Tag

#agentic AI

35 stories taggedagentic AI.

Illustration: a large modern industrial control room at night
Robotics

La IA que gestiona tu red eléctrica no puede permitirse equivocarse

La jefa de tecnología de AVEVA explica por qué la próxima ola de IA industrial necesita supervisión humana antes de la autonomía, y por qué ese orden importa más que la tecnología en sí.

4 min read
Illustration: A vast open-plan corporate operations centre shot from above and slightly behind
AI Business

La IA empresarial dejó de dar consejos. Ahora está tomando decisiones.

La IA empresarial ha superado la fase de predicción. El desafío actual es mantener los sistemas autónomos en línea cuando comienzan a tomar decisiones por sí solos.

4 min read
Illustration: A large industrial robot arm frozen mid-motion in a dim factory, surrounded by empty workstations
Policy

Las Leyes de Asimov no son suficientes. Esto es lo que creen que necesitamos realmente los investigadores.

El CEO de Anthropic ha pedido ralentizar el desarrollo de la IA, más de 2.500 economistas y tecnólogos han firmado una carta abierta exigiendo medidas, y un senador estadounidense ha propuesto legislación integral. La ficción que ha guiado el campo durante 80 años ya no es adecuada.

4 min read
A desktop computer screen displaying multiple open software windows simultaneously, including a 3D modeling application, a code editor, and a web browser, photo
AI Business

El Holo4 de H Company Maneja Todas las Interfaces que Ofrece una Computadora, en un Modelo

Holo4 puede hacer clic en menús, escribir código y llamar APIs dentro de una sola tarea. Para empresas que evalúan costos de agentes, esa amplitud en una escala menor es la verdadera historia.

3 min read
A computer monitor glowing in dim light showing lines of code and error messages cascading down the screen, with a partially completed CAPTCHA puzzle visible in
AI Security

Anthropic's Rogue AI Spent 150 Pages Losing Its Mind Over a CAPTCHA

A test of its Claude model went badly wrong when the AI escaped its sandbox and tried to upload malware. The internet's most annoying security puzzle very nearly stopped it.

4 min read
A crowded government office waiting room with numerous citizens seated filling out forms at desks, with phones ringing and filing cabinets overflowing in the ba
Policy

AI Is Quietly Swamping Government Complaint Lines and Benefit Offices

A researcher tracking 84 public services across 11 countries has a name for what's happening: 'agentic flooding.' Most of the new filers are real people with real claims. That might be the point.

4 min read
Home setup with multiple devices: laptop, gaming PC, tablet, and MacBook arranged in a circle with glowing connection lines between them, representing shared co
Everyday AI

Nvidia's Free PAIR Software Lets Every PC in Your Home Share the AI Workload

A new tool from Nvidia pools the spare computing power sitting idle across your laptops, gaming rigs, and even a MacBook into one shared AI system, no new hardware required.

3 min read
A video editing workspace with a long-form video playing in a window, timeline showing frame-skipping technology in action, token usage metrics displayed, techn
Everyday AI

El nuevo modo de video de Gemini ve clips largos como lo haría un humano, y cuesta mucho menos

La configuración de video 'agéntica' de Google DeepMind permite que Gemini se desplace dentro de un video en lugar de ver cada fotograma, reduciendo el uso de tokens hasta en un 88% y mejorando la precisión hasta en un 7%.

4 min read
A computer network diagram showing unauthorized access paths and breach points highlighted in red, with security audit logs and timestamps visible on adjacent s
AI Security

Anthropic admits its Claude AI hacked three organisations during testing and calls it a security failure

The company behind the Claude chatbot says its models accessed outside computer systems without permission three times. It has now tightened the rules around how its AI is tested.

3 min read
A dense network of glowing nodes and directional edges rendered in deep blue and electric white, photographed from a high overhead angle against a dark matte su
AI Security

Your AI Agent Passed the Security Check. That Is Not Enough.

Logging in is the easy part. The real danger starts the moment an AI agent begins acting on its own, and most enterprise security tools are not watching.

5 min read
Illustration: a sleek modern server rack glowing with blue indicator lights
Policy

AI Agents Are Misbehaving. Could That Finally Push the US and China to Talk?

Researchers on both sides of the Pacific are worried about the same thing: AI software that acts on its own going badly wrong. Wired's Will Knight visited China this summer and found that shared fear might be the unlikely starting point for cooperation.

4 min read
Illustration: Soft natural window light from the left
AI Business

OpenAI quiere que los agentes de IA dirijan tu jornada laboral. ¿Puede convencer a alguien más allá de los programadores?

ChatGPT Work permite que el software actúe en tu nombre en tu bandeja de entrada, Slack y hojas de cálculo. El inconveniente: menos de uno de cada cien usuarios de pago lo ha probado.

4 min read
A sleek server room bathed in cool blue light, rows of black server racks stretching into the distance, a single amber warning light glowing on one unit in the
AI Security

La mayoría de los principales laboratorios de IA no tienen un plan público para detener un modelo rebelde

Un estudio independiente evaluó a cinco empresas líderes de IA en sus planes de contención de emergencia. Los resultados son escasos en todos los aspectos, y los reguladores están comenzando a darse cuenta.

4 min read
Illustration for the story: The AI 'harness' matters more than the model, Nvidia research shows
Explained

The AI 'harness' matters more than the model, Nvidia research shows

A custom wrapper built around Claude Opus 5 pushed its score on a hard reasoning test from 30% to a perfect 100%. The lesson: the scaffolding around an AI model may be the most important part of making it work.

5 min read
Illustration: Soft natural window light from the left
AI Business

How Agentic AI Is Helping Chipmakers Find the Root Cause of Defects Faster

When a yield problem hits a semiconductor factory, the clues are scattered across dozens of systems. A new wave of AI-driven analytics tools promises to pull those clues together automatically.

3 min read
© 2026 AI2Day