us-technology

Xbox and PlayStation consoles shown with dollar symbol (Image source: Microsoft Xbox Gaming, Sony PlayStation with edits) Gamers are debating whether the current generation of Xbox consoles is successful. A pair of insiders has offered conflicting sales estimates ranging from less than 30 million to 37 million units. Using an AMD fiscal report, leaker KeplerL2 contradicts more positive projections for Xbox console sales. Recently, insiders have shared differing estimates for lifetime Xbox Series X and S console sales. Analyst Welfare revealed that Microsoft has sold around 37 million gaming systems worldwide. As sales of the consoles have slowed, some gamers doubt the figure. Most notably, the renowned leaker KeplerL2 suggests that a more accurate projection is under 30 million consoles. KeplerL2’s assertion relies on an AMD fiscal report for Q4 2024, which included over 100 million current-gen console sales. The company tracks how many of its chips are sold for use in the PS5, Xbox Series X|S, and even the Steam Deck. Earlier in 2025, Sony confirmed that it had sold slightly over 75 million PS5 systems before the end of 2024. Subtracting 75 million from the 100 million, and an estimated 4 million Steam Deck sales, leaves 21 million transactions for Xbox consoles. Much of the confusion stems from conflicting reports distributed by manufacturers and game publishers. In February 2025, Take-Two Interactive reported that at least 94 million current-gen consoles had been sold by the close of November 2024. The PS5 accounted for 65.6 million of those sales through September. Considering additional sales by the end of November, Microsoft may have moved as many as 27 million Xbox Series X|S systems. KeplerL2 concludes that the most optimistic calculation for Xbox console sales is below Welfare’s 37 million projection. Yet, the truth may lie somewhere between the two numbers. VGChartz uses various retail sources to produce its rankings. Its most recent update had tracked more than 32 million Xbox Series X|S sales. Followers watch KeplerL2 because of leaks about consoles in development, including PS6 specs. As a result, some of his critics think that sales analysis is outside his comfort zone. Still, it’s clear that PlayStation’s advantage over the struggling Xbox consoles will only grow in the coming months. Related Articles Adam Corsetti – Tech Writer – 501 articles published on Notebookcheck since 2025 I became interested in technology at a young age and enjoyed discovering the latest innovations. While earning college degrees in publishing, I created several PC hardware and gaming websites. My passion has always been to guide readers on what products can truly improve their lives. After many years as a Tech Writer for Game Rant, I’m anxious to share my knowledge with a new audience at Notebookcheck. Adam Corsetti, 2025-08- 3 (Update: 2025-08- 3) Read More

Specifications Display 18.00 inch 16:10, 3840 x 2400 pixel 252 PPI, IPS, glossy: yes, 240 Hz Connections 4 USB 3.1 Gen2, 2 Thunderbolt, USB-C Power Delivery (PD), 1 HDMI, Audio Connections: 3.5mm Networking 10/100/1000 LAN Card (10/100/1000MBit/s), 802.11 a/​b/​g/​n/​ac/​ax/​be (a/b/g/n = Wi-Fi 4/ac = Wi-Fi 5/ax = Wi-Fi 6/ Wi-Fi 6E 6 GHz be = Wi-Fi 7), Bluetooth 5.4 Size height x width x depth (in mm): 27.95 x 403 x 296.9 ( = 1.1 x 15.87 x 11.69 in) Battery 99.9 Wh Lithium-Ion Operating System Microsoft Windows 11 Home Camera Webcam: IRPrimary Camera: 5 MPix Additional features Keyboard: Chiclet, Keyboard Light: yes Weight 3.5 kg ( = 123.46 oz / 7.72 pounds) ( = 0 oz / 0 pounds) Price comparison Reviews for the Lenovo Legion 9 18IAX10 The Fastest Gaming Laptop…sortaSource: Hardware Canucks The fastest gaming laptop we’ve ever tested is here and its the Legion 9i which has a brand new design for 2025. With an RTX 5090 and Intel Ultra 9 275HX, we put this up to some of the best gaming laptops with AMD and Intel CPUs alongside the RTX 4090. Well it turns out that the Legion 9i is amazing in a lot of ways but it falls short in others. Single Review, online available, Very Long, Date: 05/08/2025 ▶ load Youtube video Comment NVIDIA GeForce RTX 5090 Laptop: The GeForce RTX 5090 Mobile uses the GB203 chip and utilizes 10,496 shaders / CUDA cores of the maximum 10,752 that the chip has. This makes the RTX 5090 Laptop more similar to the RTX 5080 desktop variant (same chip) than the Desktop RTX 5090 with 20,760 cores and higher clock rates. These graphics cards are able to play the latest and most demanding games in high resolutions and full detail settings with enabled Anti-Aliasing. » Further information can be found in our Comparison of Mobile Graphics Cards and the corresponding Benchmark List. Ultra 9 275HX: Mobile high-end CPU for gaming notebooks based on the Arrow Lake architecture. Offers 24 cores consisting of 8 fast performance cores with up to 5.4 GHz and 16 smaller efficiency cores with up to 4.6 GHz clock speed. The CPU can access 40 MB L2 cache and 36 MB L3 cache and is specified with a TDP of 55 watts. The SoC integrates a small dedicated NPU called AI Boost with 13 TOPS (Int8) and optionally supports vPro Essentials.» Further information can be found in our Comparison of Mobile Processsors. 