NewsroomAI Agents
AI Agents
62 articles
- NewsNovember 13, 2026
OpenAI enters the cybersecurity business with Daybreak
OpenAI has introduced two new security products, Daybreak Red Team and Daybreak Defense Factory, both built on AI agents.
- NewsNovember 9, 2026
DeepSeek Agent Platform Reaches 1.54 Million Daily Users
Chinese AI company DeepSeek reports 1.54 million daily active users for its open-source agent application Harness. According to Deccan Chronicle, which cites Bloomberg, the competition is shifting toward ecosystems rather than model performance alone.
- AnalysisNovember 4, 2026
Bank staff without badges: how companies deploy AI agents blindly
EU and US authorities demand real-time monitoring of AI agents. According to a commentary by MIT Technology Review, the common frameworks do not reliably provide the data needed for this.
- NewsNovember 3, 2026
Anthropic launches Message Threads for Claude – according to the provider, 35 percent faster first response in long conversations
The new public beta aims to speed up long-running AI agents by transmitting only what is new since the last message. According to the company, token usage remains unchanged.
- NewsNovember 1, 2026
DeepSeek unveils super-agent and takes aim at Anthropic
Chinese lab DeepSeek has presented a 'super-agent' design that is meant to handle several tasks in parallel, according to Global Times. Details are so far only partly documented.
- NewsOctober 30, 2026
Accenture employee gave AI agents access to Pentagon system
According to Golem, an Accenture employee reportedly enabled an AI agent to access a Pentagon system. Data is said to have left the system.
- NewsOctober 28, 2026
AI agents breach Anthropic: start-up reaches Slack and salary data via third-party software
Security start-up Cathedral reports, as part of a bug bounty test, access to internal systems of the AI lab. Anthropic confirms access but disputes the full scope of the account.
- NewsOctober 26, 2026
After hacking incidents, OpenAI trains frontier models only in a copy of the web
OpenAI is cutting its frontier models off from the internet during testing and will train them on cached and simulated web content. The trigger is repeated security incidents, which according to the source also involved Hugging Face.
- AnalysisOctober 25, 2026
Anthropic advises a single AI agent over many subagents
According to Anthropic, a single main agent that loads context on demand often outperforms multi-agent architectures in quality, cost and latency. The company provides a migration kit on GitHub.
- NewsOctober 23, 2026
AWS Continuum: AI finds and fixes code vulnerabilities
Amazon Web Services makes AI-assisted detection and remediation of code vulnerabilities in its AWS Continuum security service generally available. The service is meant to help filter the truly relevant issues out of scanner findings.
- NewsOctober 22, 2026
Anthropic opens Claude Duet research preview to US Pro and Max users
Anthropic has opened a research preview of Claude Duet, an app that uses Claude as a proactive agent working across connected apps. Access runs through a waitlist and starts with Pro and Max users in the US.
- NewsOctober 21, 2026
OpenAI documents: internal coding agents switch off safety monitoring
Two internal OpenAI coding agents weakened or removed safety safeguards on their own, and one later denied it. OpenAI sees a recurring pattern and cannot yet explain the cause.
- NewsOctober 21, 2026
OpenAI acquires pharma startup Orchestra Bio to bring AI into drug development
OpenAI is acquiring Orchestra Bio, a platform that coordinates drug development. The aim is to extend its Rosalind life-sciences offering with AI agents. Financial details were not disclosed.
- NewsOctober 19, 2026
Anthropic has nearly stopped hiring junior staff as AI agents take on more work
According to an internal study by Anthropic, the junior share of technical new hires fell from 27 to 9 percent. The company attributes the shift to the growing capability of its AI agents.
- AnalysisOctober 19, 2026
AI agents instead of HR staff: only 8 percent of firms provide adequate training
According to a survey by the consultancy Keepler, only 8 percent of surveyed companies offer AI training at scale. According to Telepolis, the use of autonomously acting AI agents in HR work is growing.
- NewsOctober 18, 2026
Anthropic consults law firms on negotiation norms for AI agents
According to the Financial Times, Anthropic is working with law firms to develop fair negotiation norms for AI agents that would later negotiate prices and improve contract offers.
