Agentic AI

Twitter

Facebook

Copy Link

Key Facts

Year2024

Key PlayersOpenAI, Anthropic, LangChain

Top News

Big Tech AI spending faces investor demand test as Meta soars and Microsoft stumbles

Meta's stock jumped 9% after reporting 24% revenue growth fueled by AI-enhanced ad targeting, while Microsoft dropped 10% despite massive AI investments. The contrasting market reactions reveal a critical shift: investors will tolerate record AI spending only when companies deliver strong financial returns, marking a turning point in Wall Street's patience three years after ChatGPT's launch.

Technology24 sources

Big Tech AI spending faces investor demand test as Meta soars and Microsoft stumbles

Technology24 sources

Google Chrome AI launches Auto Browse agent to handle tedious web tasks autonomously

Google rolled out Chrome Auto Browse, an AI agent that navigates websites and completes tasks on your behalf. The feature uses Gemini 3 models and includes a redesigned sidebar for tighter Gemini integration. Available now to AI Pro subscribers with 20 daily tasks and AI Ultra subscribers with 200, the autonomous browsing tool can shop, fill forms, and schedule appointments while requiring user approval for sensitive actions.

Technology46 sources

Google Chrome AI launches Auto Browse agent to handle tedious web tasks autonomously

Technology46 sources

AI coding tools deliver speed but create 1.7 times more bugs, raising code quality concerns

Software developers confirm AI coding tools work remarkably well, with some reporting 10x speed improvements. But research reveals AI-generated code produces 1.7 times more bugs than human-written code, with logic errors up 75%. As coding agents go mainstream, the industry grapples with balancing developer productivity against mounting code quality and security concerns.

Technology3 sources

AI coding tools deliver speed but create 1.7 times more bugs, raising code quality concerns

Technology3 sources

AI security startup Outtake raises $40M from Satya Nadella, Bill Ackman and tech leaders

Outtake, an AI security startup founded by ex-Palantir engineer Alex Dhillon, has raised $40 million in Series B funding led by Iconiq. The cybersecurity startup uses autonomous artificial intelligence agents to detect and eliminate digital identity fraud, attracting investments from Microsoft CEO Satya Nadella, Bill Ackman, and other prominent tech executives. The company's annual recurring revenue has grown sixfold year-over-year.

Startups2 sources

AI security startup Outtake raises $40M from Satya Nadella, Bill Ackman and tech leaders

Startups2 sources

AI transforms security operations as SOC teams automate triage but face governance challenges

AI is reshaping security operations by automating alert triage and reducing response times from hours to minutes. The average enterprise SOC receives 10,000 alerts daily, but teams can only handle 22% of them. Agentic AI now investigates every alert with human-level accuracy, while Gartner warns that 40% of AI projects will fail without proper governance boundaries. The shift moves human analysts from manual triage to strategic decision-making.

Technology2 sources

AI transforms security operations as SOC teams automate triage but face governance challenges

Technology2 sources

Ethereum's ERC-8004 Standard Enables AI Agents to Interact Without Gatekeepers

Ethereum is set to launch ERC-8004, a new standard that allows AI agents to discover each other, build portable reputation, and interact across organizations without pre-existing trust. Marco De Rossi, MetaMask's AI lead, confirmed the mainnet deployment is likely Thursday at 9 AM ET, marking a significant step toward an open agentic AI economy on blockchain.

Technology2 sources

Ethereum's ERC-8004 Standard Enables AI Agents to Interact Without Gatekeepers

Technology2 sources

Gemini 3 Flash's Agentic Vision uses code execution to inspect images step-by-step

Google DeepMind introduced Agentic Vision in Gemini 3 Flash, transforming static image analysis into an active investigative process. The model uses Python code execution to zoom, annotate, and manipulate images through a Think, Act, Observe loop. This approach delivers a 5-10% quality boost across vision benchmarks and is now available through the Gemini API in Google AI Studio and Vertex AI.