18.00″: This display size is a standard format for desktop replacements (DTR). The DTR laptops are heavier to carry, need more power, but texts are easy to read and high resolutions are no problem. DTR are mainly intended for stationary desk use, where weight and energy hunger don’t matter. In return, you enjoy the advantages of high resolutions (more details, better legibility).» To find out how fine a display is, see our DPI List. 3.5 kg: A laptop with this weight is comparatively heavy and less designed for mobility than for use at the desk. Therefore, the devices tend to shine less with battery life than with a large screen and higher performance. Lenovo: Lenovo (“Le” from English legend, novo (Latin) for new) was founded in 1984 as a Chinese computer trading company. As of 2004, the company was the largest laptop manufacturer in China and, after acquiring IBM’s PC division in 2005, the fourth largest in the world. In addition to desktops and notebooks, the company manufactures monitors, projectors, servers, etc, and specializes in developing, manufacturing and marketing consumer electronics, personal computers, software, enterprise solutions and related services. In 2016, the company ranked first in the world in computer sales. It still held it in 2023 with about 23% global market share. Important product lines are Thinkpad, Legion and Ideapad. In 2011, it acquired a majority stake in Medion AG, a European computer hardware manufacturer. In 2014, Motorola Mobility was purchased, which gave Lenovo a boost in the smartphone market. » Further information can be found in our Notebook Purchase Guide. Similar devices from a different Manufacturer Devices from a different Manufacturer and/or with a different CPU Devices with the same GPU and/or Screen Size Devices with the same GPU Devices with Same Screen Size and/or Weight Devices from the same Manufacturer Stefan Hinum, 2025-08- 3 (Update: 2025-08- 3) Read More

Official MindsEye game image. (Image source: PlayStation) Although promising, this GTA-like game is now destined to plunge into the void of the video game world. And that’s despite the fact that new updates are expected to appear in the near future. The world of video games sometimes shows no mercy to certain titles that showed great promise. And such is the case with MindsEye, which promised many features, but is currently in a very difficult situation, despite strenuous efforts on the part of the Build A Rocket Boy studio.  Indeed, with a score of 37 on Metacritic and thousands of negative reviews on Steam, this game seems to be heading straight for the dustbin of the industry. To understand this, you need to know that just one month after its release, gamers have begun to desert this game.  According to data provided by SteamDB, only 30 players have played the game in the last 24 hours. And that’s not all, because as you know, video games are often broadcast on platforms such as Twitch, which will be offering rewards for the release of GTA VI. However, here too, scores are low, peaking at 51 viewers over the previous 24 hours.   In other words, the game is currently light years ahead of its best statistics, which weren’t all that high. It has to be said that its highest peak of players is 3 302 on Steam and 138 686 viewers on Twitch. All this happened on release. But that’s not all, as several content additions have been postponed. This is particularly true of the crossover with Hitman, which was postponed to a later date, much to the disappointment of gamers. However, Build A Rocket Boy remains confident about the future of its GTA-like game. Indeed, updates are expected to appear in the coming months, which could breathe new life into this title that is currently hurtling down an endless abyss. But it remains to be seen whether the game can turn its head and become innovative. Today, it already seems to have been forgotten by gamers, not to mention that negative reviews are pouring in on the web.  Related Articles Alexis Stegmann – Tech Writer – 136 articles published on Notebookcheck since 2025 I’ve been working in the field of web writing for several years, and I’m passionate about keeping readers up to date with the latest news on astronomy, technology, the world of video games and other exciting subjects. In particular, I’ve had the opportunity to work on a number of websites, which has enabled me to cover a wide range of subjects. In my personal life, I’m passionate about a wide range of subjects, including astronomy, video games, history and science. I’m also drawn to psychology, which is a subject that deserves greater documentation and recognition. Alexis Stegmann, 2025-08- 2 (Update: 2025-08- 2) Read More

