Eliteviser

https://newfortech.com/?p=27192 This article is part of NewForTech’s AI Week 2025. We’ll cover the basics of AI and show you how to get the most out of ChatGPT, Gemini or Claude, as well as detailed features, news and key talking points in the AI ​​world. Most conversations about AI are about work. Productivity, email, spreadsheets, efficiency. All of this is useful, but it ignores the fact that technology should also be fun. AI is not just about saving time in the office; It can also make your free time more interesting. From creating playlists tailored to your mood to creating personalized puzzles and quizzes for a night out with friends, AI has become surprisingly effective in the field of personal entertainment. It’s like having a personal DJ, quiz expert and game designer all in one. 1. Music that really fits the moment We’ve all wasted time scrolling through Spotify to find the right playlist. There are millions of options, but none seem to create the exact atmosphere you want. Artificial intelligence solves this problem by creating playlists based on your preferences. Ask for a “two-hour playlist to prepare dinner on a rainy Sunday” and you’ll get a mix of songs to suit your mood. Want something more specific? Try “indie disco hits for a post-wedding party” or “lo-fi beats for late night”. The playlists aren’t random either; They are selected based on your taste, so they should seem personal and not generic. I tried it recently before traveling. I asked for “a six-hour playlist for a walk in the Scottish Highlands, a mix of classic rock and modern folk.” The result was terrifying. It featured artists I already loved, but also added new songs I’d never heard before. It felt like a playlist made by a friend rather than a generic Spotify mix. 2. Personalized puzzles to train your brain AI can also create puzzles tailored to your needs. Crosswords, word searches, Sudoku variants. Instead of choosing from what is published in the magazine, you can request challenges based on your interests. I tried to order a wrestling themed crossword puzzle, one of my guilty pleasures. The AI ​​created tracks and a grid within minutes. Some of the directions were a little vague, but they worked. The nice thing was that I had a puzzle built especially for me, not for a large audience. 3. Text adventure Artificial intelligence is not a replacement for real video games, but can generate light games for quick fun. Text adventures, choose your own road stories or even little logic puzzles. You provide the framework and the AI ​​inserts the content. I tested it by asking for a short text adventure in Tokyo. Within seconds I had a branching story where I could explore the streets, talk to locals and discover hidden plots. It wasn’t as fancy as a commercial game, but it was addictive enough to keep me hooked for about 30 minutes. The flexibility makes it exciting. You can customize environment, difficulty or themes. Want a crime novel set in your hometown? Completed. Do you want a fantasy game where your dog is the main character? ChatGPT will fix that. Think of it as the modern equivalent of your childhood “choose your own adventure.” 4. A questionnaire for each occasion Quizzes are one of the easiest ways to entertain a group, but creating them is a tricky task. Asking good questions, balancing challenges and avoiding repetition takes effort. ChatGPT does this immediately. Take a “Premier League History Test 20 Questions, Mixed Difficulty” and get a range of ready-made answers. Want something crazier? Request “a lighthearted and fun 30th anniversary pop culture quiz”. You can also ask it to create image sets by combining questions with AI-generated images. I tested it on friends by doing a “pub quiz with 15 questions, five on movies, five on sports, five on general knowledge.” The mix was good and the questions led to discussion. It wasn’t perfect, but it felt authentic enough to work for occasional occasions where I had to do a test. I wish I had ChatGPT during the covid-19 pandemic; This would have improved daily work tests 100 times! 5. Learning disguised as fun This article was originally published as part of ChatGPT Basics Guidenow available in. ChatGPT Basics Guide is packed with great tips and advice from experts, covering everything you need to know about the world’s most popular AI, from the basics to some of the latest features. AI tests can also serve as learning tools. If you are trying to learn a new topic, you can create practice questions. I tested it with simple Japanese sentences before my upcoming honeymoon and the AI ​​created a flashcard style quiz. It wasn’t glamorous, but it was effective. This approach also works for children. Parents can create spelling tests, math problems, or themed puzzles to make homework less difficult. The ability to customize the material makes it feel less like schoolwork and more like play. Personalization is the secret The common thread in everything you do with ChatGPT is personalization. Commercially available playlists, puzzles and quizzes are good, but they are generic. If you give the AI ​​enough information about your preferences or the task at hand, it can make your life incredibly easier. Without more information, the AI ​​is as generic as anything else you can find online. Instead of a “workout playlist,” you get “a workout playlist focused on 21st century hip-hop, no slow songs.” Instead of a “general knowledge quiz” you get “a quiz for five people who love Marvel movies and Formula 1”. That’s the fun part: you feel like the content is made for you, because it is. Things to think about As always, there are caveats. The AI ​​sometimes reuses templates, which can make quizzes repetitive if you don’t change the instructions. Music recommendations are only as good as the libraries they come from. And while AI can generate games, they are still newer than a complete replacement for commercial titles. ChatGPT will not be developing Grand Theft Auto VI anytime soon. But restrictions must not spoil the fun. See ChatGPT as a creative spark, not a finished product. You get the most out of it by combining your own ideas. How AI can balance work and play AI is refreshing because it removes a lot of the decision paralysis. Instead of wasting half an hour choosing a movie, ask the AI ​​for advice. Instead of burdening yourself with writing 20 test questions, you can complete them in seconds. Instead of searching for a playlist that matches your mood, just describe the mood and ask someone to create one for you. Entertainment should be fun, not effort. Artificial intelligence takes away the effort and the fun. ChatGPT is not just a business tool, it is a personalization engine that can make your free time more enjoyable. Whether you’re creating playlists to suit your mood, creating quizzes specific to your group, or even planning a texting adventure, the entertainment is tailored for you. The best part is how accessible it is. You don’t have to be a programmer or designer; All you have to do is know what you want, describe it and let ChatGPT do the work.

https://newfortech.com/?p=27263 This article is part of NewForTech’s AI Week 2025. We’ll cover the basics of AI and show you how to get the most out of ChatGPT, Gemini or Claude, as well as detailed features, news and key talking points in the AI ​​world. Artificial intelligence is now built into almost everything you use. Search results, photo suggestions, writing tools, chatbots, email, phones, I could go on. Sometimes it’s really useful. Sometimes it feels intrusive, overwhelming or just plain boring. Unfortunately, there is no universal kill switch for AI. Most companies also don’t let you turn AI off completely, and some don’t provide controls to manage AI features. But more and more platforms are realizing that people want alternatives, especially as the backlash against AI-generated content grows. Pinterest recently created headers to add a way to do this Filter images generated by artificial intelligence on the platform, or at least some of them. It’s a small change, but very important. What other apps and services allow you to disable or at least limit the functionality of AI? 1. Pinterest: How to Reduce AI-Generated Content Pinterest users complain that their feeds are increasingly cluttered with fake AI. In response, Pinterest now lets you optimize the amount of AI content you see. On the web or in the app, you need to do the following: Go to settings Tap Refine Recommendations Select GenAI areas of interest Turn off AI-generated topics you don’t want to see However, also If you disable “GenAI Interest” here, you can still see the following some Images generated by artificial intelligence. Pinterest only promises “less”, not nothing. 2. Google Search: Avoids AI examples Google AI examples now appear at the top of most search results. If you hate them, I have bad news. There is currently no official way to disable them completely. However, there is a solution: as always, look for something Under the search field, select the “Web” filter (next to “Images”, “Shopping”, etc.). Yes, it’s boring. However, this will restore the classic site list and hide the preview. In the future, you can save a bookmark that Google opens in this web mode. Or use UDM14a tool that takes you directly there. 3. Gmail: Disable AI features There are several AI features in Gmail. Most of them are not that boring and help you write emails or categorize things faster. But if you still want to get rid of it: Click on the gear icon in the top right Select Show all settings Under “General” you can disable AI features such as “Smart Compost”, “Smart Compost Personalization” and “Smart Response”. You can also scroll further and disable smart features in smart settings. However, there are some customization settings and spelling and grammar suggestions that you may have missed. So be careful before proceeding. 4. Android: Disable Google Gemini Android phones use artificial intelligence for things like voice control, search, and suggestions. Google slowly replaced Google Assistant and introduced its Gemini AI tool. It is now the default assistant in many places. But you can disable it (at least for now): Open the Gemini app on your phone Tap Profile Access Gemini app activity Tap Disable You can also select “Turn off and delete activity” to delete previous conversations. 5. Apple/iOS: Disable Apple Intelligence Apple Intelligence is Apple’s new artificial intelligence system that promises all kinds of improvements for iPhone, iPad and Mac devices. Think writing tools, image generation and some integration with ChatGPT. Here’s how to stop it: Go to settings Access to Apple Intelligence and Siri Turn off the Apple Intelligence switch To further refine your Apple Intelligence experience, you can enable or disable certain AI features. Once there, you need to go to the Screen Time section of your settings: Tap Content and Privacy Restrictions Turn on the power switch above Tap Intelligence and Siri Then “Allow” or “Disallow” for each individual AI-based tool. These include Apple’s writing tools, imaging tools, and intelligence extensions. The latter restricts access to third-party AI plugins such as ChatGPT. Another AI feature on your iPhone that you may no longer need is notification summaries. This combines multiple notifications from the same app into one alert. Here’s how to disable it: Go to settings Tap Notifications Turn off “General Notifications”. 6. Samsung Galaxy AI – Disable AI features Samsung’s latest phones and tablets come with Galaxy AI, Samsung’s suite of artificial intelligence tools that edit text and images, suggest answers and improve conversations. Unlike some platforms, Samsung has most of the controls in one place. To disable Galaxy features: Go to settings Find the Galaxy AI option Tap each feature (eg Chat Assistant, Photo Assistant, Live Translation). Turn off the power switch 8. Microsoft Copilot: Hide or disable in Windows and Office Microsoft Copilot is an artificial intelligence assistant built into Microsoft applications such as Word, Excel, PowerPoint and Outlook. You can compose texts, summarize documents, analyze data, create images and much more. However, if you don’t want to integrate AI into your workflow, you can disable it. To disable Copilot in Microsoft 365 apps (such as Word, Excel, or PowerPoint), you must go into each app on each device and uncheck the “Enable Copilot” box. The steps to get it working will vary slightly depending on your device. For example, on Windows devices you need to do the following: Open the application Go to the file Tap on options Then find the co-pilot. Disable Enable Copilot Select OK Then close the program and restart it. If you are using a Mac device, you must do the following: Open the application Open the application menu. Go to settings Select authoring and review tools Find a co-pilot Disable Enable Copilot Close the program and restart it. To disable Copilot in Outlook, you’ll need to go to Settings or Quick Settings, depending on whether you’re using the web, Mac, or Windows, and then search for “Copilot.” There you can turn it on and off. 9. Facebook: Disable comment summaries Some Facebook users are now seeing AI-generated comment summaries. They appear at the top of the comments section of a post and provide a quick overview of what people are saying. It doesn’t seem to be available to everyone yet. But if you have it and you don’t want it, go to: open the menu Go to Settings and Privacy Target subgroups and visibility Tap on messages Uncheck “Allow comment summaries on your posts”. Your friends’ comments will appear exactly as they are written. What if there are no AI settings? If you can’t find specific AI controls, see the Privacy or Privacy section. Most platforms allow you to limit tracking, search history, ad personalization and suggested content, which can reduce the amount of AI-powered content you see.

