Skip to main content
Scour
Discover
Docs
Login
Sign Up
You are offline. Trying to reconnect...
Copied to clipboard
Unable to share or copy to clipboard
The AI Security Institute (AISI)
aisi.gov.uk
AI Security Institute
·
20h
20 hours ago
Transcript analysis for AI agent evaluations | AISI Work
Covers
ReAct: Synergizing Reasoning and Acting in Language Models
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Transcript analysis for AI agent evaluations | AISI Work
AI Security Institute
·
20h
20 hours ago
Funding 60 projects to advance AI alignment research | AISI Work
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Funding 60 projects to advance AI alignment research | AISI Work
AI Security Institute
·
20h
20 hours ago
A Decision-Theoretic Formalisation of Steganography With Applications to LLM Monitoring
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for A Decision-Theoretic Formalisation of Steganography With Applications to LLM Monitoring
AI Security Institute
·
20h
20 hours ago
International joint testing exercise: Agentic testing | AISI Work
Covers
Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Models
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for International joint testing exercise: Agentic testing | AISI Work
AI Security Institute
·
20h
20 hours ago
Our First Year | AISI Work
Covers
AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Our First Year | AISI Work
AI Security Institute
·
20h
20 hours ago
Deep ignorance: Filtering pretraining data builds tamper-resistant safeguards into open-weight LLMs
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Deep ignorance: Filtering pretraining data builds tamper-resistant safeguards into open-weight LLMs
AI Security Institute
·
20h
20 hours ago
UK AISI Alignment Evaluation Case-Study
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for UK AISI Alignment Evaluation Case-Study
AI Security Institute
·
20h
20 hours ago
How do environmental factors impact AI behaviour? | AISI Work
Covers
Agentic Misalignment: How LLMs could be insider threats
Covered by
LessWrong
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for How do environmental factors impact AI behaviour? | AISI Work
AI Security Institute
·
20h
20 hours ago
Fourth progress report | AISI Work
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Fourth progress report | AISI Work
AI Security Institute
·
20h
20 hours ago
Our 2025 year in review | AISI Work
Covers
3 stories
See all stories this covers
including
A small number of samples can poison LLMs of any size
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Our 2025 year in review | AISI Work
AI Security Institute
·
20h
20 hours ago
Lessons from studying two-hop latent reasoning
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Lessons from studying two-hop latent reasoning
AI Security Institute
·
20h
20 hours ago
Investigating models for misalignment | AISI Work
Covers
Large Language Models Often Know When They Are Being Evaluated
Covered by
LessWrong
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Investigating models for misalignment | AISI Work
AI Security Institute
·
20h
20 hours ago
Safety case template for ‘inability’ arguments | AISI Work
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Safety case template for ‘inability’ arguments | AISI Work
AI Security Institute
·
20h
20 hours ago
Evidence for inference scaling in AI cyber tasks: Increased evaluation budgets reveal higher success rates | AISI Work
Covers
Frontier AI Security
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Evidence for inference scaling in AI cyber tasks: Increased evaluation budgets reveal higher success rates | AISI Work
AI Security Institute
·
20h
20 hours ago
Why I joined AISI by Geoffrey Irving | AISI Work
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Why I joined AISI by Geoffrey Irving | AISI Work
AI Security Institute
·
20h
20 hours ago
Can AI agents escape their sandboxes? A benchmark for safely measuring container breakout capabilities | AISI Work
Covered by
indiehacker.news
,
toxsec.com
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Can AI agents escape their sandboxes? A benchmark for safely measuring container breakout capabilities | AISI Work
AI Security Institute
·
20h
20 hours ago
5 key findings from our first Frontier AI Trends Report | AISI Work
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for 5 key findings from our first Frontier AI Trends Report | AISI Work
AI Security Institute
·
20h
20 hours ago
AISI Blog
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for AISI Blog
AI Security Institute
·
20h
20 hours ago
A pipeline for transcript analysis using Inspect Scout | AISI Work
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for A pipeline for transcript analysis using Inspect Scout | AISI Work
AI Security Institute
·
20h
20 hours ago
Advanced AI evaluations at AISI: May update | AISI Work
Covers
3 stories
See all stories this covers
including
ReAct: Synergizing Reasoning and Acting in Language Models
Love
Like
Not for me
Save
See related topics
Feeds
Share
Report
Spam
Misleading
Harmful Content
Block Domain
Actions for Advanced AI evaluations at AISI: May update | AISI Work
« Page 4
·
Page 6 »
Log in to enable infinite scrolling
Keyboard Shortcuts
Navigation
Next / previous post
j
/
k
Open post
o
or
Enter
Preview post
v
Post Actions
Love post
a
Like post
l
Dislike post
d
Undo reaction
u
Save / unsave
s
Recommendations
Add interest / feed
Enter
Not interested
x
Go to
Home
g
h
Interests
g
i
Feeds
g
f
Likes
g
l
History
g
y
Changelog
g
c
Settings
g
s
Discover
g
b
Search
/
Pagination
Next page
n
Previous page
p
General
Show this help
?
Submit feedback
!
Close modal / unfocus
Esc
Press
?
anytime to show this help
Like
Save
Not for me
Report