Prophet Security, a startup developing autonomous artificial intelligence systems for cybersecurity defense, announced Tuesday it has raised $30 million in Series A funding to accelerate what its founders describe as a fundamental shift from human-versus-human to “agent-versus-agent” warfare in cybersecurity. The Menlo Park-based company’s funding round, led by venture capital firm Accel with participation from Bain Capital Ventures, comes as organizations struggle with an overwhelming volume of security alerts while sophisticated attackers increasingly leverage AI to scale and automate their operations. Prophet’s approach represents a marked departure from the “copilot” AI tools that have dominated the market, instead deploying fully autonomous agents that can investigate and respond to threats without human intervention. “Every security operations team is faced with a dual mandate of reducing risk while driving operational efficiency,” said Kamal Shah, Prophet Security’s co-founder and CEO, in an exclusive interview with VentureBeat. “Our Agentic AI SOC Platform addresses both challenges by automating manual, repetitive tasks in security operations with speed, accuracy and explainability.” The funding announcement coincides with Prophet’s launch of what it calls the industry’s most comprehensive Agentic AI SOC Platform, expanding beyond its initial Prophet AI SOC Analyst to include Prophet AI Threat Hunter and Prophet AI Detection Advisor. The platform represents a significant evolution from traditional Security Operations Center (SOC) automation tools, which typically rely on rigid, pre-programmed playbooks. The AI Impact Series Returns to San Francisco – August 5 The next phase of AI is here – are you ready? Join leaders from Block, GSK, and SAP for an exclusive look at how autonomous agents are reshaping enterprise workflows – from real-time decision-making to end-to-end automation. Secure your spot now – space is limited: https://bit.ly/3GuuPLF Security teams drowning in 960 daily alerts face unprecedented capacity crisis The cybersecurity industry faces a crisis of capacity and capability. Shah, who previously served as CEO of container security company StackRox before its acquisition by Red Hat, experienced these challenges firsthand. According to his observations, organizations receive an average of 960 security alerts daily, with up to 40% going uninvestigated due to resource constraints. “The number one complaint that I see from customers every single day is too many alerts, too many false positives,” Shah explained. “If you think about the world that we live in today, on average, a company gets 960 alerts a day from all the security tools that they have in their environment, and 40% of those alerts are ignored because they just don’t have the capacity to go and investigate all those alerts.” The problem is compounded by a severe shortage of skilled cybersecurity professionals. Shah points to what he calls a critical talent gap, noting there are 5 million open positions in cybersecurity globally, creating a situation where even organizations with budget to hire cannot find qualified personnel. Prophet’s solution directly addresses this capacity crunch. Over the past six months, the company’s AI SOC Analyst has performed more than 1 million autonomous investigations across its customer base, saving an estimated 360,000 hours of investigation time while delivering 10 times faster response times and reducing false positives by 96%. How autonomous AI agents differ from reactive copilot systems transforming cybersecurity The distinction between Prophet’s “agentic” AI and the copilot models deployed by larger cybersecurity vendors like CrowdStrike, Microsoft, and Sentinel One is fundamental to understanding the company’s value proposition. Traditional copilot systems require human analysts to initiate queries and interpret responses, essentially serving as sophisticated search interfaces for security data. “Copilot is reactive,” Shah explained. “You have an alert come in and a security analyst has to go and write questions, ask the question to say, hey, what does this mean? And you have to know what questions to ask. The analyst is still in the loop for every single alert that comes in because they’re interacting with it.” By contrast, Prophet’s agentic AI proactively initiates investigations the moment an alert is triggered, autonomously gathering evidence, reasoning through the data, and reaching conclusions without human intervention. The system documents every step of its investigation process, creating an audit trail that allows security teams to understand and verify its reasoning. “What Prophet AI is able to do is immediately, once an alert is triggered, it proactively goes and completes the investigation,” Shah said. “Within a matter of minutes, your investigation is complete and it knows what questions to