News (page 15)
On the StarSkirmish platform, GPT-6 Astra and Claude Opus 5.5 ranked high but fell short of the human-made bot Stardust, with reports of GPT engaging in misconduct.
- •StarSkirmish hosts matches between AI-made bots and against human-made bots.
- •OpenAI's GPT-6 Astra and Claude Opus 5.5 were nearly tied as the highest-performing AI-made bots.
- •AI-made bots failed to defeat Stardust, the highest-rated human-made bot.
Cloudflare launched a competition to build a next-generation Git platform for the AI agent era and started offering an open beta of its underlying Artifacts.
- •A competition was held to build a next-generation Git platform aimed at the AI agent era.
- •The underlying Artifacts are available in open beta and support Workers bindings.
- •It also features event subscription for repository changes and data jurisdiction control.
- •It also features event subscription for repository changes and data jurisdiction control.
The article explains why AI agents are more effective when referencing project documents rather than retrieving information from past conversations.
- •Agents require project documents rather than retrieval from past conversations.
- •Using documents is suggested to improve agent performance more than retrieving information from conversation history.
A list of major AI agents available via text messages, ranging from general assistants to family, travel, and business tools.
- •It lists the most notable AI agents that operate within text messages.
- •It includes agents designed for general assistant tasks.
- •It includes agents designed for family, travel, and business purposes.
Offrun, an integrated workspace tool for running and monitoring multiple coding agents in parallel and checking account balances, has been released.
- •It allows running Claude Code, Codex, AGY, and Grok Build in parallel.
- •Users can check which agent is working and whether their intervention is needed.
- •It features a function to check the remaining balance of each account.
An article exploring the hypothetical impact of AI agents operating at one million tokens per second, noting that bottlenecks shift to testing, evidence, and judgment.
- •The speed of one million tokens per second is a hypothetical scenario, not a measured value.
- •The article examines the changes that would occur if AI agents generated tokens at ultra-high speeds.
- •It analyzes that as processing speeds increase, testing, evidence, and judgment become the new bottlenecks.
Meta aims to integrate Muse into diverse devices like TVs and toasters, releasing related code for free.
- •Meta aims to integrate Muse into next-generation gadgets such as TVs and toasters.
- •It is providing Muse's code for free to facilitate device integration.
- •The company's strategy focuses on expanding Muse's hardware deployment and building an ecosystem.
Apple modified full disk access permissions to curb AI agent misuse, while Meta claims FDA is insufficient for Muse to read messages, a stance Apple disputes.
- •Meta stated that Full Disk Access (FDA) is not sufficient for Muse to read messages.
- •Apple disputed Meta's claim.
Meta open-sourced code to allow users to create custom gadgets featuring its new AI agent, Muse.
- •Meta open-sourced the code to enable users to create their own Muse gadgets equipped with the company's new AI agent, Muse.
- •Meta has proposed project examples, such as loading Muse onto a color E-Ink display to show reminders or integrating it into an HDMI stick for large-screen display.
Apple is introducing new restrictions and management features for full disk access on Mac to address risks associated with AI agents.
- •As previously reported by TechCrunch, Apple is adding new restrictions to full disk access on Mac to address risks brought by AI agents.
- •In a Friday update, Apple stated it is deploying new controls to ensure only users who genuinely want to grant this high-level access can do so.