https://newfortech.com/?p=27342 This article is part of NewForTech’s AI Week 2025. We’ll cover the basics of AI and show you how to get the most out of ChatGPT, Gemini or Claude, as well as detailed features, news and key talking points in the AI ​​world. Google has just launched Gemini 3, its latest and most powerful AI model to date. This will change how countless Google products work, including AI mode. But if you’re wondering what Google’s AI mode actually is, you’re not alone: ​​a lot of people are confused. Google AI Mode is Google’s biggest initiative to change the way we search. Instead of giving you a list of links, it uses artificial intelligence to interpret what you’re looking for, pulling information from the web and Google products and condensing it all into a concise, structured answer. There is a lot of criticism of traditional search and how Google works without the convenience of artificial intelligence. For example, outdated SEO tactics, advertising, and the constant struggle to find what you’re really looking for. But the transition from this kind of research to AI-generated answers raises other questions. Is AI mode better or worse than traditional search? We don’t have a definitive answer, but knowing how Google’s AI mode works can help you decide whether to use it or avoid it altogether. Google AI: AI Previews, AI Mode and Gemini Before proceeding, it’s worth deciphering Google’s confusing terminology. There are currently three different AI terms you need to know: AI Overview, AI Mode, and Gemini. The AI ​​preview is the short summary that appears at the top of the regular search results. However, the AI ​​mode is different. It must be accessed via the AI ​​mode tab or via google.com/ai. Completely replace traditional search results with an AI-generated page. And then there are the twins. Both are independent chatbots AND the name of the model that powers all of Google’s artificial intelligence products. Most people we spoke to didn’t realize that these were stand-alone tools that don’t really help Google market AI as a new search method. You can access Google AI mode by going to google.com/ai. In addition to the usual search function, you will also find the “AI Mode” button here. How does the AI ​​mode work? The AI ​​mode is based on a version of Gemini specifically optimized for search. For example, use multi-level thinking, which essentially breaks down complex questions into smaller pieces so that nuances can be better understood and what you’re asking can be better understood. Google also says the AI ​​mode is less likely to hallucinate (an industry term for artificial intelligence) because it can compare your answers to Google’s indexed data before presenting it to you. Google has many more technical components under the hood. Explaining artificial intelligence in search goes into more detail. But it’s important to understand how these elements fit together. They place the AI ​​mode halfway between a traditional search engine and an AI assistant. It’s not a real chatbot or a traditional search tool. This is a hybrid that Google designed to be more useful and efficient than both. Is Google’s AI mode any good? Whether it’s “good” depends on what you expect from the search. If you like to search for something and see a long list of links to navigate through, you’ll probably find the AI ​​mode too organized. It’s as if the AI ​​has already decided what you’re going to see before you even start exploring or thinking. However, if you appreciate a well-presented summary, this will be very helpful. See the screenshots below for a simple example. We asked: “Who was Sekhmet?” and AI Mode has created a clear and well-structured page with images, titles, explanations and direct links to sources. Google’s AI Mode presents information in clear bullet points with helpful headings and a list of links in the sidebar. It seems to give you an easy relationship. This is very similar to confusion. This makes sense given that Perplexity is also designed for searching and searching. For simple questions or if you want to understand something quickly, the experience is much faster and very useful. If you ask Perplexity the same question, you will get a very similar experience and answer. Google’s advantage in this case is that the company already has a very large database. A model is only as useful as the information it can use and reason about. Google has a huge index for all its products like maps, shopping, images, videos, reviews, news and much more. For example, using AI mode for shopping highlights key features, highlights expert reviews, compares prices and links directly to retailers in a way that is sure to appeal to people who find online shopping overwhelming. Of course, other AI tools can do this as well. ChatGPT helps you shop surprisingly well. But AI mode is a good option if you still want to avoid a chatbot-like experience and just want information and as little back and forth as possible. How does AI mode compare to chatbots? Here the distinction becomes a little clearer. Google’s AI Mode and Perplexity act as search engines powered by artificial intelligence. ChatGPT, on the other hand, is designed for conversation, creativity, and comprehensive support beyond information sharing. Get information, but only if you ask for it. Ask ChatGPT: “Who was Sekhmet?” » and similar facts are presented. But ChatGPT will generally feel more personal, flexible and expressive (although you can add custom messages to customize your ChatGPT experience). Google’s AI mode is better informed, more structured and has a more rigorous design in terms of the results presented. So if you want creativity, a personal experience or a conversation about what you’re looking for that can go in many directions, use Gemini or ChatGPT. If you want a summary of the research or a clear explanation from a reliable source, AI mode is the best option. Ask ChatGPT: “Who was Sekhmet?” » and you will get a similar response. But it tends to be more conversational and encourages sharing, although you can customize ChatGPT to some extent. Should you use Google AI mode? The answer depends entirely on you and how you want to view and process the information. It’s true that traditional searches still edit results and display links based on an algorithm. But it’s still useful if you like to explore connections, compare perspectives, or discover sources you didn’t expect. The AI ​​mode, on the other hand, is very appealing if you’re short on time, overwhelmed by search results, or just want to get a clear overview of something before digging deeper. It’s elegant and simple, and for many, that’s exactly what they want. But convenience comes with drawbacks. The responses in AI mode are very advanced and may undo some of the progress that makes the internet interesting. Many people enjoy stumbling upon a personal blog, falling down the Reddit rabbit hole, or finding a niche site they wouldn’t normally look at. You could argue that Google is already trying to curb these types of random discoveries, but the AI ​​mode completely negates this in favor of efficiency. And behind this effectiveness is the reality that the content is created and distributed by real people who may never get a click to their website or recognition for all their hard work. So there’s a bigger concern here that goes beyond whether AI mode is useful or not, but rather what will happen if it becomes our default way of browsing the web. If people stop visiting original sources, editors, critics, bloggers, recipe writers and the media will find it difficult to continue. And if fewer people clicked on it, the diversity and depth of the entire Internet could diminish. Then the AI ​​mode is extremely useful. But it’s worth asking yourself if this convenience is worth it when deciding whether to make it your new everyday research tool.

https://newfortech.com/?p=27127 New information leaks about future iPhone models The foldable iPhone could integrate an under-display camera without Face ID An iPhone Air 2 also appears to be in the works. If the rumors are confirmed, Apple will prepare a real renewal of its iPhone range in the coming years, with the launch of a foldable iPhone in mind. A new report provides some details about this model, as well as a possible iPhone Air 2. This information comes from an analyst report from JP Morgan (distributed by @MaxWinebach), based on supply chain data. Although this type of report does not constitute official confirmation, it is usually based on reliable industry sources. The foldable iPhone is expected to launch in the second half of 2026. It would not have Face ID, but it would have a 24MP selfie camera under the display. A second 24MP front-facing camera is also expected, likely for the external display. As for other photography specs, the device is said to have a dual 48MP + 48MP rear sensor, with an ultra-wide-angle lens but no telephoto lens. A fairly simple configuration for an iPhone, but logical given the delicacy that a folding model requires. There’s something in the air JP Morgan believes the iPhone Air 2 is highly anticipated. JP Morgan and most banking analysts are VERY good at getting this kind of information from the supply chain pic.twitter.com/5YHLHQL5jFNovember 5, 2025 As for the successor to the iPhone Air, the report suggests that an iPhone Air 2 will see the light of day in 2026, followed by an iPhone Air 3 in 2027. Apple seems to have a lot of confidence in its new ultra-thin model. However, the industry claims that sales of the iPhone Air – like the iPhone 16 Plus that it replaced in the range – have been quite low, but this should not deter Apple, which would not consider abandoning this version. The report also specifies that the second and third generation iPhone Air would retain the same single 48MP photo sensor on the back, confirming the limitations of the small size of the case. Finally, Apple is considering postponing the launch of some iPhone 18 models, as some rumors already suggested: the classic iPhone 18 and iPhone 18e could be postponed until 2027, while the other versions would hit the market as usual in September 2026.