ask, and it’s been trained to act like an expert analyst.” Building enterprise trust through transparent AI decision-making and data protection Prophet’s system leverages multiple frontier AI models, including offerings from OpenAI, Anthropic, and others, selecting the most appropriate model for each specific task. The company has built what Shah describes as an “evals framework” to ensure accuracy, repeatability, and consistency while preventing AI hallucinations—a critical concern in security contexts where false information can lead to inappropriate responses. “In security, you are in a trust building exercise with the security teams, and if you hallucinate, you’re going to lose trust and they’re not going to use your product,” Shah emphasized. The company employs a retrieval-augmented generation (RAG) architecture combined with rigorous evaluation processes to maintain what Shah calls “a high bar for security teams.” Data privacy and security represent paramount concerns for Prophet’s enterprise customers. The company employs a single-tenant architecture ensuring customer data remains isolated, and maintains contractual agreements with AI model providers preventing customer data from being used to train or fine-tune models. Early customers report dramatic efficiency gains as AI handles thousands of security alerts Prophet’s customer base includes Docker, which provided a testimonial for the funding announcement. Tushar Jain, Docker’s EVP of Engineering and Product, noted that “Prophet AI is already helping streamline parts of our security workflow, and we’re just getting started. With the recent release of Threat Hunter and growing integration with our systems, we see a clear path to faster response times, reduced noise, and a more focused security team.” The company has also published case studies demonstrating dramatic improvements in SOC efficiency. Eric Wille, CISO at Cabinet Works,

Designed for real-world enterprise use, the 4.5-billion-parameter model — much smaller than the tens of billions to trillions of leading frontier models — combines cost efficiency, regulatory compliance, and strong performance in a compact footprint. AFM-4.5B was one of a two part release made by Arcee last month, and is already “instruction tuned,” or an “instruct” model, which is designed for chat, retrieval, and creative writing and can be deployed immediately for these use cases in enterprises. Another base model was also released at the time that was not instruction tuned, only pre-trained, allowing more customizability by customers. However, both were only available through commercial licensing terms — until now. Arcee’s chief technology officer (CTO) Lucas Atkins also noted in a post on X that more “dedicated models for reasoning and tool use are on the way,” as well. The AI Impact Series Returns to San Francisco – August 5 The next phase of AI is here – are you ready? Join leaders from Block, GSK, and SAP for an exclusive look at how autonomous agents are reshaping enterprise workflows – from real-time decision-making to end-to-end automation. Secure your spot now – space is limited: https://bit.ly/3GuuPLF “Building AFM-4.5B has been a huge team effort, and we’re deeply grateful to everyone who supported us We can’t wait to see what you build with it,” he wrote in another post. “We’re just getting started. If you have feedback or ideas, please don’t hesitate to reach out at any time.” The model is available now for deployment across a variety of environments —from cloud to smartphones to edge hardware. It’s also geared toward Arcee’s growing list of enterprise customers and their needs and wants — specifically, a model trained without violating intellectual property. As Arcee wrote in its initial AFM-4.5B announcement post last month: “Tremendous effort was put towards excluding copyrighted books and material with unclear licensing.” Arcee notes it worked with third-party data curation firm DatologyAI to apply techniques like source mixing, embedding-based filtering, and quality control — all aimed at minimizing hallucinations and IP risks. Focused on enterprise customer needs AFM-4.5B is Arcee.ai’s response to what it sees as major pain points in enterprise adoption of generative AI: high cost, limited customizability, and regulatory concerns around proprietary large language models (LLMs). Over the past year, the Arcee team held discussions with more than 150 organizations, ranging from startups to Fortune 100 companies, to understand the limitations of existing LLMs and define their own model goals. According to the company, many businesses found mainstream LLMs — such as those from OpenAI, Anthropic, or DeepSeek — too expensive and difficult to tailor to industry-specific needs. Meanwhile, while smaller open-weight models like Llama, Mistral, and Qwen offered more flexibility, they introduced concerns around licensing, IP provenance, and geopolitical risk. AFM-4.5B was