E 3’s linguistically extremist approach can be off-putting at times. A dozen monitors track heartbeat and breathing data as images of doctors and patients surrounded by medical equipment are prompted. One of the computers has a keyboard with about 100 keys.
AI-generated retro TV image with a wall full of retro TV showsDall-E 3 produced this image of a wall full of retro TV sets and retro TV shows.
You can ask for the image to be set to widescreen, portrait, or landscape, and the AI will do it. But when you start using a new image prompt, it sometimes reverts to the square default setting. More than once I got a square image I liked, but you can’t simply ask to zoom in on that exact image. (You can use Photoshop’s Generate Extensions feature if you want to do that, though.)
How quickly do the images arrive?I guess there’s always a benefit to waiting. The Dall-E 3 usually takes 20 or 30 seconds to take a picture. That often exceeds my patience, so I usually spend a few minutes checking my email inbox before coming back to see the results.
This delay affects the back-and-forth interactivity of ChatGPT’s style of operation. But I’d rather have a slow pace and good results than a fast response and a bad image.
GenerativeAI pushes computing technology to its limits. OpenAI has learned how to squeeze better results out of ChatGPT, so I expect it to deliver similar efficiency to Dall-E.
ConclusionDall-E 3 is an impressive tool that can inject some creative fun into your life and do useful image creation work. Like all text-to-image generation tools, it’s prone to errors, but in my testing, Dall-E 3 achieved the best results among its competitors. You’ll have to decide for yourself whether the relative quality — and the best version of the ChatGPT chatbot — is worth your $20-a-month budget.
Shop Laptop Battery the Online Prices in UK – Large Selection of Batteries and Adapters from batteryforpc.co.uk With Best Prices.
Graph-based AI models chart the future of innovation
An AImethod developed by Professor Markus Buehler can discover hidden connections between science and art to recommend new materials. Imagine using AI to compare two seemingly unrelated creations—biological tissues and Beethoven’s Ninth Symphony. At first glance, living systems and musical masterpieces seem unrelated. However, a novel AI method developed by Markus J. Buehler, MIT’s McAfee Professor of Engineering, professor of civil and environmental engineering, and professor of mechanical engineering, bridges that gap, revealing common patterns of complexity and order.
“By combining generative AIwith graph-based computational tools, this approach reveals entirely new ideas, concepts, and designs that were previously unimaginable. We can accelerate scientific discovery by teaching generative AI to make novel predictions about ideas, concepts, and designs that have never been seen before,” Buehler says.
The open-access study, recently published in Machine Learning: Science & Technology, demonstrates an advanced AImethod that integrates generative knowledge extraction, graph-based representations, and multimodal intelligent graph reasoning.
The research used graphs developed using methods inspired by category theory as a core mechanism to teach the model to understand symbolic relationships in science. Category theory is a branch of mathematics that studies abstract structures and the relationships between them. It provides a framework for understanding and unifying different systems by focusing on objects and their interactions rather than their specific content. In category theory, systems are viewed as objects (which can be anything from numbers to more abstract entities such as structures or processes) and morphisms (arrows or functions that define the relationships between these objects). By using this approach, Buehler was able to teach theAI model to systematically reason about complex scientific concepts and behaviors. The symbolic relationships introduced through morphisms made it clear that the AI was not just making analogies, but was also engaging in deeper reasoning, mapping abstract structures to different domains.
Using this new approach, Buehler analyzed 1,000 scientific papers on biomaterials and transformed them into a knowledge graph in the form of a graph. The graph revealed connections between different information and was able to find relevant ideas and key points that tie many concepts together.
“What’s really interesting is that the graph follows the scale-free property and is highly connected, which can be effectively used for graph reasoning,” Buehler said. “In other words, we teach AIsystems to think about graph-based data to help them build better representations of the world and enhance their ability to think about and explore new ideas to enable discovery.”
In another experiment, a graph-based AImodel suggested making a new biomaterial inspired by the abstract patterns in Wassily Kandinsky’s painting Composition VII. The AI suggested making a new mycelium-based composite material. “This material combines a range of innovative concepts, including a balance of chaos and order, tunable properties, porosity, mechanical strength, and chemical functionality in a complex pattern,” Buehler noted. By drawing inspiration from abstract paintings, AI created a material that balances strength and functionality while being adaptable and able to perform different roles. This application could facilitate the development of innovative sustainable building materials, biodegradable plastic alternatives, wearable technology, and even biomedical devices.