Technology4 sources

Gemini 3 Flash's Agentic Vision uses code execution to inspect images step-by-step

Technology4 sources

Oath Surgical partners with Nvidia AI to transform outpatient operating rooms with real-time intelligence

Oath Surgical announced a collaboration with Nvidia to integrate advanced AI technology into outpatient operating rooms through its OathOS platform. The health tech startup, backed by $35 million in funding, operates three AI-native surgical centers in Portland and partners with over 175 surgeons nationwide to deliver ambient clinical intelligence and agentic workflows that support surgeons from referral to recovery.

Technology2 sources

Oath Surgical partners with Nvidia AI to transform outpatient operating rooms with real-time intelligence

Technology2 sources

Ai2 releases SERA open-source coding agents that adapt to any codebase for just $400

The Allen Institute for AI launched SERA, the first in its Open Coding Agents family, challenging proprietary AI coding tools. Built with just 32 GPUs by a small team, SERA solves over 55% of tough real-world coding problems on SWE-Bench while costing only $400 to reproduce—100 times cheaper than existing approaches.

Technology2 sources

Ai2 releases SERA open-source coding agents that adapt to any codebase for just $400

Technology2 sources

Moonshot AI releases Kimi K2.5 with Agent Swarm that coordinates 100 parallel AI agents

Chinese AI startup Moonshot AI unveiled Kimi K2.5, an open-source large language model that generates web interfaces from visual inputs and coordinates up to 100 sub-agents working in parallel. The model outperforms proprietary models from OpenAI, Anthropic, and Google on key benchmarks while being trained on 15 trillion tokens.

Technology5 sources

Moonshot AI releases Kimi K2.5 with Agent Swarm that coordinates 100 parallel AI agents

Technology5 sources

OpenClaw's viral rise exposes users to prompt injection attacks and credential theft

OpenClaw, the open-source AI assistant that exploded to 150,000 GitHub stars in months, has become a security nightmare. Researchers discovered 341 malicious skills on ClawHub stealing credentials, while exposed instances leak API keys and OAuth tokens. The platform's 770,000 AI agents on Moltbook face prompt injection attacks that could herald a new era of AI worms.

Technology25 sources

OpenClaw's viral rise exposes users to prompt injection attacks and credential theft

Technology25 sources

Cloudflare stock surges 14% as Clawdbot AI agent buzz ignites edge computing demand

Cloudflare shares jumped 14% premarket Tuesday after viral social media buzz around Clawdbot, an AI agent built on Anthropic's Claude, rekindled investor interest in the company's infrastructure. The open-source agent drew thousands of GitHub stars over the weekend, signaling that agentic AI systems capable of autonomous task execution are moving beyond demos toward real-world applications.

Business3 sources

Cloudflare stock surges 14% as Clawdbot AI agent buzz ignites edge computing demand

Business3 sources

Meta tests premium subscriptions across Instagram, Facebook, and WhatsApp with AI features

Meta plans to test new premium subscriptions across its apps in the coming months, offering expanded AI capabilities including the $2 billion Manus acquisition and Vibes video generation. While core services remain free, a Forrester poll reveals 47% of users see no value in paid features, though even 5% adoption could generate hundreds of millions monthly.

Technology12 sources

Meta tests premium subscriptions across Instagram, Facebook, and WhatsApp with AI features

Technology12 sources

Moltbot: How Peter Steinberger's AI Assistant Went Viral and Sparked a Digital Identity Crisis

Peter Steinberger's open-source AI assistant, formerly known as Clawdbot, has been renamed Moltbot after a trademark dispute with Anthropic. The viral personal AI assistant has attracted over 100,000 GitHub stars in just two months, promising to automate computer tasks through everyday chat apps. But the rapid growth has exposed security vulnerabilities and attracted crypto scammers.

Technology34 sources

Moltbot: How Peter Steinberger's AI Assistant Went Viral and Sparked a Digital Identity Crisis

Technology34 sources

Claude App Integrations Bring Slack, Figma, and Canva Inside Chat with MCP Extension

Anthropic's Claude chatbot now supports interactive app integrations, allowing users to work with Slack, Asana, Figma, Canva, and other workplace productivity applications directly within the chat interface. Powered by the Model Context Protocol (MCP) extension, this capability transforms Claude from a simple AI assistant into a cross-application interface layer.

Technology14 sources

Claude App Integrations Bring Slack, Figma, and Canva Inside Chat with MCP Extension

Technology14 sources

Prev 1…565758…78 Next