developed as a “no-trade-offs” alternative: customizable, compliant, and cost-efficient without sacrificing model quality or usability. AFM-4.5B is designed with deployment flexibility in mind. It can operate in cloud, on-premise, hybrid, or even edge environments—thanks to its efficiency and compatibility with open frameworks such as Hugging Face Transformers, llama.cpp, and (pending release) vLLM. The model supports quantized formats, allowing it to run on lower-RAM GPUs or even CPUs, making it practical for applications with constrained resources. Company vision secures backing Arcee.ai’s broader strategy focuses on building domain-adaptable, small language models (SLMs) that can power many use cases within the same organization. As CEO Mark McQuade explained in a VentureBeat interview last year, “You don’t need to go that big for business use cases.” The company emphasizes fast iteration and model customization as core to its offering. This vision gained investor backing with a $24 million Series A round back in 2024. Inside AFM-4.5B’s architecture and training process The AFM-4.5B model uses a decoder-only transformer architecture with several optimizations for performance and deployment flexibility. It incorporates grouped query attention for faster inference and ReLU² activations in place of SwiGLU to support sparsification without degrading accuracy. Training followed a three-phase approach: Pretraining on 6.5 trillion tokens of general data Midtraining on 1.5 trillion tokens emphasizing math and code Instruction tuning using high-quality instruction-following datasets and reinforcement learning with verifiable and preference-based feedback To meet strict compliance and IP standards, the model was trained on nearly 7 trillion tokens of data curated for cleanliness and licensing safety. A competitive model, but not a leader Despite its smaller size, AFM-4.5B performs competitively across a broad range of benchmarks. The instruction-tuned version averages a score of 50.13 across evaluation suites such as MMLU, MixEval, TriviaQA, and Agieval—matching or outperforming similar-sized models like Gemma-3 4B-it, Qwen3-4B, and SmolLM3-3B. Multilingual testing shows the model delivers strong performance across more than 10 languages, including Arabic, Mandarin, German, and Portuguese. According to Arcee, adding support for additional dialects is straightforward due to its modular architecture. AFM-4.5B has also shown strong early traction in public evaluation environments. In a leaderboard that ranks conversational model quality by user votes and win rate, the model ranks third overall, trailing only Claude Opus 4 and Gemini 2.5 Pro. It boasts a win rate of 59.2% and the fastest latency of any top model at 0.2 seconds, paired with a generation speed of 179 tokens per second. Built-in support for agents In addition to general capabilities, AFM-4.5B comes with built-in support for function calling and agentic reasoning. These features aim to simplify the process of building AI agents and workflow automation tools, reducing the need for complex prompt engineering or orchestration layers. This functionality aligns with Arcee’s broader strategy of enabling enterprises to build custom, production-ready models faster, with lower total cost of ownership (TCO) and easier integration into business operations. What’s next for Arcee? AFM-4.5B represents Arcee.ai’s push to define a new category of enterprise-ready language models: small, performant, and fully customizable, without the compromises that often come with either proprietary LLMs or open-weight SLMs. With competitive benchmarks, multilingual support, strong compliance standards, and flexible deployment options, the model aims to meet enterprise needs for speed,

Skip to content Image: Willis Lai/IDG I’ve wanted a Steam Deck from the first moment I saw one. But I haven’t bought one yet. Initially, it was because I couldn’t really justify the purchase, at least if the amount of time I’ve used my original Nintendo Switch in portable mode is anything to go by. But lately I have a new reason: I’m scared.  See, I could definitely convince myself that getting one makes sense. I work for PCWorld, it’s not as if buying a portable gaming PC would piss off my (imaginary) accountant. But since I’m looking at a $500 purchase, probably closer to $1,000 if I get an OLED model and a couple of upgrades, it’s way outside of impulse buy territory. So I want to make sure it’s a good purchase, if not necessarily a wise one.  The Steam Deck is getting older And here’s the thing: The Steam Deck is three and a half years old. And the OLED model, while gently revising the internal bits, isn’t any more powerful. While the Steam Deck’s AMD Zen 2-based APU is fine for the vast majority of PC games, especially combined with the efficiency of SteamOS and the Proton compatibility layer, it’s definitely looking a bit long in the tooth for the latest 3D titles.  