An AImethod developed by Professor Markus Buehler can discover hidden connections between science and art to recommend new materials. Imagine using AI to compare two seemingly unrelated creations—biological tissues and Beethoven’s Ninth Symphony. At first glance, living systems and musical masterpieces seem unrelated. However, a novel AI method developed by Markus J. Buehler, MIT’s McAfee Professor of Engineering, professor of civil and environmental engineering, and professor of mechanical engineering, bridges that gap, revealing common patterns of complexity and order.
The research used graphs developed using methods inspired by category theory as a core mechanism to teach the model to understand symbolic relationships in science. Category theory is a branch of mathematics that studies abstract structures and the relationships between them. It provides a framework for understanding and unifying different systems by focusing on objects and their interactions rather than their specific content. In category theory, systems are viewed as objects (which can be anything from numbers to more abstract entities such as structures or processes) and morphisms (arrows or functions that define the relationships between these objects). By using this approach, Buehler was able to teach theAI model to systematically reason about complex scientific concepts and behaviors. The symbolic relationships introduced through morphisms made it clear that the AI was not just making analogies, but was also engaging in deeper reasoning, mapping abstract structures to different domains.
Using this new approach, Buehler analyzed 1,000 scientific papers on biomaterials and transformed them into a knowledge graph in the form of a graph. The graph revealed connections between different information and was able to find relevant ideas and key points that tie many concepts together.
“What’s really interesting is that the graph follows the scale-free property and is highly connected, which can be effectively used for graph reasoning,” Buehler said. “In other words, we teach AIsystems to think about graph-based data to help them build better representations of the world and enhance their ability to think about and explore new ideas to enable discovery.”
In another experiment, a graph-based AImodel suggested making a new biomaterial inspired by the abstract patterns in Wassily Kandinsky’s painting Composition VII. The AI suggested making a new mycelium-based composite material. “This material combines a range of innovative concepts, including a balance of chaos and order, tunable properties, porosity, mechanical strength, and chemical functionality in a complex pattern,” Buehler noted. By drawing inspiration from abstract paintings, AI created a material that balances strength and functionality while being adaptable and able to perform different roles. This application could facilitate the development of innovative sustainable building materials, biodegradable plastic alternatives, wearable technology, and even biomedical devices.
“Graph-based generative AI is more innovative, exploratory, and technically detailed than traditional methods, and establishes a broadly useful innovation framework by revealing hidden connections,” said Buehler. “This research not only contributes to the field of biomimetic materials and mechanics, but also lays the foundation for future interdisciplinary research driven by AI and knowledge graphs to become a tool for scientific and philosophical inquiry. We look forward to more research results in the future.
Graph-based AI models chart the future of innovation
Imagine using artificial intelligence to compare two seemingly unrelated creations—biological tissue and Beethoven’s Ninth Symphony. At first glance, living systems and musical masterpieces seem unrelated. However, a novel AI approach developed by Markus J. Buehler, MIT’s McAfee Professor of Engineering and professor of civil and environmental engineering and professor of mechanical engineering, bridges that gap, revealing common patterns of complexity and order.
“By combining generative AIwith graph-based computational tools, this approach reveals entirely new ideas, concepts, and designs that were previously unimaginable. We can accelerate scientific discovery by teaching generative AI to make novel predictions about ideas, concepts, and designs that have never been seen before,” Buehler says.
The open-access study, recently published in Machine Learning: Science & Technology, demonstrates an advancedAIapproach that integrates generative knowledge extraction, graph-based representations, and multimodal intelligent graph reasoning.
The study uses graphs developed using methods inspired by category theory as a core mechanism to teach models to understand symbolic relationships in science. Category theory, a branch of mathematics that studies abstract structures and the relationships between them, provides a framework for understanding and unifying disparate systems by focusing on objects and their interactions rather than their specific content. In category theory, systems are viewed as objects (which can be anything from numbers to more abstract entities like structures or processes) and morphisms (arrows or functions that define the relationships between these objects). By using this approach, Buehler was able to teach the AImodel to systematically reason about complex scientific concepts and behaviors. The symbolic relationships introduced through morphisms made it clear that the AI was not just making analogies, but was also engaging in deeper reasoning, mapping abstract structures to different domains.