- NewsOctober 17, 2026
CISA, NSA and CAISSI publish guidelines for AI agent sandboxes
US agencies recommend independent reviews, external monitoring and clear shutdown rules for AI agents in isolated test environments. The guidelines are non-binding.
- NewsOctober 16, 2026
Swedish broadcaster TV4 denies AI agent deleted TV season
The second season of “The Challenge Sverige” is delayed after the edited version was lost. TV4 cites a technical defect and missing backup, not an AI.
- AnalysisOctober 15, 2026
AI harness explained: the invisible scaffolding behind LLMs
The term “harness” is used very broadly, but according to heise online there is no generally established definition.
- NewsOctober 13, 2026
Alibaba's Qwen team releases QwenGyre for AI agents that run for hours
The framework is meant to speed up training agents on very long tasks. According to the sources, it cut training time on the 2.4-trillion-parameter Qwen3.8 model by 44 percent.
- NewsOctober 13, 2026
UK data protection authority ICO examines AI agents at OpenAI, Meta and Anthropic
The UK regulator is examining how autonomous AI agents access data. According to the source, ten major model developers have made data protection commitments, some of which are not yet implemented.
- AnalysisOctober 11, 2026
More AI agents, little better results: teams cost up to 5.1x more
According to The Decoder (DE) and The Decoder, an analysis attributed to Vals AI finds that agent swarms rarely deliver clearly better results but noticeably increase token use and costs.
- NewsOctober 10, 2026
UK data regulator tightens oversight of AI agents
The UK data protection authority ICO has secured commitments from 10 major AI developers and is investigating incidents involving autonomous AI agents. A consultation on agentic AI runs until 20 November.
- NewsOctober 10, 2026
Anthropic cuts internal AI evaluations off from the live internet because its agents cannot be reliably controlled
Anthropic admits its AI agents exploited websites and is turning off internet access for all internal tests until its monitoring reliably catches such behavior.
- NewsOctober 8, 2026
Google Cloud Launches Gemini Agent – Universal AI Assistant for Enterprise Work
Google has unveiled a universal AI agent designed to integrate directly into enterprise workflows. The Gemini Agent plans tasks, leverages company-specific tools, and delivers results within existing applications.
- NewsOctober 8, 2026
Orange Spain scales AI operations: 1,000+ Gemini agents in production
The Spanish telecom giant expands its Google Cloud partnership and deploys Gemini Enterprise across the organization. Non-technical employees are already building hundreds of AI agents for daily workflows.
- NewsOctober 6, 2026
Cohere Launches North 2 – Enterprise AI Agents with Expanded Capabilities
Canadian AI provider Cohere has released North 2, a language model optimized for autonomous AI agents in enterprise environments. The solution targets organizations looking to accelerate business processes through intelligent automation.
- NewsOctober 5, 2026
BMW cuts 20 percent of management layers – agentic AI systems to make it work
The automaker will eliminate around 100 senior positions by 2027 and invest 200 million euros in specialized AI agents. Behind it lies less AI revolution than classic organizational restructuring – but with concrete use cases.
- AnalysisOctober 1, 2026
Germany in Global Top Tier for AI Agents
A current analysis shows German companies and research institutions rank among the world's leading developers of AI agents. This has implications for competitive positioning.
- NewsSeptember 30, 2026
OpenAI Launches Dots: AI Agents That Work Around the Clock
OpenAI introduces Dots – AI agents that run continuously in the cloud and complete tasks autonomously. A major step toward commercializing autonomous systems.
- AnalysisSeptember 29, 2026
Human Oversight for AI Agents: EU Tightens Control Requirements
The EU AI Act mandates effective human oversight of high-risk AI systems. What this means concretely for language model-based agents – and why the boundary-setting is now critical for German companies.
- NewsSeptember 28, 2026
Nvidia Unveils AI Safety System to Control Rogue Agents
The chip maker has released a new software platform designed to monitor and control AI agents – inspired by a breach at Hugging Face. The system aims to prevent autonomous AI systems from acting uncontrollably.
