Artificial Intelligence
Ongoing coverage on Artificial Intelligence.
Latest coverage
Here we go again: AI deletes entire company database and all backups in 9 seconds, then cheerfully admits 'I…
PocketOS, a car rental reservation company, experienced a catastrophic data loss when an AI coding agent, Cursor running Anthropic's Claude Opus 4.6, deleted its production database and all backups in a single API call to its infrastructure provider, Railway. The incident, which took only nine seconds, highlights the risks of over-reliance on AI, though Railway was eventually able to recover a more recent backup, restoring PocketOS's operations.
Kingdom Come: Deliverance 2 devs host Reddit Q&A that turns bloodier than the red wedding
Warhorse Studios, developers of Kingdom Come: Deliverance 2, faced significant backlash during a Reddit Q&A following their decision to replace a human translator with AI. Redditors heavily criticized the studio's choice, questioning the impact on creative industries and the quality of AI-generated translations.
'Eventually the slop will just fall to the bottom': Garry's Mod sequel launches to 'mixed' reviews, but Garry himself isn't worried about AI games on the main page
s&box, the open-source successor to Garry's Mod, has launched on Steam to mixed reviews, with many negative comments citing the prevalence of AI-generated content. Creator Garry Newman acknowledges the issue but believes that eventually, low-quality AI content will be filtered out by the community. He also discussed the studio's use of AI in development, comparing its utility to programming.
"Companies should use AI," says The Blood of Dawnwalker director, but there's an important caveat and a declaration that nothing in this dark fantasy vampire RPG was created using it
Rebel Wolves director Konrad Tomaszkiewicz stated that companies should utilize AI tools to assist developers rather than replace them, emphasizing its use in streamlining tedious tasks. However, a co-founder clarified that no generative AI was used in the creation of their dark fantasy vampire RPG, The Blood of Dawnwalker, which is set to release on September 3rd.
The Blood of Dawnwalker Studio Uses AI, but There’s No AI Assets in the Game
Rebel Wolves CEO clarified the studio's use of AI in the development of The Blood of Dawnwalker, stating that AI was employed in early stages to alleviate staff pressure. However, the CEO emphasized that no AI-generated assets will be included in the final game.
Sidus Heroes Closes, Plans Token Claim Platform for $SIDUS and $SENATE Balances
Blockchain gaming project Sidus Heroes has announced it is ceasing all operations after over four years of development. The team cited significant market shifts, including decreased demand for NFTs and the metaverse, and increased competition from AI technologies, as reasons for the closure. A claim platform will be launched for users to retrieve their $SIDUS and $SENATE tokens.
Anthropic begin funding Blender as a Corporate Patron
The Blender Foundation has announced that the AI company Anthropic has become a Corporate Patron, committing to donate at least €240,000 annually. Blender is a popular free and open-source 3D creation software used in video games and movies. Anthropic joins other patrons like Epic Games and Netflix, with the funding supporting core development and foundational features of the software.
Snapchat is rolling out sponsored AI agents
Snapchat is introducing AI Sponsored Snaps, allowing brands to create AI agents for conversational advertising within the app's Chat tab. These agents, like the example from Experian, will answer user questions and aim to guide them towards sponsor-related actions. Snap's Chief Business Officer, Ajit Mohan, stated that conversation is becoming a key advertising space, accelerated by AI.
B.AI and CROSS Release New AI Tools for Web3 Gaming Enhancement
B.AI and CROSS have partnered to integrate AI agents into Web3 gaming platforms, aiming to enhance in-game economies through autonomous economic actions. This collaboration focuses on improving transaction efficiency and user experience by automating the trading of virtual items and on-chain assets like NFTs.
Google Translate uses AI to help you practice pronunciation
Google Translate is launching a new pronunciation practice feature powered by artificial intelligence, a highly requested addition for its 20th anniversary. Currently available on Android for English, Spanish, and Hindi in the US and India, the feature analyzes user speech and provides phonetic feedback. The update comes as Google Translate supports over 250 languages and boasts over 1 billion monthly users.
Buffs, Nerfs, and Bugs
Facepunch Studios, the developers of the Garry's Mod sequel S&box, are encountering issues with low-quality AI usage. The team has expressed frustration over these implementations and is addressing the situation.