Using this new approach, Buehler analyzed 1,000 scientific papers on biomaterials and transformed them into a knowledge graph in the form of a graph. The graph revealed connections between disparate information and was able to find relevant ideas and key points that tie many concepts together.
“What’s really interesting is that the graph follows the scale-free property and is highly connected, which can be effectively used for graph reasoning,” Buehler said. “In other words, we teach AIsystems to think about graph-based data to help them build better representations of the world and enhance their ability to think about and explore new ideas to enable discovery.”
Researchers can use the framework to answer complex questions, discover gaps in current knowledge, propose new material designs, predict how materials will behave, and connect concepts that have never been connected.
The AI model found unexpected similarities between biomaterials and the Ninth Symphony, suggesting that both follow complex patterns. “Similar to the way cells in biomaterials interact in complex but organized ways to function, Beethoven’s Ninth Symphony arranges notes and themes to create a complex but coherent musical experience,” Buehler said.
In another experiment, a graph-based AI model suggested making a new biomaterial inspired by the abstract patterns in Wassily Kandinsky’s painting Composition VII. The AI suggested making a new mycelium-based composite material. “This material combines a range of innovative concepts, including a balance of chaos and order, tunable properties, porosity, mechanical strength, and chemical functionality in a complex pattern,” Buehler noted. By drawing inspiration from abstract paintings, AI created a material that strikes a balance between strength and functionality while also being adaptable and able to perform different roles. The application could facilitate the development of innovative sustainable building materials, biodegradable plastic alternatives, wearable technology, and even biomedical devices.
With this advanced AI model, scientists can draw insights from music, art, and technology, analyzing data from these fields to identify hidden patterns that could lead to endless innovative possibilities for material design, research, and even music or visual art.
“Graph-based generative AIis more innovative, exploratory, and technically detailed than traditional methods, and establishes a broadly useful innovation framework by revealing hidden connections,” said Buehler. “This research not only contributes to the field of biomimetic materials and mechanics, but also lays the foundation for future interdisciplinary research driven by AI and knowledge graphs to become a tool for scientific and philosophical inquiry. We look forward to more research in the future.
Shop Laptop Battery the Online Prices in UK – Large Selection of Batteries and Adapters from batteryforpc.co.uk With Best Prices
Decart’s AI simulates a real-time, playable version of Minecraft
Decart, an Israeli AIcompany that went public today with $21 million in funding from Sequoia Capital and Oren Zeev, has released what it says is the first AI model that can run in an “open world.”
The model, called Oasis, is available for download and trial on the Decart website: It’s a Minecraft-like game that generates end-to-end games on the fly. Trained on Minecraft gameplay videos, Oasis can generate keyboard and mouse movement frames in real time, simulating physics, rules, and graphics. Oasis belongs to an emerging class of generative AI models called “global models.” Many of these models can simulate games, but few can match Oasis’ frame rates.
I tried the demo out of curiosity, but I think it has a long way to go before it can be a truly fun experience. The resolution is fairly low, and Oasis tends to “forget” the layout of a level quickly – I turned my character around only to see the rearranged scene.
Decart has added new features, however, such as the ability to upload images to create custom “worlds.” Future versions of Oasis will reportedly be optimized for Etched’s upcoming AI acceleration chip (the demo currently runs on an Nvidia H100 GPU) and will be able to generate gaming footage at up to 4K.
“These models could even improve modern entertainment platforms by dynamically generating content based on user preferences,” Decart wrote in a blog post. “Or it could be a gaming experience that offers new possibilities for user interaction, such as text and audio… prompts that guide the game.”
I’d like to know more about the copyright issue. Decart doesn’t claim to have a license from Microsoft to use the Minecraft videos for training purposes. (Microsoft owns Minecraft.) Is Oasis essentially creating an unauthorized copy of Minecraft? That’s for the courts to decide.
Shop Laptop Battery the Online Prices in UK – Large Selection of Batteries and Adapters from batteryforpc.co.uk With Best Prices.
OpenAI adds search capabilities to ChatGPT to challenge Google
AI developers’ chatbots can now search and summarize SAN FRANCISCO —AI giant OpenAI has announced a major overhaul of ChatGPT, enabling the chatbot to search the web and provide answers based on what it finds.