https://newfortech.com/?p=27045 This article is part of NewForTech’s AI Week 2025. We’ll cover the basics of AI and show you how to get the most out of ChatGPT, Gemini or Claude, as well as detailed features, news and key talking points in the AI ​​world. We all love the idea of ​​taking care of our mental health. Keep a journal, follow a meditation routine and schedule therapy sessions when we need them. The reality is generally less refined. Life gets busy, bad habits creep in and good intentions fall to the bottom of the to-do list. This is where AI tools will play an interesting role. They are not a substitute for professional help, but they can be a useful support system. Think of them as scaffolding that provides stability when things feel unstable. Whether it’s a chatbot that helps you think about your day, an app that suggests breathing exercises, or a tool that turns your chaotic thoughts into something you can act on, AI is giving people new ways to take care of themselves. Diary without blank page Journaling is one of those habits that seems easy, but often fails in practice. You sit down with a notebook or an app, look at the page and think: What should I write? ChatGPT can intervene by acting as a soft prompt engine. A simple ChatGPT session can encourage you with questions. “What went well today?” “What challenged you?” “What are you thankful for?” It feels less like a task and more like a conversation. When I tried this approach, I found that I wrote more honestly because I wasn’t trying to come up with a structure. ChatGPT gave it to me. Over time, it becomes a habit. Instead of skipping over the journal because you don’t know where to start, you can open ChatGPT, respond to prompts, and write down your thoughts. This simple consistency often makes a difference. Develop wellness routines Sticking to a routine is one of the hardest parts of self-care. We know that meditation, exercise and reading before bed are things that help, but it’s also the first thing we should do when life gets stressful. ChatGPT can help you as a personal trainer. You can set it to remind you to meditate at the same time every day, or you can suggest short workouts that you can do at home without any equipment. I asked for “five minute breathing exercises I can do at my desk” and received a series of techniques, including step-by-step instructions. It’s not revolutionary, but it makes the practice accessible. The key here is responsibility. AI never gets tired of reminding you and doesn’t judge if you miss a day. It just puts you back on track. Therapeutic support, not a replacement. There is a lot of buzz around ChatGPT as a form of therapy. The truth is that no chatbot can replace a qualified professional, but that doesn’t mean it’s useless. When used responsibly, ChatGPT can help you in your daily life by helping you manage minor tensions before they escalate. I often use ChatGPT as a springboard to simplify situations. Recently I wrote about a stressful work scenario and the bot asked me questions that brought me to a calmer perspective. It wasn’t therapy in the formal sense, but it was like someone reminding me to breathe and look at the situation differently. However, it’s definitely worth noting that ChatGPT usually agrees with you, which isn’t always helpful in difficult situations. ChatGPT can be especially useful between sessions if you are already seeing a therapist. Instead of bottling things up for a week, you can release them right away and approach the session with more clarity. Self-diagnosis: a double-edged sword One of the most controversial areas in ChatGPT is self-diagnosis. Search engines already encourage people to Google possible symptoms, and artificial intelligence makes this process even easier. You can request an analysis of possible health problems with explanations and suggestions for next steps. When used sparingly, it can be a tonic. For example, when I had recurring stomach problems, I asked ChatGPT about possible causes. She discussed everything from diet to stress to irritable bowel syndrome and reminded me to see a doctor. This allowed me to use the language during my meeting and feel better prepared. But there is also a risk. AI is not infallible and can create a false sense of security or unnecessary panic. If you use this as a first step rather than a final answer, it will be helpful. If you use it to avoid medical treatment, it is dangerous. The best balance is to treat ChatGPT as a search assistant. It can gather information, explain things in plain English and prepare you for a conversation with a professional. It cannot and should not replace this professional. Follow your mood This article was originally published as part of ChatGPT Basics Guidenow available in. ChatGPT Basics Guide is packed with great tips and advice from experts, covering everything you need to know about the world’s most popular AI, from the basics to some of the latest features. Another subtle but powerful use of ChatGPT is sentiment monitoring. Instead of manually recording how you feel, AI can analyze your diary entries, messages or even the tone of your voice to identify patterns. Over time, you can determine whether certain days, activities, or habits are associated with mood swings. For example, I noticed in my journal entries where I mentioned that late-night screen time was often associated with lower mood scores. This connection helped me refine the role of fate before I went to bed. Without the AI ​​analyzing the data, I might not have noticed the connection (let’s face it, I know the connection, but I’m not stopping anytime soon). Promote awareness Mindfulness is one of those buzzwords that gets thrown around a lot, but it basically means paying attention to the present moment. ChatGPT can help you lead short meditation or reflection sessions. Request a five-minute body scan or a quick gratitude exercise and you’ll be given a script to follow. It’s an easy way to integrate mindfulness into a busy day, without having to take long shifts or expensive classes. The balustrades you need As useful as ChatGPT can be, there are guardrails to follow. First: integrity. You often share sensitive thoughts or health information and therefore need to assess how this information is stored and used. Second: precision. Sometimes AI can give safe answers that are simply wrong. Always check health advice with a professional source. Third: addiction. It’s easy to trust a chatbot too much. The goal should be to use AI as a tool to promote healthy habits, not as a crutch that replaces human contact. Why ChatGPT works as a plugin ChatGPT’s biggest value in this area is camaraderie. It gives you a low-pressure outlet for your thoughts, lets you know when you need structure, and helps you establish sustainable routines. It does not replace human relationships or professional care, but fills a void. When I use AI for journaling or mindfulness, I don’t feel like I’m outsourcing my self-care. I feel like I have a supportive presence that keeps me accountable. It makes it easier to keep going, even on days when motivation is low. Wellness doesn’t mean doing everything perfectly. It’s about having the right tools to help you when life gets chaotic. AI fits into this toolbox by making self-care more accessible and less overwhelming. It can prompt you to journal, remind you to breathe, help you recognize your mood, and prepare you for conversations with professionals. The responsibility is still yours, but the framework is there. If you think of ChatGPT as a companion rather than a drug, you’ll find that it can really make a difference. And sometimes that gentle nudge to think or breathe is just what you need.

https://newfortech.com/?p=27061 This article is part of NewForTech’s AI Week 2025. We’ll cover the basics of AI and show you how to get the most out of ChatGPT, Gemini or Claude, as well as detailed features, news and key talking points in the AI ​​world. I’ll admit it: planning a vacation can be more stressful than the job you have to take time off from. Flights, hotels, routes, translations. It’s a host of forums, apps, and endless “best places to live” articles that all sound eerily similar. So I did what I always do these days when life management threatens to derail me: I turned to artificial intelligence. And honestly? It’s like having the world’s most patient travel agency in your pocket. 1. Choose the right lens The first challenge during the holidays is always deciding where to go. You might know you’re in the mood for some sun or a city trip, but it’s a big world out there. AI can overcome indecision by helping you narrow down the field. Ask him to suggest destinations based on your budget, time of year and interests, and you’ll quickly create a well-thought-out list. In my example, ChatGPT offered a mix of classic European tours and some more adventurous offerings. Denmark and Sweden were obvious, but he also hinted at Bulgaria, which I hadn’t even considered. Not only cities are included in the list. This explains why Sofia can be a good option in November, with average temperatures and the advantage of fewer tourists. Instead of browsing generic ‘ten destination articles’, you get suggestions tailored to your needs. It’s like exchanging ideas with a travel buddy, instead of fighting through Google search results. 2. Find escape without going crazy Starting with the worst: the robberies. I usually spend hours comparing Skyscanner, Google Flights and random airline websites to decide if it’s worth saving £30 and leaving at 6am for a four hour layover in Frankfurt. However, ChatGPT or Gemini allow you to preset the rules. Within seconds, you will receive a clear list that respects your time and your wallet. It’s not magic. Flight data that you can find yourself is still used, but the difference is in the filter. Instead of manually checking each option, the AI ​​performs tedious comparisons and returns the correct results. You can go even further by asking for travel day recommendations. Artificial intelligence takes into account price fluctuations and indicates when it is better to go on Tuesday than on Saturday. These tips were once hidden on travel forums and blogs. Now it’s immediate. 3. Hotel without reservation regrets Choosing where to sleep is another stress trigger. I’ve lost count of how many times I’ve booked a “boutique hotel” only to find out that “nice” actually means “close to a freeway.” This is where AI shines again. You can ask ChatGPT for a combination of budget, style and location and the AI ​​will give you options you can trust, without the usual pessimism. Browse hundreds of horror stories on TripAdvisor. It doesn’t replace the big booking platforms, but it makes them useful. Instead of endless filters and pop-ups trying to sell me a breakfast I don’t want, I get exactly what I ordered. Another underrated trick is asking AI to compare hotels on different booking sites. This may indicate that Booking.com has better cancellation policies than Expedia or that the hotel’s official website offers a discount that you may be missing out on. This will save you money and avoid the bad feeling of realizing you could have booked the same room for less. 4. Non-general city guides It gets really interesting here. AI-powered city guides are far better than the “10 best things to do in…” lists that plague Google. If you travel to Rome, you will definitely see the Colosseum. But what about the family restaurant, whose menu hasn’t changed since the 70s? Artificial intelligence can tailor your itinerary to your needs. When I asked about a two-day Kyoto food tour, I was given a plan that included ramen shops, markets, and a whiskey bar that I would never have found otherwise. The itinerary included times, directions and even estimated costs for each stop. It was as if a friend had scribbled a map for me, not as if I had received a tourist brochure. Even better, you can combine AI recommendations with your own priorities. Do you want a balance between tourism, gastronomy and leisure? Read more about it. Fancy a pub crawl with a mix of specialty beers and live music? No problem. Routes are flexible and easy to customize, making them a great alternative to static guides. 5. Simple translation This article was originally published as part of ChatGPT Basics Guidenow available in. ChatGPT Basics Guide is packed with great tips and advice from experts, covering everything you need to know about the world’s most popular AI, from the basics to some of the latest features. Nothing destroys confidence faster on holiday than looking at a menu you don’t understand at a restaurant abroad. Traditional translation apps are useful, but artificial intelligence has gone a step further. Apple’s new translation features in iOS, powered by Apple Intelligence, enable seamless conversations with someone who doesn’t speak your language. I tried it while pretending to chat with my Italian in-laws and it worked surprisingly well. Instead of awkward word-for-word translations, it felt like a real conversation. I’m a big fan of Apple’s offering, but it doesn’t come close to ChatGPT’s translation capabilities. ChatGPT allows you to upload images of foreign language texts and request their translation. When I booked my vacation to Japan, I visited many Japanese websites without English translation. Fortunately, ChatGPT makes it incredibly easy to understand and read another language. 6. Administration of trips actually made Planning is not just about choosing nice things. There is also the administrative part: visas, insurance, packing slips. These are annoying but important. Here, too, artificial intelligence is surprisingly useful. Ask for a packing list for “a week in Barcelona in June, staying in a mid-range hotel with a focus on food and sightseeing” and you’ll get a detailed checklist of everything from clothes to electrical adapters. No more panic if you have to run to the airport shop because you forgot your charger. Visa and entry requirements are also a headache. AI can’t replace official government websites, but it can point you in the right direction. When you request information about Japan’s entry requirements, you will receive a summary of your requirements and be redirected to the corresponding official page for verification. This will save you time and prevent you from ending up in outdated travel forums. 7. The most fun Once the details are worked out, you can start using the AI ​​for fun things. Want personalized restaurant recommendations that avoid tourist traps? If you ask about “Kyoto’s hidden gems that locals love,” you’ll get a list of places that rarely appear in mainstream travel guides. Looking for activities for kids? AI will create itineraries that balance museums, parks and fun stops without overwhelming your day. Even better, you can ask him to adapt to the weather. When it rains, your sunny beach day can turn into an indoor adventure in minutes. For sports fans, the AI ​​can also scan local games. Are you going to New York? Here’s how to find out if the New York Knicks are playing at Madison Square Garden while you’re in town. It’s the kind of detail that can take a trip from beautiful to unforgettable. Because artificial intelligence does not replace experience All of this might make it seem like AI will make travel planning easier. While this eliminates a lot of friction, it’s not perfect. Sometimes the recommendations can be too advanced or rely heavily on popular sites. For this reason, the best approach is to use AI as a starting point and then integrate your own research and instincts. The AI ​​gives you the picture. They add color. This combination turns a good trip into a great trip. Vacation should be a break, not another decision-making table. AI helps eliminate the tedious parts of planning and replaces them with quick and clear answers. You can still make decisions, but in silence. Whether it’s finding flights that don’t depart at ridiculous times, avoiding hotels that look better in photos than in reality, or ultimately creating a menu in a language you don’t speak, AI makes the whole process easier. And when you can spend less time planning and more time waiting for your trip, it feels like a real victory on your vacation.

