Happy Halloween! This year, we’re celebrating some of your most terrifying creations. Mark your calendars for Halloween, as we’ll be sitting down with two incredible horror devs – Chantal Ryan (We Have Always Lived in the Forest) and Sam Barlow (Half Mermaid). Together, we’ll play games created by the Unity community and break down the game design strategies and horror narrative techniques that make them so effective. All in costume, of course.Catch us live on Halloween day at 12 pm ET / 9 am PT.Weird Horror – Community Choice Steam Curator List: October 2024Horror takes many forms, and this month, you helped us choose one – weird horror came out on top. Check out our Steam Curator list of games with unconventional horror design.Learn to create your own horror roguelikeFeeling inspired enough to make your own horror game? Check out the new 2D Roguelike course on Unity Learn. It's free and chock full of zombies and skeletons to get you started.Working on a game in Unity? We’d love to help you spread the word. Be sure to submit your project.We asked folks here at Unity for some of their favorite (or most anticipated) Made with Unity horror games. Check them out below, or add to the list by sharing your favorites.Projekt Z: Beyond Order, 314 ArtsSorry We’re Closed, à la mode gamesPhasmophobia, Kinetic GamesSons of the Forest, Endnight Games LtdWho’s Lila?, Garage HeathenCrow Country, SFB Games IMMORTALITY, Half MermaidFear the Spotlight, Cozy Game Pals Mouth Washing, Wrong OrganHollowbody, Headware GamesDoki Doki Literature Club Plus!, Team SalvatoMy Friendly Neighborhood, John Szymanski, Evan SzymanskiContent Warning, thePetHen, Skog, Zorro, Wilnyl, PhilipSIGNALIS, rose-engineDREDGE, Black Salt GamesInscryption, Daniel Mullins GamesTormenture, Croxel StudiosSUNLESS SEA, Failbetter GamesCROWDED. FOLLOWED., NIGHT DIALLethal Company, ZeekerssCult of the Lamb, Massive MonsterCultist Simulator, Weather FactoryMADiSON, BLOODIOUS GAMESBroken Spectre, Games by StitchLamentum, Obscure TalesThat’s a wrap for Halloween 2024! Want more Made with Unity and community news as it happens? Don’t forget to follow us on social media: X, Facebook, LinkedIn, Instagram, YouTube, or Twitch.

Anthropic CEO Dario Amodei published an essay Thursday highlighting how little researchers understand about the inner workings of the world’s leading AI models. To address that, Amodei set an ambitious goal for Anthropic to reliably detect most AI model problems by 2027. Amodei acknowledges the challenge ahead. In “The Urgency of Interpretability,” the CEO says Anthropic has made early breakthroughs in tracing how models arrive at their answers — but emphasizes that far more research is needed to decode these systems as they grow more powerful. “I am very concerned about deploying such systems without a better handle on interpretability,” Amodei wrote in the essay. “These systems will be absolutely central to the economy, technology, and national security, and will be capable of so much autonomy that I consider it basically unacceptable for humanity to be totally ignorant of how they work.” Anthropic is one of the pioneering companies in mechanistic interpretability, a field that aims to open the black box of AI models and understand why they make the decisions they do. Despite the rapid performance improvements of the tech industry’s AI models, we still have relatively little idea how these systems arrive at decisions. For example, OpenAI recently launched new reasoning AI models, o3 and o4-mini, that perform better on some tasks, but also hallucinate more than its other models. The company doesn’t know why it’s happening. “When a generative AI system does something, like summarize a financial document, we have no idea, at a specific or precise level, why it makes the choices it does — why it chooses certain words over others, or why it occasionally makes a mistake despite usually being accurate,” Amodei wrote in the essay. In the essay, Amodei notes that Anthropic co-founder Chris Olah says that AI models are “grown more than they are built.” In other words, AI researchers have found ways to improve AI model intelligence, but they don’t quite know why. In the essay, Amodei says it could be dangerous to reach AGI — or as he calls it, “a country of geniuses in a data center” — without understanding how these models work. In a previous essay, Amodei claimed the tech industry could reach such a milestone by 2026 or 2027, but believes we’re much further out from fully understanding these AI models. In the long term, Amodei says Anthropic would like to, essentially, conduct “brain scans” or “MRIs” of state-of-the-art AI models. These checkups would help identify a wide range of issues in AI models, including their tendencies to lie or seek power, or other weakness, he says. This could take five to 10 years to achieve, but these measures will be necessary to test and deploy Anthropic’s future AI models, he added. Anthropic has made a few research breakthroughs that have allowed it to better understand how its AI models work. For example, the company recently found ways to trace an AI model’s thinking pathways through, what the company call, circuits. Anthropic identified one circuit that helps AI models understand which U.S. cities are located in which U.S. states. The company has only found a few of these circuits but estimates there are millions within AI models. Anthropic has been investing in interpretability research itself and recently made its first investment in a startup working on interpretability. While interpretability is largely seen as a field of safety research today, Amodei notes that, eventually, explaining how AI models arrive at their answers could present a commercial advantage. In the essay, Amodei called on OpenAI and Google DeepMind to increase their research efforts in the field. Beyond the friendly nudge, Anthropic’s CEO asked for governments to impose “light-touch” regulations to encourage interpretability research, such as requirements for companies to disclose their safety and security practices. In the essay, Amodei also says the U.S. should put export controls on chips to China, in order to limit the likelihood of an out-of-control, global AI race. Anthropic has always stood out from OpenAI and Google for its focus on safety. While other tech companies pushed back on California’s controversial AI safety bill, SB 1047, Anthropic issued modest support and recommendations for the bill, which would have set safety reporting standards for frontier AI model developers. In this case, Anthropic seems to be pushing for an industry-wide effort to better understand AI models, not just increasing their capabilities.

