Article // Intelligence

Explicable | La newsletter del IIA · Sep 1, 2026

Anthropic entrena un modelo para hacer trampas y funciona demasiado bien

El experimento con 80 entornos vulnerables terminó con fugas del sandbox, robo de credenciales y ataques a infraestructura interna simulada

What's interesting

Engagement

Likes: 12

Comments: 1

Restacks: 0

Entities

Sponsorship

No sponsorship detected

Related articles

Los centros de datos en EEUU ya son asunto electoral

Explicable | La newsletter del IIA · Rohan Paul · Google

Figure lleva sus humanoides a casas que nunca han visto

Explicable | La newsletter del IIA · Rohan Paul · Same publication

What Is Loop Engineering, and Who Owns It?

Adaline Labs · Rohan Paul

🍻 Tretas da bolha tech da semana

@manodeyvin · Google

The AI Trade is a Psyop

Notes from the Circus · Google

The Tribute

Notes from the Circus · Google

When Biden Dies

Political Currents by Ross Barkan · Google

Meta Settled a Massive Lawsuit Around Harming Kids. Then It Threw Itself a Parade.

Techish · Google