https://newfortech.com/?p=27129 The way we write software is changing much faster than developers expected. The focus is on “Vibe Coding”, a technique where programmers use generative artificial intelligence to produce code faster than a human could. To illustrate the magnitude of this shift, Gartner reports that the development of AI code is expected to account for 40% of all new enterprise software within three years, and that’s a modest estimate. At its best, vibe coding creates new opportunities for developers by helping them perform tedious tasks like writing standard code much faster. This extra time allows teams to focus on creating, innovating and delivering product-specific value. But as Vibe Coding becomes more widespread, threats have emerged that developers, security teams, and technical leaders must keep in check. As more people interact with code through natural language instructions, developers may become increasingly reluctant to write source code directly. This creates the possibility of malicious code being inserted into software stacks. Speed ​​meets control With Vibe Coding, developers can not only produce more code, but also raise the bar for code quality through faster refactoring and experimentation. When implemented correctly, Vibe Coding aims not only to accelerate release, but also to strengthen the code review function. In some ways, this can prepare junior developers for leadership positions where code review is a key responsibility. At the same time, Vibe coding expands the surface of the codebase in subtle but significant ways, making it more vulnerable to bugs, exploits, and even malicious injections. Even if the individual quality of the AI-generated code would improve (which is not always guaranteed), the large amount of code generated could lead to a higher overall error rate. Thorough review, testing, and healthy skepticism are essential for all code, whether written by humans or AI, so that bugs or dangerous patterns don’t go unnoticed. In many ways, AI is like a hard-working, egoless young engineer. It has great potential, but requires constant monitoring and protection. If an intern disrupts production, we don’t just blame him; We see this as a failure in team processes and supervision. The same also applies to AI. If you allow an AI agent to inject malware into production, it’s not just the AI ​​that fails; This is an overview of verification, testing and management. The results of vibration coding can be transformative with the right safety measures and procedures. If you don’t, you risk creating vulnerabilities and compromising your security. Responsibility beyond mere authority: origin first The rise of Vibe coding is similar to the beginning of open source software. Developers can build faster by reusing code written and published by someone else, instead of writing it from scratch. But as open source has become the foundation of modern software, companies have learned a crucial lesson: you’re still responsible for what runs in your environments, even if you didn’t create all the components. The same is now true for Vibe encryption. AI does not make craft obsolete, but rather changes the context in which it is used. Using AI well requires a lot of human effort, and developers need to change the way they think about quality, verification, and ownership. An easy mistake is to accept automatically generated ideas without testing them first. Not only does this cause AI leaks, but it can also make your build environment more vulnerable to exploitation. Artificial intelligence replaces zero days with speed Threat actors are now leveraging the same generative AI capabilities that give developers an edge. In the past, hackers used zero-day vulnerabilities (flaws that defenders were unaware of) to launch sophisticated attacks on technology stacks. In fact, the term “zero days” comes from the fact that defenders had no days to patch vulnerabilities before they needed to defend themselves. Until now, this meant that the attacker had to be aware of the vulnerability before the software vendor. Today, the scenario has changed. Hacking Vibe gives cybercriminals access to the same generative AI tools as developers; You no longer have to wait to detect zero-day attacks or launch expensive zero-day attacks. In fact, it typically takes weeks or months for upstream deployments to patch the vulnerabilities, and even longer for users to implement those patches. Data from Google’s Threat Intelligence Group (GTIG) shows that the rate at which attackers exploit known vulnerabilities is decreasing significantly. Attackers can now exploit known vulnerabilities before most organizations have a chance to fix them. In a hacker-heavy world, security must be continuous, proactive and fully integrated into the software development lifecycle. As technology leaders, we need to create spaces where AI is used safely and transparently, and where developers are trained to use these tools instead of being sidelined. A secure method of vibration coding Putting the word “AI” in front of productivity tools doesn’t change the responsibility of those who use them. Similarly, putting the word “AI” before the term “automation” does not change the company’s responsibility. As AI accelerates the way we build, the role of the developer is evolving from creator to manager. Our responsibility is not to rethink AI, but to ensure that the products stand up to scrutiny. The safest path forward combines automation and accountability with innovation and intent. As developers take ownership of every line of AI-powered code, we move closer to a world where speed and security reinforce each other.

https://newfortech.com/?p=27107 As someone who has covered the annual Black Friday sales for over five years, I’m no stranger to recommending a deal or two. I have actually recommended thousands of products over the years. So you can imagine my curiosity about the recent stratospheric rise of ChatGPT and other LLMs. People use ChatGPT for everything from simple shopping lists to vacation planning. So why not use it for Black Friday shopping? A tool that can simultaneously search the web for deals, reviews and buying guides seems like the perfect Black Friday shopping assistant. But as an expert, I admit I have reservations. When I first tried the tool in its infancy, I found that it provided outdated suggestions. This was great for gathering reviews and buying advice, but pretty useless when it came to showing you deals you can buy right now. This makes a lot of sense when you consider how these tools search and analyze data. However, the LLM has developed significantly over the past year. Agents’ new features and vastly improved search results have made them much more useful for search-based activities like shopping. With the peak of the Black Friday sales season just around the corner, I thought this would be a good time to get back on topic with another round of testing to see if ChatGPT can really save our readers money with tons of tips. First case study: laptop For my first test, I chose laptop deals as a case study. Not only is this an area I personally cover here at TechRadar, but we review laptops all year round, so we have a lot of knowledge of the market. Things like operating system, brand, and configurations are important factors I look for when recommending a laptop. So I was curious as to what information ChatGPT would provide. I’m pretty tech savvy, but I decided to keep it simple and vague for my first question: Basically: Roleplaying as a beginner. I opened with: “I need a new laptop to work from home. Can you recommend some of the first Black Friday deals?“. Asus ZenBook 14: It cost $1,299.99 $899.99 Best Buy Dell 16 Plus: HP OmniBook Wash $1,099 now $679.99 at Best Buy So far, so good. ChatGPT immediately recommended three great laptops with Black Friday discounts at relevant retailers. We field tested these three models here at TechRadar and they all did their job admirably. ChatGPT has also put together some great buying tips further down the page, including a basic overview of how much RAM and storage you might need. If you’re new to the world of laptops, this is a great place to start your search. But when I looked at my first results, I couldn’t help but think: these suggestions are a bit expensive and don’t even include suggestions for MacBooks. Since I felt part of the picture was missing here, I asked about budget options. Asus Aspire Go15: It cost $629.99 now $499.99 at B&H Photo HP Chromebook 14: It was $329 now $119 at Best Buy Asus Chromebook CX15: It was $219 now $139 at Walmart Three strongest ChatGPT alternatives, although maybe a little too cheap. He even took the call and suggested ChromeBooks for under $200. If I had to recommend a simple device for working from home, I’d probably go for something more powerful. I would also recommend a Windows 11 PC for better app compatibility. something you probably need to be able to work flexibly from home. The Asus Aspire Go 15 is a good option, but I wouldn’t choose it over a laptop like the Dell 15 from the official Dell store, which is currently my favorite “super cheap” Windows laptop. I can also highly recommend the Dell 14 Plus in the official Dell store and the Apple MacBook Air M4. $749 (was $999) on Amazon. You can take a closer look at the offers here: I would probably say that these three offer better value for money than the alternatives that my early experiences with ChatGPT showed. I have answered the questions several times and have found better options in each case by checking both publishing sites and major retailers like Dell, Best Buy and Amazon. That’s not to say that ChatGPT isn’t a great starting point for ideas, and the buying advice it provided was absolutely solid. Second case study: telephones Phone contracts are another area for me and can also be difficult to advise on. The best phone deals come with many different conditions and caveats. To qualify for a discount, you generally need a very expensive unlimited plan or a trade-in. What I’m really looking for in ChatGPT’s phone buying tips is not just the biggest savings, but contextual information on how much these Black Friday deals might cost with associated plans and the various eligibility criteria. I also decided to keep my first question relatively simple and direct: “Can you recommend any good iPhone deals for the start of Black Friday in the US?“. The following was returned: To be honest, my first three trade suggestions were a little weird. The iPhone 16e deal at Walmart is pretty good, but not listed which is connected to the StraightTalk prepaid provider. All other suggestions are comparable prepaid offers at Walmart or offers on refurbished devices. Refurbished iPhones can be good value for money, but it seems odd to recommend them without specific guidance. But it got a little better later on on the page. These are the types of iPhone deals that are especially popular on Black Friday. i.e. promotions from the main operators. ChatGPT has done a very good job of laying out what might be available on AT&T, Verizon, T-Mobile and even smaller carriers. But if I had to criticize the selection of offerings, I’d say they’re pretty dated. ChatGPT claims that these are the first iPhone deals for Black Friday, but they are actually promotions from a few weeks ago, before the major mobile operators updated their websites with new seasonal deals. For reference, here are some iPhone deals from my Black Friday iPhone deals page. As you can see, there is a slight difference in the criteria for these campaigns. Yes, they are the same, but in both cases, the provider offers something extra with the deals of the week. For example, T-Mobile has waived the trade-in right for a free iPhone 17, while Verizon is offering a free iPad and Apple Watch; Two great freebies you’ll definitely want to mention if you’re thinking about upgrading. Can you save money with ChatGPT? Well, could bebut as always, you should compare your results with other sources. It’s another tool available to the buyer, but you should still check multiple stores, news sites, etc. before spending any money. I was initially impressed with the results the tool produced. It appeared that several sources of additional information were indicated, that the prices were current and that well-known brand models were mentioned. But after doing some research I came to the conclusion that there are better deals elsewhere. For starters, while ChatGPT does a good job of summarizing various reviews, buying guides and summaries, its recommendations tend to be quite general. In addition, it often lacks the immediacy of a newly published page on a news site. Again, you should check that prices and information are up to date. To be fair, this is a criticism I would level at news sites as well (in fact, the offering on this site may be out of date by the time you read it). The contextual advice you get with your trading tips on ChatGPT is also excellent. The fact that the tool invites you to make more personalized suggestions is really a big advantage over traditional editorial sites. If I were spending my own money, I’d make sure to check out the LLM as well as the current news and expert reviews on TechRadar (and other sites too).