Twitch has revealed more information about its streaming trends of 2024, and GTA V is its most popular game, while VTubers are on the rise.Read More

Lip-Bu Tan, the new CEO of Intel, sent out a blunt message to employees saying the company has to reorganize to be more efficient.Read More

Introduction AWS is a popular cloud provider that enables the deployment and scaling of large applications. Mastering at least one cloud platform is an essential skill for software engineers and data scientists. Running an application locally is not enough to make it usable in production — it must be deployed on a server to become accessible to end users. In this tutorial, we will walk through an example of deploying a FastAPI application. While the example focuses on core EC2 networking concepts, the principles are broadly applicable to other types of applications as well. Please note that this tutorial does not cover best practices for AWS usage. Instead, the goal is to give readers a hands-on introduction to application deployment using EC2 instances. # 01. Instance creation Navigate to the Ec2 dashboard in the AWS service menu and choose to create a new instance. This will open a page where we can define instance parameters. Select the corresponding instance type. In this tutorial, we will launch a very simple server with minimal technical requirements, so t3.nano should be sufficient for our needs. For its containers, AWS uses SSH authentication. When creating a new instance, it is necessary to create a new key pair that will allow us to log in from the local machine using the SSH protocol. Click on Create new key pair. Assign a name to the new key. We will not dive into the possible options here, so we will choose RSA as the key pair type and .pem as the private key file format. To save time, in our demonstration application we will not worry about security. For the network settings, tick all the checkboxes corresponding to SSH, HTTP, and HTTPS traffic. Great! By clicking Launch instance, AWS will create a new instance. After the instance is created, a .pem file will be downloaded to your local machine. This file contains the private key that allows SSH authentication. As a good practice, store this file in a safe location because AWS does not provide a way to recover it if it is lost. By opening the EC2 dashboard, you will notice that the created instance has an associated IP address. This IP is shown under the label “Public IPv4 address”. For example, in the image below, it is “16.16.202.153”. Once we deploy our application, it will be accessible from a browser using this IP address. # 02. SSH connection AWS offers several ways to perform authentication. In our case, we will use the SSH mechanism. In the instance menu, click Connect and select SSH client from the top bar. Open the local terminal and, using the screenshot above as reference, copy and execute command #3 (chmod 400 "<key_name>.pem") along with the command shown below the “Example” label. Make sure your current terminal directory matches the location where the .pem key was downloaded in the previous step. During the SSH connection, the terminal might prompt whether to proceed. If it does, type “yes”. At this point, we are successfully connected from the local terminal to the EC2 instance. Any commands entered into the terminal will now be executed directly in the EC2 container. # 03. Environment configuration After connecting to the instance from the local terminal, the next step is to update the package manager and install Python along with Nginx. sudo apt-get update sudo apt install -y python3-pip nginx To redirect traffic to our application, we need to create an Nginx configuration file. This file should be placed in the directory /etc/nginx/sites-enabled/ and can have any custom name. We will add the following configuration to it: server { listen 80; server_name <Public_IP_Address>; location / { proxy_pass http://127.0.0.1:8000; } } Basically, we are specifying that any external request sent to the EC2 instance’s IP address on the default port 80 should be redirected via a proxy to the application running inside the EC2 container at the address http://127.0.0.1:8000. As a reminder, this is the default HTTP address and port assigned by FastAPI. To apply these changes, we need to restart Nginx: sudo service nginx restart If we have a FastAPI server that we would like to launch, the simplest