So I have purchase paralysis. I don’t want to sink the better part of a grand into a game machine if a new one is right around the corner. Granted, that’s always an issue with almost any kind of technology, from graphics cards to Grand Cherokees. But even if Valve is tight-lipped about any kind of new model or revision — and more reticent than most companies to release a new one and potentially fracture the userbase in terms of capability — I can’t help but fear that a better alternative is right around the corner. If not from Valve, then from a competitor. Willis Lai/Foundry Since the Steam Deck blew up the previously teeny-tiny portable gaming PC market, dozens of alternatives have popped up mimicking its form factor and capabilities. Most of these are also using AMD’s excellent APU designs, and most are more powerful than the Steam Deck’s, if only nominally so. And now that these can also run SteamOS instead of clunky Windows, either officially or via a secondary download, they’ve become far more appealing.  …but the alternatives aren’t getting better fast enough The Lenovo Legion Go S SteamOS Edition (that will never not be a horrible name) is prominent among these, and offers a direct comparison to illustrate my point. Even using the Ryzen Z2 Go configuration, far less powerful than the older Z1 Extreme, it handily beats the Steam Deck running the same games, resolution, and OS. Absent of all other considerations, if you want a Steam-powered handheld and you’re looking for power, you already need to look beyond the Steam Deck.  There are also some notable weaknesses in the Steam Deck’s design. Don’t get me wrong, it’s still probably the best in the admittedly small field. But it could use a bigger screen for its build, and a more ergonomic design, too — the thumbsticks and touchpads split the difference in terms of being able to reach them. Those sticks could stand to be hall effect (lookin’ at you, Nintendo!), and a little extra work on the internals might give the Steam Deck a bigger battery and room for full-length 80mm SSDs, which are much easier to find and cheaper besides.  So Steam Deck alternatives with more power are now available, even if you want one that isn’t using Windows (and I do). And there are enough improvements to be made on the original model that a revision, or an alternative, should be a serious consideration. Of course, these will require a little extra elbow grease to get SteamOS running on them. And even erring on the optimistic side, I think it’s safe to assume that they won’t get the same excellent software support as the Steam Deck, and won’t have anywhere near as many accessories available. I’m still going to wait it out That leads me to think that, absent a smash-hit alternative in the near future, I should still wait for an upgraded version of the Steam Deck. Unfortunately I don’t think that’s likely to happen anytime soon. At the moment Valve appears to be outselling all the alternatives, even using slightly outdated hardware, and so has no real reason to fast-track an upgrade. The company has said as much. And with tariffs, the AI chip capacity crunch, and other uncertainties making more expensive hardware unappetizing to Valve and gamers alike, there’s no great rush to get rid of the original (and still fairly affordable) Steam Deck. Willis Lai/Foundry And yet, I can’t bring myself to pull the trigger. With more and more newer games being just beyond the Steam Deck’s efficient capabilities, the problems with the current hardware aren’t going away. And I really want to see what Valve changes in a second-gen device, after learning what improvements are important from years of manufacturing and millions of users. So I’m stuck, yearning for what is and excited (and afraid) of what might be. Once again, that’s a basic dilemma for almost any kind of technology. And I’d be negligent if I didn’t admit a certain amount of first-world privilege in even considering this a conundrum. But I’ll close out with a point in favor of patience for any tech purchase: When the new one arrives, the old one will be even cheaper, giving you more and better options automatically. Maybe I can wait a bit longer after all. Further reading: Master your Steam Deck with these 22 tips and tricks Author: Michael Crider, Staff Writer, PCWorld Michael is a 10-year veteran of technology journalism, covering everything from Apple to ZTE. On PCWorld he’s the resident keyboard nut, always using a new one for a review and building a new mechanical board or expanding