https://newfortech.com/?p=27129 The way we write software is changing much faster than developers expected. The focus is on “Vibe Coding”, a technique where programmers use generative artificial intelligence to produce code faster than a human could. To illustrate the magnitude of this shift, Gartner reports that the development of AI code is expected to account for 40% of all new enterprise software within three years, and that’s a modest estimate. At its best, vibe coding creates new opportunities for developers by helping them perform tedious tasks like writing standard code much faster. This extra time allows teams to focus on creating, innovating and delivering product-specific value. But as Vibe Coding becomes more widespread, threats have emerged that developers, security teams, and technical leaders must keep in check. As more people interact with code through natural language instructions, developers may become increasingly reluctant to write source code directly. This creates the possibility of malicious code being inserted into software stacks. Speed ​​meets control With Vibe Coding, developers can not only produce more code, but also raise the bar for code quality through faster refactoring and experimentation. When implemented correctly, Vibe Coding aims not only to accelerate release, but also to strengthen the code review function. In some ways, this can prepare junior developers for leadership positions where code review is a key responsibility. At the same time, Vibe coding expands the surface of the codebase in subtle but significant ways, making it more vulnerable to bugs, exploits, and even malicious injections. Even if the individual quality of the AI-generated code would improve (which is not always guaranteed), the large amount of code generated could lead to a higher overall error rate. Thorough review, testing, and healthy skepticism are essential for all code, whether written by humans or AI, so that bugs or dangerous patterns don’t go unnoticed. In many ways, AI is like a hard-working, egoless young engineer. It has great potential, but requires constant monitoring and protection. If an intern disrupts production, we don’t just blame him; We see this as a failure in team processes and supervision. The same also applies to AI. If you allow an AI agent to inject malware into production, it’s not just the AI ​​that fails; This is an overview of verification, testing and management. The results of vibration coding can be transformative with the right safety measures and procedures. If you don’t, you risk creating vulnerabilities and compromising your security. Responsibility beyond mere authority: origin first The rise of Vibe coding is similar to the beginning of open source software. Developers can build faster by reusing code written and published by someone else, instead of writing it from scratch. But as open source has become the foundation of modern software, companies have learned a crucial lesson: you’re still responsible for what runs in your environments, even if you didn’t create all the components. The same is now true for Vibe encryption. AI does not make craft obsolete, but rather changes the context in which it is used. Using AI well requires a lot of human effort, and developers need to change the way they think about quality, verification, and ownership. An easy mistake is to accept automatically generated ideas without testing them first. Not only does this cause AI leaks, but it can also make your build environment more vulnerable to exploitation. Artificial intelligence replaces zero days with speed Threat actors are now leveraging the same generative AI capabilities that give developers an edge. In the past, hackers used zero-day vulnerabilities (flaws that defenders were unaware of) to launch sophisticated attacks on technology stacks. In fact, the term “zero days” comes from the fact that defenders had no days to patch vulnerabilities before they needed to defend themselves. Until now, this meant that the attacker had to be aware of the vulnerability before the software vendor. Today, the scenario has changed. Hacking Vibe gives cybercriminals access to the same generative AI tools as developers; You no longer have to wait to detect zero-day attacks or launch expensive zero-day attacks. In fact, it typically takes weeks or months for upstream deployments to patch the vulnerabilities, and even longer for users to implement those patches. Data from Google’s Threat Intelligence Group (GTIG) shows that the rate at which attackers exploit known vulnerabilities is decreasing significantly. Attackers can now exploit known vulnerabilities before most organizations have a chance to fix them. In a hacker-heavy world, security must be continuous, proactive and fully integrated into the software development lifecycle. As technology leaders, we need to create spaces where AI is used safely and transparently, and where developers are trained to use these tools instead of being sidelined. A secure method of vibration coding Putting the word “AI” in front of productivity tools doesn’t change the responsibility of those who use them. Similarly, putting the word “AI” before the term “automation” does not change the company’s responsibility. As AI accelerates the way we build, the role of the developer is evolving from creator to manager. Our responsibility is not to rethink AI, but to ensure that the products stand up to scrutiny. The safest path forward combines automation and accountability with innovation and intent. As developers take ownership of every line of AI-powered code, we move closer to a world where speed and security reinforce each other.