way would be to publish it on GitHub and then clone the repository onto the EC2 instance. git clone <GitHub_URL> <directory> cd <directory> Create and activate a virtual environment: python3 -m venv venv source venv/bin/activate Install the necessary Python requirements (assuming that the cloned repository contains a requirements.txt file): pip3 install -r requirements.txt Run the server: python3 -m uvicorn <server_filename>:app Open the browser and enter the IP address of the instance. Make sure to use the HTTP (not HTTPS) protocol. For example: <a href="http://16.16.202.153/" rel="noreferrer noopener" target="_blank">http://16.16.202.153</a>. The firewall might block your connection, but you should proceed to open the web page. Add /docs after the URL to open Fast API Swagger. Exercise If you would like to run a FastAPI example, you can create a simple repository consisting of just a main.py file and a requirements.txt. main.py from fastapi import FastAPI app = FastAPI() @app.get("/") def read_root(): return {"message": "Hello, World!"} requirements.txt fastapi uvicorn Uploading files If you try to upload a file to a server and receive a 413 status with the error message “Error: Request Entity Too Large”, it is likely because Nginx has a limit on the maximum file size that can be uploaded. To resolve this issue, go to the Nginx configuration file and specify the maximum allowed file size by using the client_max_body_size directive (setting it to 0 indicates no limits on input file sizes): server { listen 80; server_name <PUBLIC_IP_ADDRESS>; location / { proxy_pass http://127.0.0.1:8000; client_max_body_size 0; } } After changing the configuration file, do not forget to restart Nginx. Conclusion In this article, we have learned how to quickly create a running EC2 instance using a FastAPI server as an example. Although we did not follow the best deployment and security practices, the main goal of the article was to provide minimal information for beginners to launch their first server on AWS. The next logical step in the AWS study roadmap would be creating multiple EC2 instances and connecting them to each other. All images unless otherwise noted are by the author. Connect with me Medium LinkedIn The post AWS: Deploying a FastAPI App on EC2 in Minutes appeared first on Towards Data Science.

Les camions sans conducteur sur route, c'est pour très bientôt. Aurora Innovation, entreprise américaine spécialisée dans le développement de...

Vagabond Definitive Edition Vol. 1 (Hardcover) $29.79 (was $55) See at Amazon One of the most notable premium manga releases of the past year is on sale for nearly 50% off at Amazon. Vagabond Definitive Edition Vol. 1, a gorgeous 728-page hardcover published by Viz Media in January, is up for grabs for just $29.79 (was $55). This massive display-worthy book collects the first three volumes in Takehiko Inoue's legendary series that's widely regarded as one of the best manga within its genre. Vagabond Definitive Edition Vol. 1 (Hardcover) $29.79 (was $55) This new edition of Vagabond is presented in oversized hardcover format, which should make Inoue's beautiful and brutal visuals really shine. So far, 37 individual volumes of Vagabond--collected in 12 books--were published before Inoue took a break from the series to focus on his health and other projects. The pursuit of perfection and weekly publishing schedule took its toll on Inoue, and a year after the hiatus, he returned to his basketball roots with the series Real, an inspiring sports story focused on wheelchair basketball.If you're new to Vagabond and don't want to drop $43 to check it out, most of the previously released paperbacks, which contain three volumes each, can be purchased from Amazon for around $20 each.There's no indication that Inoue will return to Vagabond in the future, but the current volumes are worth reading for their intricate plotting and mesmerizing visuals. See at Amazon If you've never read it, Vagabond follows the tale of Shinmen Takezou through 16th-century Japan. Also known as the legendary swordsman Miyamoto Musashi, the manga is a fictionalized account of his life and is based on Eiji Yoshikawa's novel Musashi. A decades-spanning story, Vagabond focuses not only on the battles that made Musashi a legend but also on the more intimate moments of his life as he strives for enlightenment, through stunning art.Continue Reading at GameSpot

