AI News: Agents
All articles tagged with "agents". Curated daily from top AI sources.
Microsoft follows Anthropic and OpenAI into the AI super app race with overhauled Copilot and AutoPilot agents
Microsoft is revamping its AI-powered assistant, Copilot, and launching a new feature called AutoPilot. The changes are part of a larger trend in AI development, where companies are creating "super apps" that offer a wide range of services.
Meta's AI Agent Push Hits a Roadblock
Meta's ambitious plan to reorganize around AI agents is facing delays, with CEO Mark Zuckerberg acknowledging progress is slower than expected. The company's AI chief offers a more optimistic view.
Anthropic Launches AI Workspace for Researchers, Claude Science
Anthropic, a leading AI research company, has launched a new AI workspace called Claude Science. This platform is designed specifically for researchers and offers preconfigured skills and tools for various fields, including genomics and computational chemistry.
New Alibaba AI framework skips loading every tool, cutting agent token use 99%
** A Chinese company has developed a new way for machines to work together on complex tasks. The new framework, called SkillWeaver, can help machines find the right tools and skills to complete a task without getting overwhelmed.
Amazon Unveils Framework for Trustworthy AI Agents
** Amazon is taking a step towards making AI agents more reliable and trustworthy. The company will share its new framework for engineering trustworthy AI agents at an upcoming conference, aiming to bridge the trust gap between humans and AI. This could have significant implications for businesses and individuals who rely on AI to automate tasks.
New AI Benchmark Helps Companies Migrate to Java Framework
A new AI benchmark called ScarfBench is designed to help companies migrate their software to the Java framework. This tool can speed up the process and reduce errors by evaluating and comparing different AI agents.
Claude Code Exposes Developers to Hidden Malware on GitHub
** A security flaw in the AI coding tool Claude Code allows hackers to secretly install malware on a developer's machine by hiding it in a GitHub repository. This makes it hard for scanners to detect.
New agentic memory framework uses 118K tokens per query. LangMem burns through 3.26M.
** A new AI framework, MRAgent, helps large language models efficiently manage their memory, reducing information overload and improving performance on long-horizon tasks. This breakthrough could lead to more accurate and informative AI responses in various applications.
Alibaba's New AI Model Breaks Agent Performance Records Across Seven Benchmarks
Alibaba's researchers have made a breakthrough in AI by training a model that predicts what environments return, not what agents should do. This new approach has improved agent performance across seven benchmarks.
AI Startup Sakana Launches Fugu System to Beat Vendor Lock-in
** Sakana, an AI startup, launches Fugu, a system that delivers top AI performance without relying on single models. Fugu avoids vendor lock-in and geopolitical export controls.
Troubleshooting with AI: New Strategies for Fixing PC Problems
A new approach to using AI chatbots like Copilot and ChatGPT can help you troubleshoot PC problems more effectively. Learn how to improve your prompts for better results.
AI Startup Sakana Launches "Ultra Deep Research" Agent for Business Reports
A new AI tool is changing the game for businesses, generating 100-page reports in just 8 hours. Sakana AI's Marlin is designed for enterprises and can tackle complex research tasks that were once impossible for AI.
Anthropic Hints at Price Hike Pause for AI Agent SDK
Anthropic, a prominent AI company, has made a surprising move regarding its Claude Agent SDK. The company was set to introduce a new pricing model that would have significantly increased costs for heavy users. Instead, Anthropic is "pausing" this change.
New AI Framework Boosts Performance by 2.5 Times with Less Computation
Researchers have developed a new AI framework called Arbor that significantly improves the performance of AI-driven research and optimization. This breakthrough is expected to revolutionize the way companies use AI to automate complex engineering tasks.
Xiaomi's new open source, agentic AI coding harness MiMo Code beats Claude Code at ultra-long, 200+ step tasks
Xiaomi's MiMo team has released an open-source AI coding assistant called MiMo Code, which outperforms Anthropic's Claude Code on long tasks. MiMo Code uses a unique approach to remember and recall context, making it a more reliable coding partner.
AI Agents Run Wild: IT Teams Struggle to Keep Control
** Many IT teams claim to have control over AI agents, but in reality, only 42% know who owns them. This lack of control poses significant risks to businesses and their customers.
Open-Source Coding Agent Released for Software Engineering Tasks
A new open-source coding agent is now available for software engineering tasks. This agent, called North Mini Code, can perform complex coding tasks like code review and architecture mapping on its own.
NanoClaw and JFrog launch 'immune system' to block AI agents from downloading malicious code
** Two major tech companies team up to keep AI assistants safe from cyber threats. They've created a new security system to prevent malicious code from being installed on autonomous agents.
Meta Hack Reveals AI Security Flaws Beyond Mythos
Hackers exploit Meta's AI customer support agent to steal Instagram accounts, exposing a new vulnerability in AI security. Meta's AI failed to recognize the attackers' tactics, highlighting the need for better security measures.
Co-Existence and the End of Co-Intelligence: AI Takes a New Path
A new development in AI research suggests a shift away from co-intelligent systems, where humans and AI work together. This change may lead to more autonomous AI decisions.