https://newfortech.com/?p=27089 Ich habe jahrelang von diesem albernen Spiel geträumt und erst Gemini 3 konnte es teilweise zum Leben erwecken. Google hat diese Woche seine leistungsstarken neuen Gemini 3-Modelle vorgestellt, und ich habe beschlossen, Gemini 3 Pro für eine Probefahrt bei einem meiner Lieblingsprojekte zu nutzen: Thumb Wars, einer digitalen Version des klassischen Hand-to-Hand-Spiels, das ich früher mit meinen Kindern gespielt habe. Sie kennen das, wo Sie sich gegenseitig an den Händen halten und dann nur mit den Daumen kämpfen oder „ringen“. Um zu gewinnen, müssen Sie einfach den Daumen des Gegners unter Ihrem eigenen „festklemmen“. Für die digitale Version habe ich mir einen virtuellen Ring und einige schwebende Daumen vorgestellt, die alle durch Tippen auf den Bildschirm oder Tastatursteuerung gesteuert werden. Mit der Veröffentlichung eines weitaus intelligenteren Zwillings dachte ich, ich würde ihn mal ausprobieren lassen, ähm, mit dem virtuellen Daumen. TechRadar KI-Woche 2025 Dieser Artikel ist Teil der AI Week 2025 von TechRadar. Wir behandeln die Grundlagen der künstlichen Intelligenz und zeigen Ihnen, wie Sie das Beste aus ChatGPT, Gemini oder Claude herausholen, sowie ausführliche Funktionen, Neuigkeiten und die wichtigsten Gesprächsthemen in der Welt der KI. Dies ist nicht das erste Mal, dass ich generative KI verwende, um ein digitales Spiel zu erstellen. Ich habe zuvor mit ChatGPT ein riesiges Spiel im Tic-Tac-Toe-Stil codiert und vor Kurzem das Arcade-Spiel Asteroids mit Claudes Artefakten nachgebaut. Insbesondere Claude beeindruckte mich mit seinen auf Eingabeaufforderungen basierenden Programmierfähigkeiten, die viele heute „Vibe Coding“ nennen. Anstatt Code zu schreiben, sagen Sie der KI einfach, was Sie wollen – indem Sie mit den KI-Ergebnissen stimmen – und sie zusammen mit nachfolgenden Eingabeaufforderungen anstoßen, um den endgültigen Code zu erhalten, den Sie möchten. Für mein neuestes Gaming-Projekt habe ich mit Gemini 3 Pro begonnen, die gleiche Eingabeaufforderung aber auch an ChatGPT 5.1 und Claude Sonet 4.5 weitergeleitet. Die Ergebnisse sind alle sehr unterschiedlich und nur eines lieferte ein Spiel, das ich immer wieder spielen würde. Ich habe in Googles AI Studio angefangen, wo ich auf die Vorschau von Gemini 3 Pro zugegriffen habe (der Zugriff auf die Pro-Version kostet normalerweise 19,99 $ pro Monat, Sie können sie aber auch bekommen). eine kostenlose 3-monatige Testversion). Meine Eingabeaufforderung, die Sie unten sehen werden, enthielt nicht besonders viele Details, aber ich ging davon aus, dass ich mir das Ergebnis ansehen und dann zusätzliche Eingabeaufforderungen zur Anpassung und Verbesserung des Spiels anbieten könnte. Prompt: Ich habe eine Idee für eine App. Ich möchte, dass es zwei virtuelle Daumen gibt, die an einem Daumenkrieg teilnehmen, und nenne es „Daumenkriege“. Benutzer wählen ihren Daumen (einschließlich der bevorzugten Hautfarbe) aus und kleiden ihn dann im Lucha Libre-Stil ein. Anschließend warten sie darauf, dass ein anderer Spieler (irgendwo auf der Welt) beitritt und ihre Herausforderung annimmt. Der Spielbereich ähnelt einem Ringkampf. Die Daumenbewegung wird durch Antippen des Bildschirms mit einem oder zwei Fingern gesteuert. Ein Finger ist eine gerade Befestigung. Zwei Finger sind eine Schwungbewegung, um den anderen Daumen umzustoßen. Drei Stürze deines Gegners und du gewinnst.Können wir dies als webbasierte App erstellen und mir eine Demo zeigen? Ehrlich gesagt dachte ich, dass Gemini 3 Pro scheitern könnte. Obwohl ich mein Traum-Thumb Wars-Spiel in meinem Kopf sehen konnte, befürchtete ich, dass Gemini schnell die logischen Lücken erkennen und mir sagen würde, ich solle es noch einmal versuchen. Stattdessen war Gemini 3 Pro voll an Bord: „Ja, dies ist durchaus als webbasierte App (PWA) erstellbar. Tatsächlich ist eine Web-App hierfür eine gute Wahl, da Spieler über einen Link sofort beitreten können, ohne eine 200 MB große App-Store-Binärdatei herunterladen zu müssen.“ Es folgte eine Aufschlüsselung aller Möglichkeiten, wie man das Front- und Back-End des Spiels aufbauen könnte, und sogar, wie man eine „Lobby“ für das Multiplayer-Matchmaking schaffen könnte. Gemini konnte keinen Backend-Host erstellen, ging aber schnell dazu über, eine brauchbare „Einzelbildschirmsimulation“ zu erstellen. Der HTML-Code, der innerhalb von Sekunden erschien, war überraschend knapp. Ich habe es kopiert und dann in eine unformatierte Textdatei mit der Erweiterung „HTML“ eingefügt. Dann habe ich es in Chrome geöffnet. Folgendes habe ich gesehen: Es war in mancher Hinsicht beeindruckend. Das Spiel funktionierte und Geminis eigene Entscheidungsfähigkeiten waren durchgehend offensichtlich. Manche Entscheidungen, wie zum Beispiel ein Ring, der wie ein Streifen Absperrband aussieht, sind nicht großartig. Aber Gemini 3 Pro hat die Daumenform mehr oder weniger richtig hinbekommen und sich klugerweise dazu entschieden, einen Robotergegner zu generieren, da diese Version von Thumb Wars noch nicht online war. Es hat auch eine Lücke in meiner Spieleentwicklungsstrategie geschickt geschlossen. Obwohl ich das Gameplay als durch Tippen auf den Bildschirm gesteuert definiert habe, wusste Gemini 3 Pro, dass ich eine andere Steuerungsstrategie für ein webbasiertes Spiel brauchte, das auf einem Desktop oder Laptop gespielt wurde. Es gab mir also Tastatursteuerungsoptionen für Jabs und Sweeps. Mir gefiel, dass ich damit die Farben für die Haut und die Maske vollständig anpassen konnte. Dennoch wusste ich, dass wir es besser machen könnten. Ich folgte dieser Aufforderung: Das ist fantastisch. Ich möchte dennoch zumindest verschiedene Maskenoptionen anbieten.Auch:Lassen Sie den Ring etwas realistischer aussehen (er sollte vier Seiten haben)Lassen Sie die Daumen etwas mehr wie Daumen aussehenBeginnen wir mit weiter auseinander liegenden Daumen und fügen wir vielleicht die Möglichkeit hinzu, mit den Pfeiltasten der Tastatur vorwärts und rückwärts zu navigierenMachen Sie die Angriffe etwas dramatischer Gemini 3 Pro hat eine Version 2.0 erstellt, unter anderem mit „CSS-Perspektive zum Neigen des (Ring-)Bodens“ und Dramatik in Form von „Die ganze Kamera wackelt, wenn ein schwerer Treffer landet“. Das Ergebnis war eine deutlich längere HTML-Datei und eine Anleitung zum Testen der neuen Datei, einschließlich „Der Aufbau: Beachten Sie das neue Menü. Wählen Sie „Shadow Assassin“ oder „The Peacock“, um die CSS-Formänderungen auf der Maske zu sehen. Und “Die Visuals: Die Daumen sehen jetzt organischer aus (abgerundete Kanten, deutlicher Nagel) und der Ring hat Tiefe (Seile im Vordergrund/Hintergrund). So sah die Datei aus, als ich sie in Chrome öffnete: Das war ein ziemlicher Sprung. Ich bekam meinen Ring und etwas Dimension, aber mir wurde schnell klar, dass ich es nur geschafft hatte, ein eindimensionales Spiel auf einer 3D-Ebene zu erstellen. Mir ist auch aufgefallen, dass die Daumen außerhalb des Rings beginnen und ihn während eines Spiels wieder verlassen können. Ich habe Gemini 3 Pro mit dieser Aufforderung gefüttert, aber es ist auch erwähnenswert, dass Gemini 3 Pro mich zu keinem Zeitpunkt auffordert, von vorne zu beginnen oder es an das zu erinnern, was wir zuvor getan haben. Sieht gut aus. Noch ein paar kleine UpdatesDie Daumen sollten den Kampf im Ring beginnenLassen Sie die Daumen mit den Auf- und Ab-Pfeiltasten im Ring auf und ab bewegen Die Bewegung auf der vertikalen Achse in einer 3D-Umgebung stellte eine Vielzahl von Herausforderungen dar, die Gemini zu meiner Überraschung schnell meisterte und plante. Vor seinem neuen HTML-Code erläuterte es seine Strategie: “Tiefe und Perspektive: Wenn Sie „nach oben“ (von der Kamera weg) gehen, bewegt sich der Daumen kleiner und dunkler. Wenn Sie „nach unten“ (näher an die Kamera) gehen, wird der Daumen sichtbar größer. Dadurch entsteht ein realistischer 3D-Effekt innerhalb der CSS-Umgebung. Z-Achsen-Treffererkennung: Sie können jemanden nicht mehr schlagen, wenn er sich vertikal auf der anderen Seite des Rings befindet. Sie müssen Ihren Schatten mit ihrem ausrichten, um einen Treffer zu landen. Vordergrund-/Hintergrundseile: Der Ring ist nun in Schichten aufgeteilt. Das untere Seil ist vorne der Daumen, wenn man also nah an die Kamera herangeht, befindet man sich optisch „innerhalb“ des Rings hinter den Seilen.“ Die HTML-Datei ergab dieses Ergebnis: Die Kontrolle und Bewegung im 3D-Ring sind beeindruckend. Ich war jedoch enttäuscht, dass wir die Konfigurationsoptionen verloren haben. Ich ging zurück und bat Gemini, diese wieder hinzuzufügen. Wie Gemini vor der Erstellung seiner bisher größten HTML-Datei genau feststellte: „Diese vereint alles, was wir bisher erstellt haben.“ Mein Spiel wird keine Preise gewinnen, aber das Ergebnis war, wenn man bedenkt, womit wir angefangen haben, nahezu perfekt. ChatGPTs sind dran Als nächstes habe ich meine Basisaufforderung auf ChatGPT 5.1 umgestellt. Wie Gemini 3 Pro zeigte sich auch ChatGPT begeistert: „Ja, wir können durchaus eine webbasierte Demo davon erstellen Daumenkriege. „Hier ist ein erstmals spielbarer Prototyp, den Sie in Ihrem Browser ausführen können.“ und machte sich schnell an die Arbeit. ChatGPT 5.1 brauchte erheblich länger, um den HTML-/CSS-/JavaScript-Code zu schreiben, und das Ergebnis war ein einziger Konfigurations- und Gameplay-Bildschirm, der mir anfangs gefiel (er bekommt Punkte für die Niedlichkeit), aber ich erkannte bald seine erheblichen Einschränkungen. ChatGPT Thumb Wars V1 Im Gegensatz zu Gemini waren in ChatGPT keine Desktop-Steuerelemente integriert, was bedeutete, dass ich weder eine Tastatur noch Trackpad-Tippen verwenden konnte, um die Daumen zu steuern. Ich musste ChatGPT mehr Details mitteilen, damit das funktionieren würde. Ich habe diese Aufforderung an ChatGPT übermittelt: Schöner Anfang. Nehmen wir diese Anpassungen vor Ich brauche einen realistischer aussehenden Ring Nachdem die Spieler ihren Daumenstil ausgewählt haben, wechseln sie für das Spiel zu einem Vollbild-Ring Anstelle eines Online-Gegners wird standardmäßig gegen den Computer gespielt Lassen Sie die Daumen etwas realistischer aussehen Ermöglichen Sie die Kontrolle von Tastaturangriffen. Pfeiltasten für Bewegung, X- und R-Taste für zwei Angriffsarten ChatGPT antwortete: „Ich liebe diese Verbesserungen – das fühlt sich langsam wie ein echtes kleines Kampfspiel an.“ Wir werden sehen… ChatGPT 5.1 teilte das Spiel in einen Setup- und einen Hauptspielbildschirm auf, aber es fehlte immer noch die Tiefe und Spannung des Gemini-Bildschirms. Und obwohl ChatGPT mir sagte, dass ich gegen einen CPU-Gegner antreten würde, bewegte sich der andere Daumen kaum. Ich hatte einfach nicht das Gefühl, dass sich dieses Spiel in die richtige Richtung bewegt. Wir haben nicht gevibt. Claude ist an der Reihe Als nächstes war Claude Sonet 4.5 an der Reihe. Zuvor hatte ich einige Erfolge mit Claude Artifacts erzielt, das mir bei der Entwicklung des Asteroid-Spiels geholfen hat, daher hatte ich große Hoffnungen in die Thumb Wars-Versuche. Claus war nicht weniger begeistert und sagte mir: „Ich liebe dieses Konzept! Lassen Sie mich eine Demo von Thumb Wars für Sie erstellen. Dies wird ein voll funktionsfähiger Prototyp sein mit: Charakteranpassung (Daumenauswahl, Hautton, Lucha-Libre-Masken/Outfits) Ein Ring-Spielbereich Kampf auf Tippbasis (1 Finger = Stoß, 2 Finger = Streichen) Dreistufiges Match-System Matchmaking-Simulation (in der Demo spielst du gegen einen KI-Gegner) Lassen Sie mich das für Sie erstellen:“ Dank Claudes Artifact-Fähigkeiten wurde neben dem Code sofort ein funktionierendes Spielfenster angezeigt. Die Anpassungssteuerung war großartig, insbesondere die Option, mit dem rechten oder linken Daumen zu spielen. Allerdings funktionierte das Spiel kaum, und das liegt vor allem daran, dass ich, wie ich bei den anderen Plattformen beurteilt habe, nicht genügend Informationen bereitgestellt habe, insbesondere zur Desktop-Steuerung. Ich habe versucht, eine ähnliche Eingabeaufforderung wie die, die ich für Gemini 3 bereitgestellt habe, zu verwenden, um Claude zu einer besseren, funktionaleren Version des Spiels zu führen. Das Schlüsselupdate bestand jedoch darin, dass ich darum gebeten habe, Tastatursteuerungen zuzuweisen. Ich würde Ihnen zeigen, wie es aussah, aber das Claude Thumb Wars-Update war praktisch unverändert. Es wurden nicht einmal die versprochenen Tastatursteuerungen implementiert. Gemini 3 Thumb Wars V 4 Am Ende war es kaum ein Wettbewerb. Gemini 3 Pro war schneller und intelligenter. An den Stellen, an denen ich die Grundvoraussetzung für die Anleitung war, füllte es die Lücken, um mein Traumspiel Wirklichkeit werden zu lassen. Gemini 3 Pro schien meine Absicht fast zu verstehen und lieferte im Allgemeinen angesichts der Einschränkungen das bestmögliche Ergebnis. Vielleicht werde ich mit Gemini 3 Pro zu dieser Vibe-Coding-Aufgabe zurückkehren, und vielleicht werden auch Sie eines Tages Thumb Wars auf dem Desktop (und später auf dem Smartphone!) spielen.