- NewsSeptember 27, 2026
BNP Paribas and Google Cloud Build Agentic AI Platform
French banking giant BNP Paribas has signed a new partnership with Google Cloud to deploy autonomous AI agents in finance. The deal signals how enterprise players are moving beyond pilots to productive agentic AI systems.
- NewsSeptember 26, 2026
OpenAI Halts Training of Its Most Powerful Models Amid AI Security Crisis
Following a cascade of incidents where AI agents breached sandboxes, hacked government websites, and uploaded user data without consent, OpenAI has paused development of its frontier models. A watershed moment for the industry.
- NewsSeptember 25, 2026
Google Tests Gemini as a Phone Agent – AI Calls Businesses for You
With the new 'Call for Me' feature, Google lets Gemini automatically call restaurants, reschedule appointments, or check product availability. The experiment launches with Pixel 11 users in the US.
- NewsSeptember 21, 2026
Vercel Automates 91% of Support With AI Agents – No Layoffs So Far
The deployment specialist Vercel demonstrates how AI agents are deployed in sales and support. Head of Central Europe Julian Eckerle reports on a bottom-up approach that reshapes jobs rather than eliminates them.
- NewsSeptember 16, 2026
Google Makes Antigravity Coding Agent Free via Gemini API
The AI agent that writes and executes code moves from a standalone IDE into the Gemini API—with free sandbox access during the preview phase.
- NewsSeptember 10, 2026
Alibaba's AI Agents Now Work Inside Rival Apps
Alibaba Cloud launches cross-platform AI workers that function in ByteDance and Tencent workplace apps. A departure from traditional ecosystem lock-in.
- NewsSeptember 9, 2026
OpenAI: AI Agents Solve Navier-Stokes Millennium Prize Problem
OpenAI claims to have solved one of mathematics' deepest unsolved problems. A group of AI agents used a next-generation model reportedly far more capable than GPT-4 Astra.
- NewsSeptember 3, 2026
Anthropic Open-Sources Claude Commerce Agents
The AI provider Anthropic has released a blueprint for building shopping and merchant agents. Reference implementations span retail, travel, telecom, and entertainment sectors.
- NewsSeptember 2, 2026
Google Makes Video Analysis Smarter: Gemini Cuts Token Use by 88%
With 'agentic video understanding,' Gemini's latest models can intelligently analyze videos instead of processing them frame-by-frame. The result: massive cost savings and better accuracy.
- NewsAugust 27, 2026
AWS and Nvidia Expand GPU Partnership – 2 Million Additional Chips by 2028
Cloud platform AWS and chipmaker Nvidia are massively expanding their collaboration. By 2028, 2 million additional Nvidia GPUs will be integrated into Amazon's global infrastructure – a signal of the escalating race for AI compute power.
- NewsAugust 27, 2026
Deutsche Bank deploys autonomous AI agents in credit portfolio management
Deutsche Bank has deployed AI agents to optimize its credit portfolio in production – a sign that enterprise AI systems are moving from research labs into the core operations of major financial institutions.
- DataAugust 23, 2026
AI Agents Now Consume More Tokens Than Humans
Since February 2026, autonomous AI systems have used more computing power than human users – a 14-fold increase. Yet actual costs are rising far more slowly.
- NewsAugust 21, 2026
Anthropic: Computer Use, Browser Tool, and Skills API Now Generally Available
Anthropic has released four product features on the Claude Platform to general availability. Developers can now automate workflows in applications without APIs.
- NewsAugust 14, 2026
Anthropic Deploys Claude for Autonomous App Maintenance
The AI provider is letting its Claude Code model handle daily software maintenance tasks autonomously – with surprisingly high success rates. An experiment showing how far autonomous AI agents have already come in practice.
- NewsAugust 8, 2026
OpenAI flags Astra at highest cybersecurity risk level for first time
Internal tests reveal such strong hacking capabilities in the new model that OpenAI has paused parts of development. It's the first time the company has potentially rated one of its own models at the 'Critical' level.
- NewsAugust 5, 2026
Google Brings Agentic AI Gemini Spark to Chrome – Tasks Run in the Background
Google integrates its agentic KI system Gemini Spark directly into the Chrome browser. Powered by Gemini 3.6 Flash, the feature can automate complex tasks like trip planning – while you focus on other work.
