1. Skip to content
  2. Skip to main menu
  3. Skip to more DW sites
In focus
IsraelEthiopiaUN
Latest audioLatest videos
Live TV
Digital WorldGlobal issues

Nvidia’s Cage for AI: can it prevent disaster?

04:04

This browser does not support the video element.

October 1, 2026

Nvidia has launched a new safety platform meant to catch AI agents overstepping their limits and quarantine them. Could this be the safety net autonomous AI needs?

Skip next section Similar stories

Similar stories

OpenAI model bypasses internet safeguards

During a test, a model managed to bypass a supposedly secure system and obtain responses from a chatbot.
TechnologySeptember 29, 202609:57 min

OpenAI shelves new AI model amid safety concerns

The decision comes amid a broader push within the industry to slow AI development until safety measures can catch up.
TechnologySeptember 29, 2026

Trump says AI companies agree to 'self-police'

The agreement comes amid mounting concerns over AI safety after recent hacking incidents.
PoliticsSeptember 30, 2026
Show more
Skip next section More from this show

More from this show

ChatGPT-6 Astra: Almost human or just hyped?

Will AGI change your life? The reality behind OpenAI’s claim
Digital WorldSeptember 10, 202603:23 min
Skip next section About the show

About the show

Image: Colourbox

TopTechNews

Dive into the fast-paced world of tech with the TopTechNews. Stay ahead of the game as we unpack the latest AI breakthroughs, dissect social media trends and tackle deepfakes and misinformation.

Go to show TopTechNews
Skip next section DW's Top Story

DW's Top Story

EU ministers mull new ways to curb irregular migration

The proposals include setting up so-called return hubs outside the EU and strengthening the border agency Frontex.
MigrationOctober 2, 2026

EU states plan first deportation center in Africa for 2027

PoliticsSeptember 30, 2026
Skip next section More stories from DW

More stories from DW

Go to homepage