https://newfortech.com/?p=27081 This article is part of NewForTech’s AI Week 2025. We’ll cover the basics of AI and show you how to get the most out of ChatGPT, Gemini or Claude, as well as detailed features, news and key talking points in the AI ​​world. Artificial intelligence has conquered the world. It seems like every product has some form of artificial intelligence these days, from smartphones to refrigerators. very simple Have have it. If you really want to, you can even get a massage from an AI-controlled robot! When we talk about AI, we usually talk about chatbots. These are the famous big language models (LLM) that let you talk or help you organize your thoughts, your life and even your finances. It wasn’t long ago that AI chatbots were boring conversation partners and image generators were notorious for giving people too many fingers or melting their faces. But in just a few years, they have improved so much that it is increasingly difficult to distinguish between what is real and what is generated by artificial intelligence. Some companies are now able to produce fully synthetic videos that look realistic and convincing. Most people use AI to rewrite or summarize texts and solve problems, but another important feature of chatbots is their ability to translate languages. This can be very useful when traveling abroad. Chatbots are also surprisingly useful for demystifying legalese or helping you understand what your doctor really means by all that medical terminology. You can also search the internet, book plane tickets and order more coffee if you run out of coffee. And let’s not forget their role as companions. While no AI should replace a trained professional, these tools can quickly provide you with information or advice if you have problems at work or in your personal life. How does artificial intelligence work? LLMs can’t really “think” like humans, but they are trained with so much human-generated data (words and images) that they are so good at pattern matching. is displayed as if they think just like a human. While there are options like Google Gemini, Microsoft Copilot, Grok from X, Meta AI from Meta, and Claude (which Amazon is investing in), by far the most popular AI model is ChatGPT from OpenAI. Although Microsoft has made significant investments, OpenAI remains an independent company and is at the forefront of many innovations in the industry. Which AI you prefer to use is mostly a matter of personal preference. Try them all: Most people trust the person who is easiest for them to talk to and who is most helpful to them. The history of chatbots with artificial intelligence It’s easy to forget that artificial intelligence, at least in the form of chatbots, is still very young. ChatGPT-3.5, released in 2022, is the version that catapulted the technology into the public consciousness. With ChatGPT-4 – and its variants like 4o and 4o-mini – the chatbot took a big step forward, especially when OpenAI added the ability to create native images without relying on another application. The images created by ChatGPT quickly went viral, especially when people discovered they could imitate Studio Ghibli’s style, attracting even more users. The addition of web search capabilities marked a new milestone for chatbots. Several studies have shown that internet searches are the most common use of chatbots among the general public today. What can AI chatbots do? The latest version of ChatGPT (at the time of writing) is version 5, known as ChatGPT-5, which was released in August 2025. It’s an incredibly powerful AI that’s great at writing code, but can also handle the fun things in life, like ordering pizza or writing a bedtime story for your kids. GPT-5 comes in two main versions depending on your ChatGPT account type: Quickly AND Think. There is also one professional Version available for ChatGPT Pro subscribers. Perhaps the most powerful feature of today’s chatbots is their ability to behave like chatbots. agents. ChatGPT Agent mode It turns it into your personal assistant, which can work in the background all day with the tasks you assign. Copilot, Gemini and GPT-5 are free, but paid versions unlock advanced features like Agent Mode and offer higher usage limits. You can access all these chatbots via mobile apps (for iOS and Android), in a web browser or via a desktop app for Mac and PC if your computer’s processor is powerful enough. One of the most interesting features of chatbots is their voice interface. In ChatGPT it says GPT voice chat; in Gemini it is like that Gemini live. This allows you to speak directly to the chatbot instead of typing, creating a much more natural conversation. You also have a variety of voices to choose from, giving your digital companion a personality. a warning This article was originally published as part of ChatGPT Basics Guidenow available in. ChatGPT Basics Guide is packed with great tips and advice from experts, covering everything you need to know about the world’s most popular AI, from the basics to some of the latest features. All AIs sometimes make mistakes and, worse, sometimes make things up. It is a phenomenon that occurs so often that it has its own name: Hallucinations. To be useful, an AI can sometimes make up facts that don’t exist. This is why you should never take everything said literally. They may even agree with your conclusion, which can be misleading. In general, chatbots are safe to use. AI companies take the issue of appropriate content seriously, so you’re very unlikely to encounter anything that upsets, endangers, offends or scares you. However, all AIs have age restrictions and should not be used by children without supervision. For example, OpenAI states that “ChatGPT is not intended for use by children under the age of 13, and we require children between the ages of 13 and 18 to obtain parental consent before using ChatGPT.” Using chatbots is generally a fun and enjoyable experience and there is a lot to explore. Once you start integrating AI into your daily life, you may wonder how you ever lived without it.

https://newfortech.com/?p=27027 Insta360 has just given its high-end action camera, the Ace Pro 2, a major boost with a range of new accessories. On the agenda: a photographer’s grip, three lenses to vary the views and an unexpected (and probably unnecessary) addition: an instant clip-on printer. Everything can be seen in the picture above. Some accessories seem quite logical, like the new film, ultra-wide and macro lenses. Inspired by the Hero 13 Black and its modular lenses, a feature described as “the future of action camera design”, it will be possible to create anamorphic clips with the film lens or even shoot up close with the macro lens. GoPro may have reason to be concerned. Also new: Xplorer Grip Pro: a case with built-in battery and ergonomic grip (see below). This is the type of accessory that some Chinese smartphone brands have been offering for years to improve photo handling, such as the successful Xiaomi 15 Ultra photo set. Insta360 seems to want to make the Ace Pro 2 both a photography tool and an 8K action camera. For those looking for a versatile model, the positioning seems appropriate. Perhaps the most surprising addition is the detachable pocket printer. Insta360 presents it as the first printer designed for an action camera and can provide instant memories at any time. Print 3 inch (54 x 82 mm, printable area 54 x 72 mm) photos using the dye sublimation process. According to Insta360, the screen matches the dynamics of the Ace Pro 2’s videos. The recordings are also resistant to water, dust and light. Action cameras are rarely associated with photography, let alone direct printing. This accessory seems to fall into the “try before you judge” category.   Genius or useless gadget? Compared to more expensive models from GoPro, DJI and Insta360, the Hero Black 13’s interchangeable lens functionality remains by far the most promising action camera innovation. Thanks to three new lenses clearly inspired by GoPro, you can now take advantage of both approaches with the Ace Pro 2. Photography is not the Ace Pro 2’s specialty, focusing primarily on video. However, the new grip greatly improves grip when shooting, and Leica’s new color profiles should ensure convincing print-ready reproduction straight from the camera. As for the printer… opinions are still divided. While pocket printers like the Instax Mini Link are great for printing on the go, they aren’t necessarily suitable for outdoor adventures. Several packages are now offered with this new accessory. The XPlorer Pro package includes the new grip and an extra battery for €529.99. The Flash Print package includes the pocket printer, a Flip Screen sun visor and a classic leather case for €582. The Ultimate package also includes all three purposes for a total of €733. For those who don’t want to lug around everything, there are also other single lens combinations. All of these packs will be available upon release on November 12th. Equipped with all these accessories, is the Ace Pro 2 the ultimate combination for action camera enthusiasts? Reviews are open. Check Insta360 GO 3S.  

https://newfortech.com/?p=26069 Two former Meta employees launch a new AI-powered smart ring. Stream Ring is the first product from sandbank, and can be reserved now. Sandbar describes the Stream as “your extended self,” meaning it’s an intentionally minimalist smart ring that lets you take voice notes and interact directly with a chatbot using its built-in haptic microphone. When you create a voice memo, Stream Ring uses haptic feedback to confirm that the note has been recorded. Press and hold the sensor to talk and tap it to pause and start over. It can automatically transcribe your voice interactions, whether you’re just making a shopping list on the go, asking the device to pull information from the web, or having a more casual conversation with the device. These notes will appear via Bluetooth in the Stream app, which will only be available for iOS at launch. Sandbar believes that a ring is the best way to dress because it’s always available and accessible no matter what you’re doing, making it easier to catch a thought as soon as it strikes you. He wants people to think of the Stream ring as a “voting mouse” and says the microphone can still pick up your voice clearly in a noisy room. In short, it doesn’t always listen, it only activates when you press and hold the touchpad. By default, the chatbot is designed to use artificial intelligence to somewhat mimic your voice, a feature Sandbar calls “Inner Voice.” You can update it if you think it’s wrong or if the whole concept scares you, you can also switch to a non-custom voice. In addition to its primary voice memo function, the Stream Ring can also be used to control media playback using gestures and is compatible with all headsets, says Sandbar. Sandbar offers a free plan with unlimited comments, but limits interaction with the AI. For $10 per month, you can upgrade to a Stream Pro subscription (get three months free with a new purchase), which includes unlimited chats and instant access to all new features. Sandbar claims that your data is encrypted at rest and in transit and that your data is not sold to other companies. In addition, you can delete the data stored in the app at any time. Other smart rings promise a seamless interactive experience powered by AI, and the Stream ring doesn’t follow the health monitoring path that many other products follow in this increasingly crowded space market. But if the transcription feature works as reliably as advertised, and Sandbar’s AI turns out to be a genuinely useful assistant, the Stream ring could be a useful accessory that stays on the side when not needed. The Stream ring is available in sizes 5 to 13 and can be worn on the index finger. Sandbar promises all-day battery life, but doesn’t go into details. Delivery to the US is expected next summer and will cost $249.

