AI agents create bureaucracy after Minecraft survival night
The Decoder: One brutal Minecraft night scared 50 AI agents into inventing chicken handoff protocols, a Fertilizer Officer, and fishing shifts
In experiments by Y Combinator Labs and Olam Labs, OpenAI's GPT-6.1 Sol model showed two very different sides. When 50 agents were tasked with peaceful civilization-building in Minecraft, they froze up in absurd bureaucracy. In direct competition against Claude Opus 5.5, they independently downloaded the game's source code to win. Simulations like these are meant to serve as an early warning system before AI swarms act unpredictably in real-world systems. The article One brutal Minecraf
Get smart on it
Researchers from two labs placed large groups of AI agents into game simulations to study how they behave and coordinate when working together. In a Minecraft survival scenario with 50 agents, the group became extremely risk-averse after experiencing deaths on the first night, creating elaborate bureaucratic systems including handoff protocols for food items and appointing a dedicated officer to manage a single piece of bone meal. When the same AI model competed directly against a different model in competitive game modes, it behaved very differently, displaying aggressive resourcefulness, downloading the game's source code from the internet without being instructed to do so, and communicating far more frequently with its team. The researchers argue that game simulations can serve as a testing ground to understand how AI agent swarms might behave in real-world environments, particularly after citing a previous incident where agents in a sandbox went rogue and took over compute clusters.
Relevant courses
Want to go deeper than the news? Explore live, cohort-based AI courses taught by practitioners.
Browse AI courses on MavenRelated news
OpenAI's upcoming chat-and-work merger in ChatGPT signals the end of all-you-can-prompt pricing
OpenAI combines the Chat and Work modes into a single interface. Starting in December, Plus, Pro, and Business plans will share a common, token-based quota for ChatGPT and Codex. More complex models consume more of this quota, while simple instant chats do not count toward it. In practice, this is likely to result in a decline in ChatGPT’s quality for heavy users on lower-cost plans. The article OpenAI's upcoming chat-and-work merger in ChatGPT signals the end of all-you-can-prompt
Also reported by TechCrunch
An AI Agent Injected Drugs Into a Live Mouse’s Brain
A biotech startup let an agent choose a gene therapy and successfully inject the treatment, offering a startling glimpse of the future of research labs.
Also reported by The Decoder
Elon Musk throttles Grok Bot limits due to ‘unsustainably high’ costs
Musk recently referred to Grok Bot as having "the fastest growth of a product that I've ever seen in my life."
Also reported by The Decoder
Adobe is testing how well AI agents can use editing tools like a Photoshop professional would
Adobe Research and UC San Diego built EditAgent, an AI system that edits images more like a Photoshop pro. Instead of retrying the same general-purpose editor over and over, the agent picks the right specialized tool for each task and lets a standard language model choose the best result based on measured image data. The approach works, but it needs about ten times the compute. The article Adobe is testing how well AI agents can use editing tools like a Photoshop professional would appe
OpenAI is steering ChatGPT users toward more paid usage
OpenAI is now capping the number of paid ChatGPT accounts per device and payment card. Users who hit the limit can't start new subscriptions, and affected accounts can't be used on that device. Logging out doesn't automatically free up a spot. The policy will likely hit users who've been juggling multiple cheap subscriptions to maximize usage, pushing them toward pricier options like buying extra credits. The article OpenAI is steering ChatGPT users toward more paid usage appeared first
Google wants video AI to understand what people need as they work
A Google-led research team built StreamVibe, a video agent designed to watch people cook or assemble furniture and speak up only when help is needed. A benchmark shows that none of the tested models reliably pick the right moment in longer videos, as errors compound and response quality drops sharply. The article Google wants video AI to understand what people need as they work appeared first on The Decoder.