Schedule 1 developer TVGS teased potential map expansions coming to the game. Released in early access in March 2025, Schedule 1has gained significant popularity, especially among content creators, receiving "overwhelmingly positive" reviews on Steam for its fun approach to manufacturing and distributing illegal substances.

What is an “Xbox game”? Fairly recently, the answer was simple: An Xbox game is a game that you play on your Xbox. But now, with Microsoft pushing a multiplatform strategy — and openly challenging preconceived notions of what even counts as a platform — determining the best Xbox games isn’t as straightforward as it was. For us, though, the answer is still pretty simple, if somewhat evolved. Some games (Halo, Forza) will always be Xbox games regardless of how many platforms they land on, the same way a mobile Mario game is still a Nintendo game. But others fall under the broader category of, if an Xbox is the primary way you play games, you shouldn’t miss these — regardless of who published or developed them. What follows is our list of the best Xbox games as of April 2025. How we pick the best games on Xbox The Polygon staff plays a lot of video games, and everything in this list comes personally recommended by at least one of us. We determined what should be on our list of the best Xbox games by looking at the quality of each title, but also with an eye for breadth and variety — so you should find something on the list you’ll enjoy, no matter what genres of game you like, how much time you have, or what vibe you are after.

Back in December, Sony rolled out themes for the first time on PS5. While the selection was nowhere near the variety offered by other consoles, they were enticing all the same: These themes were modeled after the UI designs of all previous PlayStation consoles, including the PSOne, PS2, PS3, and PS4. These themes are, frankly, really cool. They adopt each console's unique sound design, like the clicks you hear when switching between menu items. If you had a PS3 back in the day, you'd probably like the wave background this PS5 theme adds here. Similarly, PS2 fans would appreciate the second generation's menu shapes. If you had a particular favorite, you could inject a little nostalgia into your PS5 experience—at least while on the home screen.However, the experience was short-lived. At the end of January, the company decided to remove the themes, to the disappointment of many 90s and 00s kids. Luckily, there was a silver lining to the situation: Sony said the themes would return in an unspecified number of months, following some work "behind the scenes." As it turns out, that number of months was almost exactly three: Starting Thursday, April 24, classic PlayStation themes are back on PS5, as Sony announced in a blog post on Wednesday. The company says the feature is now called "Appearance" and contains four nostalgic themes—now without the 30th anniversary branding. How to access the classic PlayStation themes Credit: Sony These themes are rolling out as part of Sony's latest system update for PS5. In order to access them, you'll need to make sure your console is fully updated. To do so, go to Settings > System > System Software > System Software Update and Settings. If you aren't fully updated, you'll see Update Available. Choose Update System Software, and you're set.Once updated, head back to Settings, then choose "Appearance and Sound." Default is the default PS5 theme, but you'll be able to choose from PlayStation, PlayStation 2, PlayStation 3, or PlayStation 4.There's another new feature in the latest PS5 update Credit: Sony In addition to these new themes, Sony is also rolling out "Audio Focus," a new setting that lets you boost soft sounds in your games when using headphones with your PS5. There are four different categories you can adjust: Low Pitch (engines and rumbles); Voices (voice chats, in-game dialogue, mid-frequency sounds; High Pitch (footsteps and metallic sounds); and Quiet Sounds (low-volume sounds across frequencies).Once you've updated to the latest version, you'll find these options in Settings > Sounds > Volume > Audio Focus. For each, you can choose whether to adjust the sounds for the left and right channel, as well as adjust the overall level of the boost. When you're finished, you can test your settings via an "Audio Preview" option in this settings menu, so you know whether you need to make adjustments.