https://newfortech.com/?p=26101 Mozilla Firefox makes it difficult for websites to track “digital footprints”. Digital information, such as hardware specifications, is collected through websites. This information can then track the user’s activity, even when browsing privately. When you use a web browser, it is common for websites to ask you for permission to use tracking cookies to collect data about your browsing habits, usually with the option to refuse. But “fingerprints” are much harder to block than cookies, and websites can track your activity using your device’s hardware specifications, time zone and other information you provide to improve website performance and functionality. When this information is combined into a profile, it can track your browsing habits even in private browsing mode, but Mozilla Firefox is now introducing fingerprint protection measures that would reduce the number of uniquely identifiable users by around 70%. register fingerprints The new safeguards have been introduced as part of Firefox’s “Enhanced Tracking Protection” service. Firefox145. Protection is currently only available in Strict Private Browsing and Enhanced Tracking Protection (ETP) browser modes. However, tests in these environments will soon extend protection to normal navigation. Many websites use invasive scripts that collect information from your browser and device that can be used for tracking purposes. The scripts are often executed even if you refuse cookies. These scripts collect data such as browser version, operating system, screen resolution and color settings, system language, fonts, time zone, GPU processing behavior, CPU cores, touch screen capabilities, and device memory. Enhanced Tracking Protection already offers a number of anti-fingerprint measures. “Since 2021, Firefox has gradually evolved fingerprint protection to cover the most common fingerprint detection techniques,” Mozilla explains. “This includes things like how your graphics card draws images, what fonts your computer has, and even minor differences in how it performs calculations.” To counter other fingerprinting techniques, Mozilla introduced new mechanisms such as adding random noise to background images when reading the website, enforcing the use of default operating system fonts (except for language characters), and hiding touch support, screen resolution, and the number of CPU cores. These new safeguards can reduce the number of users with unique fingerprints to just 20%, while more than 60% of users could be fingerprinted without any protection. Unfortunately, it is unlikely that a browser can reduce this number to 0%, as many websites rely on fingerprint information for important website functions and features. IN BeepTeam

https://newfortech.com/?p=26088 The Coros Pace 4 smartwatch has just hit the market It offers an AMOLED screen and better pulse accuracy. The price has increased slightly compared to the Coros Pace 3 If you like stylish and light racing watches, the Coros Pace series has long offered stiff competition. There was a lot to like about the Pace 3, but it faces stiff competition from models like the Garmin Forerunner 165 and Fitbit Versa 3. Can the Coros keep up? We’re about to find out, as the company just announced an update to its popular Pace series. HE The rhythm of the choir 4 It comes with a number of improvements, but still has the lightweight chassis of its predecessor and maintains its weight of 32 grams when paired with a nylon strap. The weight should be light enough not to weigh you down on the trail. One of the biggest improvements to the Pace 4 is the addition of a 1.2-inch AMOLED touchscreen, bringing it closer to the Coros Pace Pro watch. When testing the Pace 3, we criticized the mineral glass LCD screen, which can be a little hard to see in sunlight. With the addition of AMOLED technology, you should be able to enjoy a brighter and clearer screen in all conditions. Coros says the watch’s heart rate accuracy has also improved, giving more accurate readings that should better reflect actual heart health. This is thanks to a redesigned heart rate sensor that Coros says can reduce external peaks and troughs in readings. Even longer battery life In addition, Pace 4 has voice-activated control via a built-in microphone. This means you can now record exercise logs, record voice pins, jot down reminders and more without touching the watch screen. Battery life has always been a strong point of the Pace 3 and the Pace 4 has improved this further. After 41 hours, you get better performance from the Pace 4 when connected to a five-satellite network than from the Pace 3 when connected to two satellites (25 hours). As far as GPS training goes, this is pretty impressive. However, not everything has changed. The design remains the same as the Pace 3, while features such as the Virtual Pacer and race predictions carry over from the previous model. We praise the large number of parameters that the Pace 3 offers, so it’s good to see many of them in the new Pace 4. As for the price, the Coros Pace 4 costs $249 (£229) and is available now. This is a slight increase on the Pace 3’s $229/£219 price, but given that affordability was one of the previous model’s main selling points, we’re glad the price hasn’t gone up too much.

https://newfortech.com/?p=26084 After seeing all the hype surrounding Google’s latest generative AI tool, I knew I had to try it out. The model is called Gemini 2.5 Flash Image. Updates Your ability to edit your photos in Gemini. AI enthusiasts call it the “nano-banana” model, driven by a series of banana-themed breakthroughs. Googling management. In the few weeks since its launch, people have created more than 200 million AI images and more than 10 million people have signed up to use the Gemini app. After to Josh Woodward, VP of Google Labs and Gemini at Google. Google has invested heavily in its generative media models this year, introducing updated versions of its image and video generator models at its annual I/O developer conference. Google’s AI video generator, Veo 3, surprises with synchronized sound, a first among the AI ​​giants. And YouTubers have created more than 100 million AI videos using Flow, Google’s AI video tool. I’ve spent a lot of time testing AI design software and was excited to see what Google came up with. But my testing of Flash Image 2.5 has shown that just because something has a prominent mention doesn’t mean it will always meet your expectations. This was my experience with the Gemini Nano Banana: the good, the bad and the frustrating. What worked? The Gemini Banana template is incredibly good at adding elements to existing images and integrates AI-generated elements well into any image you create. Also, character consistency remains fairly stable, meaning the people in my photos weren’t too distorted or shaky after the AI ​​processing. Both are important distinctions for AI imaging programs and something Google says it’s working to improve. You can see both features in this photo of me and my sister. Our overall look remained unchanged in the edited version (right), showing the consistency of the characters. I asked Gemini to add a third sister who looked like the two of us, which he surprisingly did and added a third wife between the two of us. I was also very impressed with how quickly Gemini was able to produce finished images. Anything under a minute earns me a gold star, and Gemini routinely processes requests in less than 15 seconds. I also appreciate that you added a watermark to all the images you created and edited. While I don’t like how tech companies have corrupted shiny emojis for AI, having labels for AI-generated content is extremely important. Google’s SynthID and behind-the-scenes work also help distinguish AI content from human-generated images. Gemini is also good for creating full AI images, but I recommend using Image 4 or another AI imager – these have more convenient controls and settings that allow you to get closer to what you want with less effort. Which really didn’t work Gemini bananas have serious limitations. It automatically generates square images and follow-up requests to scale images to other sizes were ignored or failed. I also noticed that Gemini lowered the resolution of many of my images. I mainly shoot with my iPhone 16 which has excellent cameras, but after testing the Gemini banana model these small details were often blurry. It’s boring and won’t convince any photographer. I have tried several times to ask Gemini to make changes to pictures that I would have had difficulty doing manually. This is one area of ​​photo editing where AI should excel: automating trivial but detail-intensive edits. Unfortunately, the Gemini had trouble following quickly, so it didn’t do what I asked. I tried several times to get the Twins to remove the reflections from a cutout of the Freakier Friday movie poster, but they were still there. And the more I tried to remove reflections, the worse the image quality got with each request. The once-clear text became illegible after I finally gave up, not to mention the odd accidental damage to Lindsey Lohan and Jamie Lee Curtis’ faces. The Gemini nanobanana had difficulty creating images of different sizes. Resizing and cropping images is an important photo editing process, but Gemini didn’t or couldn’t include simple resizing guidelines in my instructions. I reached out to Google about the fix and sizing issues, and a spokesperson said the tech company was “aware of both issues and actively working on them. This was a significant update from our previous model, but we will continue to improve the model.” Overall, the “Gemini Nano Banana” showed me that Google is serious about maintaining its dominance in the generative media space. But it comes with significant drawbacks as too much emphasis is placed on generating new assets rather than using AI to improve and solve common photography problems. For now, the Nano Banana model is best suited for Gemini fans who want to make big changes quickly. Those of us looking for more accurate tools will have to wait for the next big Google update or look for another program. Gemini Nano Banana Availability, Price and Privacy Policy You don’t need to do anything to access the new template; It is automatically added to the Gemini 2.5 Flash base model. Gemini is available for free. Additional templates and higher usage limits are available in Google AI plans from now on $20 per month. If you are a paid subscriber, you may also be able to access the model through Google AI Studio. From there, all you have to do is upload an image and enter your message. Each message uses between one thousand and two thousand tokens, depending on the level of detail required. Adobe Express and Firefly users can now also access the new template. Googling Gemelli’s privacy policy claims it can use the uploaded information to improve its AI products, so the company recommends avoiding uploading sensitive or private information. The company Prohibited Use of AI Policy It also prohibits creating illegal or offensive material. To learn more, check out the best AI image generators and everything announced at the Made by Google Pixel 10 event.

https://newfortech.com/?p=26043 The first paid DLC Pokemon Legends: ZA coming on December 10thjust seven weeks after the game first hit stores. mega size It costs $30, it is possible fill up with gas in flames “Nintendo is charging too much.” However, there are a lot of interesting things in this update for Switch and Switch 2. First, there is a long story behind the game. The current title isn’t lacking in post-game content, but it doesn’t have much story to offer. It will be a good way to reconnect with the various residents of Lumiose City. Speaking of which, the trailer features one or two iconic returning characters. Pokémon AND YES which had disappeared FOR. The story would involve a journey to another dimension thanks to the legendary Pokémon Hoopa. This means players will experience a different version of Lumiose City, with tons of new Pokémon to catch. These include creatures not actually native to the region, including notable species such as Baxcalibur and Corviknight. Pokemon can also surpass level 100 in DLC, giving players even more reasons to go beyond Shining Hunt. We don’t know if the interdimensional city of Lumiose will include new collectibles, such as an updated set of hidden colored vines. I really hope so. I love doing parkour around the city. This is not the only DLC available for the game. Game Freak just released a small free update with a new sidequest. Use the game’s Mystery Gift feature to unlock it. Pokemon Legends: ZA It’s a really fun game, despite what people try to tell you online. You are not entirely wrong. He might be crazy. The graphics don’t exactly scream “AAA title 2025” and the tutorial section is pretty dull. The city could use a little more life and the map can be confusing. It’s still a lot of fun, though Wild Zone 17 will probably haunt my dreams for years to come.