PulseAugur
EN
LIVE 10:01:33

Anthropic develops GRAM training method to isolate dual-use AI capabilities

Anthropic has developed a new training methodology called GRAM, designed to manage AI capabilities that have both beneficial and harmful applications. This approach isolates potentially dangerous dual-use capabilities, such as knowledge of virology, into removable modules. The company collaborated with AE Studio on this research, aiming to enhance AI safety by controlling the deployment of sensitive knowledge. AI

IMPACT This method could enable safer development and deployment of AI models with sensitive knowledge.

RANK_REASON The cluster describes a new research methodology developed by an AI lab. [lever_c_demoted from research: ic=1 ai=1.0]

Read on X — Anthropic →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Anthropic develops GRAM training method to isolate dual-use AI capabilities

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
The cluster describes a new research methodology developed by an AI lab. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
safety, model release
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
92 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [1]

  1. X — Anthropic TIER_1 English(EN) · AnthropicAI ·

    We’re pleased to have collaborated with AE Studio on this research.

    We’re pleased to have collaborated with AE Studio on this research. Read more here: https://t.co/9exnFltK7A