The upgrade changes the experience of using the popular chatbot. It puts OpenAI in more direct competition with Google, offering an alternative way to find and use information online.
What it looks like: Some ChatGPT queries will now be answered with the help of web search results. AI-generated text includes links to sources in a gray box at the end of the sentence and in a list of citations at the bottom or right.How it works: Search results are shown only when ChatGPT’s algorithm determines they’re relevant, or when you click the globe icon in the prompt box.You can reply to ChatGPT search results to refine your query.
What’s missing: Product shopping links and ads (at least for now).Who will get it: Initially only available to paid ChatGPT Plus and Team users, but OpenAI says it plans to open it to free users eventually.
Starting Thursday, paid subscribers to ChatGPT will be able to activate a mode that lets the AI tool respond to queries by searching the web for the latest information and summarizing what it finds, rather than providing answers based on potentially outdated data used to create the chatbot.
The search feature is powered by Microsoft’s Bing search engine, an OpenAI backer. It also draws on articles from publishers that have signed deals with the AI developer, such as Wall Street Journal owner News Corp. Google added AI-generated summaries and quotes to its traditional search results this year in response to growing competition from chatbots. Startup Perplexity offers a similar AI-enhanced search engine and has received more than $400 million in funding, according to venture capital research firm Pitchbook.
The changes will make ChatGPT more useful and accurate, OpenAI head of media partnerships Varun Shetty said in an interview. “We think because it’s online, it will increase reliability and reduce illusions.”
In addition to helping users find useful information, chatbots and AI search engines are beginning to fundamentally change the online economy. For more than a decade, Google has been the web’s main gateway, making news outlets, bloggers and other publishers dependent on people clicking on search results to sell advertising or subscription traffic. AI-powered search tools offered by Google, OpenAI and others can summarize web pages and answer questions directly, helping people find information without clicking on other sites. That has raised growing concerns among publishers who fear they are being left behind by tech companies. Some publishers have accused AI developers of unfairly copying and plagiarizing their content to build AI tools that upend their industries. Some news organizations have sued OpenAI, alleging copyright infringement.
ChatGPT’s search overhaul will be a test of how AI search engines can impact publishing.
Queries that trigger the web search feature will deliver AI-generated paragraphs of text followed by links to websites that summarize the content.
ChatGPT’s paying users will use the new search feature during and in the days following the Nov. 5 presidential election, a period when disinformation researchers say the web will be rife with lies and political falsehoods. The AI’s search tool will direct queries related to the election results to sources such as the Associated Press and Reuters. “We are trying to prioritize and promote the highest quality and authoritative sources we can,” Shetty said.
Shop Laptop Battery the Online Prices in UK – Large Selection of Batteries and Adapters from batteryforpc.co.uk With Best Prices
Arc Search AI mobile browser confirmed to be launching Android app soon
The artificial intelligence (AI ) mobile browser Arc Search is coming to Android, the company confirmed on Saturday. The browser made its debut early this year and has gained popularity for its AI features such as Browse for Me, where AI reads multiple web pages to display relevant information, AI summaries, and more. In May, the company released a Call Arc feature that offers two-way communication where users can verbally ask queries and the AI responds.
Arc Search confirms it is developing an Android appIn response to a post by a user on Threads who asked if the app is developing an Android version, the official Threads account of Arc Search said that an Android app is coming soon. This is the first time the company behind the app has officially confirmed that it is developing an Android app.
Arc Search was launched on the App Store by the Browser Company in January 2024. Later, the platform was expanded to Mac devices and a version for Windows was launched. However, the browser is still not available on Android. The browser has several unique features such as the Browse for Me mode where, upon receiving a search query,AI browses relevant web pages to find information in an easy-to-read format. It also comes with features like AI summarization, auto-archiving, ad blocking, private browsing mode, safe tags, reading mode, and more.
The company is also regularly adding new features to the app. Since its launch, the Arc Search browser has gained a “Share Browsing Me” mode and a “Call Arc” feature. The latter is a hands-free feature where users can talk to the app and the AI then answers the query. Notably, Arc Search uses the OpenAI API, along with several other AI models, to power the AI features in the browser. For now, the company has not revealed if there will be any Android-specific features. There is no word yet on when the browser will debut on the Android operating system.