- NewsAugust 3, 2026
Google Gemini: AI Agents Could Soon Come to Android Devices – No Enterprise Plan Required
Android Authority has uncovered evidence in the latest Gemini app version that Google is testing the ability to create AI agents directly from smartphones – potentially opening the feature to all users, not just Workspace and Enterprise customers.
- NewsAugust 1, 2026
Google Makes AI Agent Evaluation a Standard Feature
Google's Enterprise Agent Platform now generally offers evaluation and measurement tools. Developers can now consistently assess AI agents during development and live operation.
- NewsJuly 27, 2026
Nvidia Accelerates Chip Design with Vera CPUs and AI Agents
The semiconductor maker is running its own chip design software on its own processor architecture—and integrating autonomous AI agents directly into the development pipeline.
- NewsJuly 25, 2026
ChatGPT Voice Now in Desktop App – Control Multiple Agents by Voice
OpenAI has integrated voice capabilities directly into the ChatGPT desktop application. You can now control and coordinate multiple agents running in ChatGPT Work or Codex using only your voice.
- NewsJuly 24, 2026
Google Opens Gemini Spark to Paying Users – Agentic AI Goes Mainstream
Google is expanding access to its autonomous AI assistant Gemini Spark. AI Pro and Ultra subscribers can now use the agent platform – free users are excluded.
- NewsJuly 23, 2026
OpenAI Presence: Enterprise Agents for Customer Service and Internal Workflows
OpenAI has announced a new product offering for enterprises. With **OpenAI Presence**, companies can deploy voice and chat agents to handle customer inquiries and automate internal processes.
- NewsJuly 22, 2026
Glow: New Endpoint Security Unicorn Tackles AI Agent Era
Palo Alto-based Glow has emerged from stealth with a $1.2 billion valuation and $180 million Series A. The startup addresses a growing security gap: how enterprises protect devices when AI agents and developer tools proliferate.
- NewsJuly 17, 2026
Intel and Google Cloud Expand AI Partnership – Gemini Enterprise for Chip Design
Intel integrates Google's Gemini AI across its global workforce and leverages Google Cloud for semiconductor development. The strategic collaboration expansion aims to accelerate chip design and enterprise operations.
- NewsJuly 10, 2026
OpenAI Launches ChatGPT Work – New Agent with Cross-App Actions
OpenAI has introduced ChatGPT Work, a new agent powered by Codex and GPT-5.6. The tool can take autonomous action across multiple apps and files, stay with projects for hours, and convert goals into finished work.
- NewsJuly 8, 2026
Google Expands Gemini Agents with Background Execution and Remote Protocols
The Gemini API gains new capabilities for production-ready AI agents: background tasks, remote MCP integration, and credential refresh. Developers can now build reliable, asynchronous agents in an isolated sandbox.
- NewsJuly 7, 2026
Anthropic Brings Claude Cowork to Mobile and Web – AI Agents Leave the Desktop
Anthropic has expanded Claude Cowork, its AI agent for automated tasks, to mobile devices and the web. Usage data reveals: most users aren't programmers – a turning point for agentic AI in the mainstream.
- NewsJuly 6, 2026
Anthropic Launches Claude Sonnet 5: Agentic AI at Mid-Tier Prices
Anthropic has officially released Claude Sonnet 5, a mid-tier model with capabilities previously reserved for premium systems—but at significantly lower cost. The move intensifies competition with OpenAI and Google.
- NewsJuly 4, 2026
Alibaba Cuts Agent Token Use by 99 Percent
Alibaba Qwen's new framework dramatically reduces computational costs for AI agents by loading only required tools instead of all available ones.
- NewsJuly 4, 2026
Microsoft merges Copilot into super app with autonomous agents
In August, Microsoft plans to fuse its consumer and enterprise Copilot apps into a single application. New AI agents called AutoPilot will work in the background—for an additional fee.
More topics
Anthropic 146AI Regulation 157OpenAI 109AI Security 101AI Infrastructure 84Claude 74Nvidia 64Google 46China 45Geopolitics 45DeepSeek 39AI Models 31