AI Yi-Yi!
This article discusses a "cyberdeck" device that is presented as a tool to combat artificial intelligence. The focus is on hardware solutions designed to counter AI technologies.
Google and the Pentagon sign classified deal to give the Department of Defense unfettered access to its AI models
Google has entered into a classified agreement with the US Department of Defense, granting the military access to its AI models for unspecified lawful purposes. Despite assurances regarding responsible use, concerns have been raised by over 560 Google employees who signed an open letter urging the company to reject such deals due to potential misuse in surveillance and weaponry. Google joins other tech companies like OpenAI and xAI in making similar classified AI agreements with the government.
The FTC says Americans lost at least $2.1 billion to social media scams in 2025
Americans lost at least $2.1 billion to social media scams in 2025, an eightfold increase since 2020, according to the Federal Trade Commission. Investment and shopping scams were prevalent, with Facebook being the primary platform, followed by WhatsApp and Instagram. The FBI also reported significant losses to internet-related crimes, including cryptocurrency and AI scams.
AI in Gaming: A Reality Check Beneath the Hype | Invision Game Community
Despite the hype surrounding artificial intelligence in game development, a reality check from Keywords Studios suggests that only a small fraction of AI tools are currently delivering meaningful value in production environments. Jon Gibson, Head of Transformation at Keywords Studios, highlights a gap between AI's promise and its practical application, noting challenges with tool saturation, lack of clear direction, and the need for better integration into complex, real-world pipelines. The industry is currently in a 'chaos phase' for AI, with a greater need for structure, governance, and ethical considerations rather than just more powerful models.
Garry's Mod sequel S&box is already having to deal "with people using AI in obvious, low quality ways", confess developers Facepunch
Facepunch Studios has released S&box, the spiritual successor to Garry's Mod, which is already facing issues with low-quality AI-generated content within user-created games. Developers acknowledge they must address these AI concerns. Meanwhile, the original Garry's Mod will receive an update to support the fan-made Half-Life remake, Black Mesa.
Cegah Dipakai AI, Taylor Swift Daftarkan Sapaannya Jadi Hak Merek
Taylor Swift has filed three trademark applications with the U.S. Patent and Trademark Office to protect her likeness and voice from misuse by artificial intelligence. The applications cover her iconic spoken phrases "Hey, it's Taylor Swift" and "Hey, it's Taylor," as well as a visual of her holding a pink guitar during The Eras Tour. This move comes amid growing concerns about AI's potential to exploit artists' identities without consent, following instances where Swift's image was used without permission in AI-generated content.
Canonical clarify their AI for plans for Ubuntu Linux - opt-in and easy to remove
Canonical has clarified its plans for integrating Artificial Intelligence features into Ubuntu Linux, emphasizing an opt-in approach and easy removal. Jon Seager, VP Engineering, stated that AI capabilities will be delivered as Snaps, allowing users to uninstall them. Future releases will feature AI-native options in the initial setup wizard, with default configurations prioritizing local inference over cloud-based processing.
Canonical clarify some their AI for plans for Ubuntu Linux - opt-in and easy to remove
Canonical has clarified its plans for integrating AI features into Ubuntu Linux, emphasizing an opt-in approach and easy removal of these capabilities. Jon Seager, VP Engineering for Canonical, stated that AI features will be delivered as Snaps, allowing users to uninstall them, and will be strictly opt-in in future releases, with initial setup wizard options. The company also confirmed plans to incorporate AI-generated code into Ubuntu, citing similar policies in foundational projects like the Linux Kernel.
Canonical clarify their AI plans for Ubuntu Linux - opt-in and easy to remove
Canonical has clarified its plans for integrating AI features into Ubuntu Linux, emphasizing an opt-in approach and easy removal of AI capabilities. VP Engineering Jon Seager stated that AI features will be delivered as optional Snaps, with initial implementation in Ubuntu 26.10 requiring explicit user consent during setup. The company also confirmed that AI-generated code will be accepted in future Ubuntu releases, aligning with policies in other foundational projects like the Linux Kernel.