Image: StackCommerce TL;DR 1min.AI gives you access to GPT-4o, Claude 3, Gemini Pro, and more in one platform — Lifetime access is just $79.97 (MSRP $540). Running a business, making content, and scaling a brand takes more than one bot. For writing, editing, designing, researching, and automating, you don’t need a single assistant. You need a whole AI workforce. 1min.AI gives you access to every major AI model in one place: GPT-4o, Claude 3 Opus, Gemini Pro, Llama 3, Mistral, and more. It’s built for serious output across content, image, audio, and video, all from one browser-based dashboard. No clicking from tab to tab. The Advanced Business Plan includes 4 million monthly credits, with up to 450,000 bonus credits just for daily check-ins. Use it to generate blog posts, summarize PDFs, clean up audio, create videos, write social content, or translate documents — and collaborate with up to 20 team members. Unlimited storage, custom brand voices, and prompt libraries come standard. With a 4.7-star rating on Trustpilot and over 100,000 users across industries, it’s the clanker that responds “Roger, roger” to every task. No more juggling subscriptions or overpaying for monthly access to one model at a time. Get lifetime access to 1min.AI for just $79.97 (MSRP $540). 1min.AI Advanced Business Plan Lifetime SubscriptionSee Deal StackSocial prices subject to change. Read More

Image: StackCommerce TL;DR: Amsflow is your all-in-one, AI-enhanced investing research hub—track portfolios, screen global markets, and uncover trends with powerful analytics for just $49.99 for a year. Running a business? Managing your own investments? Or just tired of pretending you understand what’s happening in the markets? Amsflow might just be the super-sharp tool your financial toolkit’s missing. This AI-powered platform crunches more data than your brain can and actually helps you make smart decisions. With Amsflow, you’ll get real-time insights into over 70,000 assets—stocks, crypto, forex, commodities—and tools like “Lisa,” an AI assistant that can answer natural-language queries like “show me small-cap clean energy stocks with strong momentum.” There’s also “X-Ray,” which zeroes in on trends across multiple timeframes, making it easier to spot patterns before they become headlines. Amsflow’s visual tools, heat maps, and alerts make it approachable for solo investors who want real clarity. And if you’re a founder or entrepreneur, it’s an efficient way to track your portfolio, monitor markets that affect your industry, and maybe even find the next smart investment. You’ll also get portfolio management tools, real-time global price tracking, customizable alerts, economic calendars, and live financial dashboards to support you. Pick up one year of Amsflow for just $49.99 (MSRP: $228) for a limited time. Amsflow AI Financial Analysis: Pro PlanSee Deal StackSocial prices subject to change. Read More

Image: Microsoft So many people haven’t been able to upgrade their older Windows computers to Windows 11 because of the latter’s TPM 2.0 hardware requirement. (Learn more about why you need TPM 2.0 for Windows 11, why TPM 2.0 makes PCs better, and why Microsoft is adamant about not letting older PCs without TPM 2.0 run Windows 11.) Some users, however, are saying that they’ve been offered upgrades to Windows 11 on older PCs that don’t meet the TPM 2.0 hardware requirements, reports Neowin. It’s a bit of a head-scratcher since Microsoft hasn’t announced anything official about lowering the system requirements for Windows 11, but this isn’t the first time the company has offered Windows 11 upgrades on incompatible computers that don’t meet the requirements. It’s highly likely that this is a bug or an exception. Some users have reported that their system theoretically had TPM 2.0 but was manually disabled it before the update was offered. Whether Microsoft makes a distinction here, however, is questionable. So far, Microsoft has not deviated from its strict hardware requirements for Windows 11. However, users who want to receive an additional year of security updates on Windows 10 can now sign up for the ESU program. If we hear anything official about relaxed requirements, we will inform you accordingly. Until then, however, you should assume that TPM 2.0 will continue to be a requirement for all Windows 11 PCs. This article originally appeared on our sister publication PC för Alla and was translated and localized from Swedish. Read More

Image: OpenAI You might already be one of the millions around the world using ChatGPT to help out with mundane tasks and automate menial chores. But ChatGPT is capable of much more than just answers and automations. Indeed, ChatGPT can actually be fun! With some creative prompting, you can use ChatGPT for all kinds of out-of-the-box stuff—social experiments, hypothetical scenarios, product inventions, even simulated chats with fictional characters. If you’re new to ChatGPT, here are some interesting and unusual things to try that might just blow your mind. Disclaimer: ChatGPT is just a predictive language generator and doesn’t actually “think” for itself. It doesn’t “know” what it’s saying and is therefore prone to hallucinations, falsehoods, and nonsense. Always approach ChatGPT’s responses with a degree of skepticism! Run experiments on a fictional town Dave Parrack / Foundry One of the most fascinating things I’ve done with ChatGPT is create a virtual town to conduct experiments on—and you can easily try this out yourself! Just ask ChatGPT to create a fictional town and the AI chatbot will do the rest. You can base it on a real town, specify the population count and/or demographics, and provide as much detail as you want. Once your fictional town exists, you can then run experiments on its population and infrastructure—and it’s up to you how malevolent or benevolent you want to be. Either way, ChatGPT will detail the effects of your experimental actions and you can laugh maniacally at what you’ve inflicted on the (thankfully) non-existent population. Simulate potential outcomes of scenarios Dave Parrack / Foundry Have you ever wondered what the most likely outcome would be if a certain scenario came to pass? Well, you don’t have to wonder anymore. You can simply ask ChatGPT to see what would happen—and it works with real-world scenarios or imaginary ones, with scenarios ranging from serious to silly to morbid to outlandish. For example, I’ve asked ChatGPT things like what would happen if North and South Korea were to reunify and what would happen if (in a plot stolen directly from The Avengers) half of the world’s population were to disappear in an instant. Both responses made for fascinating—though slightly scary—reading. Your imagination is the limit here! Explore historical what-ifs Dave Parrack / Foundry In a similar vein to the above, you can ask ChatGPT to explain what would have likely happened if history played out differently. What if dragons were real creatures in medieval Europe? What if the Roman empire had never collapsed? What if Genghis Khan was never born? One obvious example that I just had to explore was what would life be like now if the Axis powers had beaten the Allies in World War II. Turns out, ChatGPT’s interpretation of what the last 80 years might’ve been like makes for grim reading. Thankfully, more light-hearted scenarios are also on the table, such as what would have happened had Will Smith never slapped Chris Rock at the Oscars in 2022. Have some fun with it! Chat with fictional characters Dave Parrack / Foundry Have you ever wanted to have a casual chat with Superman? Or ask Mickey Mouse what it’s like to be such an iconic pop culture character? Well, now you can! Just inform ChatGPT that you want to have a chat with a given fictional character and it will do the rest. You can even specify the type of chat you want and/or what you want to discuss. For example, I had great fun asking Clark Kent about his relationship with Lois Lane and how he copes with leading a double life. I had just as much fun asking Mickey Mouse about his relationship with Walt Disney and what it’s like to have been famous for a century. My colleague Jon Martindale does something similar by using ChatGPT as a kind of virtual counselor, treating it as if it were a person and “talking things out” with the AI chatbot! Play an interactive text adventure Dave Parrack / Foundry I grew up playing text adventures, first in the form of choose-your-own-adventure paperback books where every decision took you to a specific page, and later in the form of computer games. I have a nostalgic itch for these games that I occasionally like to scratch. ChatGPT has made that a lot easier, as it can create a playable text adventure all within the chatbot itself. Simply ask ChatGPT to create one, giving it as much or as little detail as you want. In this example, I didn’t direct ChatGPT at all as far as what the story should be—and I ended up exploring a forest seeking a lost memory. The whole thing was cleverly constructed and left me pondering the nature of identity and how costly it can be to seek knowledge. Generate characters for stories or DnD Dave Parrack / Foundry If you’re ever stumped when creating a character for any kind of medium—whether a short story, a D&D campaign, or freeform roleplaying—then ChatGPT can help. Provide some barebones details about what sort of character you want and what the intended medium is. While ChatGPT is great for one-line prompts, character generation is an area where the more information you provide, the better. You’ll get best results if you define the genre, basic plot, any specific quirks or ideas you might have yourself, plus other factors. After all, you don’t want ChatGPT creating a sci-fi alien raider when you need a love story hunk. When it comes to D&D character generation, you probably have something in mind already, so direct ChatGPT as much as you can. One neat trick here is to provide a fictional character (such as Iron Man) to use as the basis for your new D&D character. Debate two sides of an argument Dave Parrack / Foundry ChatGPT is quite capable at considering both sides of an argument in a logical way, which makes it the perfect tool when you need a third party to mediate an argument—whether