AI Security Institute (@AISecurityInst) / X

AI Security Institute

442 posts

AI Security Institute

@AISecurityInst

We conduct scientific research to understand AI’s most serious risks and develop and test mitigations.

United Kingdom

Joined February 2024

Pinned
AI Security Institute
@AISecurityInst
Dec 18, 2025
📈 Today, we’re releasing our first Frontier AI Trends Report: evaluation results on 30+ frontier models from the past two years, showing rapid progress in chemistry and biology, cyber capabilities, autonomy, and more. ▶️Read now: aisi.gov.uk/frontier-ai-tr…
68K
AI Security Institute
@AISecurityInst
May 22, 2024
We are announcing new grants for research into systemic AI safety. Initially backed by up to £8.5 million, this program will fund researchers to advance the science underpinning AI safety. Read more: gov.uk/government/new…
237K
AI Security Institute
@AISecurityInst
Aug 12, 2025
How can open-weight Large Language Models be safeguarded against malicious uses? In our new paper with @AiEleuther, we find that removing harmful data before training can be over 10x more effective at resisting adversarial fine-tuning than defences added after training 🧵
36K
AI Security Institute
@AISecurityInst
Jul 30, 2025
📢Introducing the Alignment Project: A new fund for research on urgent challenges in AI alignment and control, backed by over £15 million. ▶️ Up to £1 million per project ▶️ Compute access, venture capital investment, and expert support Learn more and apply ⬇️
The Alignment Project
The Alignment Project
The Alignment Project
The Alignment Project
From aisi.gov.uk
125K
AI Security Institute
@AISecurityInst
Nov 19, 2024
We've released a technical report detailing our pre-deployment testing of @AnthropicAI's upgraded Claude 3.5 Model with the U.S. AI Safety Institute. Read our blog for a high-level overview.
Pre-deployment evaluation of Anthropic’s upgraded Claude 3.5 Sonnet | AISI Work
From aisi.gov.uk
85K
AI Security Institute
@AISecurityInst
Nov 5, 2024
We’re looking for talented individuals and organisations to help us build evaluations. We’ll reward bounties for new evaluations and agent scaffolding tools that assess the risks of autonomous AI systems. Find out more and apply by 30 November:
Bounty programme for novel evaluations and agent scaffolds
From aisi.gov.uk
67K
AI Security Institute
@AISecurityInst
Feb 14, 2025
From the start, we've been dedicated to providing a scientific understanding of AI’s risks to protect people's safety and security🔎🔐 Today we’re crystallising that mission and changing our name to the AI Security Institute. 1/3
110K
AI Security Institute
@AISecurityInst
May 6, 2025
🧵 Today we’re publishing our first Research Agenda – a detailed outline of the most urgent questions we’re working to answer as AI capabilities grow. It’s our roadmap for tackling the hardest technical challenges in AI security.
AISI Research Agenda
AISI Research Agenda
AISI Research Agenda
AISI Research Agenda
From aisi.gov.uk
29K
AI Security Institute
@AISecurityInst
Apr 22, 2025
🚨 New AISI research 🚨 RepliBench is a novel benchmark that measures the ability of frontier AI systems to autonomously replicate. Read the full blog here: aisi.gov.uk/work/replibenc…
30K
AI Security Institute
@AISecurityInst
Mar 5, 2025
🚨 Introducing the AISI Challenge Fund: £5 million to advance AI security & safety research. Grants of up to £200,000 are available for innovative AI research on technical mitigations, improved evaluations, and stronger safeguards. 🛡️🤖
22K
AI Security Institute
@AISecurityInst
Sep 5, 2024
Jade Leung (our CTO) and @geoffreyirving (our Research Director) have been nominated in the @TIME top 100 most influential people in AI 2024 We're incredibly proud of this team. They're proof of the immense impact technologists can have by joining the government.
6K
AI Security Institute
@AISecurityInst
Dec 11, 2024
🎉 Huge congratulations to AISI researcher @hannahrosekirk for winning a Best Paper Award at #NeurIPS2024! 🏆
Hannah Rose Kirk
@hannahrosekirk
Dec 11, 2024
A real honour and career dream that PRISM has won a @NeurIPSConf best paper award! 🌈 One year ago I was sat in a 13,000+ person audience of NeurIPs '23 having just finished data collection. Safe to say I've gone from feeling #stressed to very #blessed 😁
34K
AI Security Institute
@AISecurityInst
Aug 29, 2025
🚨Open-weight AI models are becoming more powerful, now knocking on the door of today’s closed-weight frontier. This poses critical safety challenges – how can we prevent the misuse of models whose parameters are free to download online? 🧵
24K
AI Security Institute
@AISecurityInst
Oct 22, 2025
🔒How can we prevent harm from AI systems that pursue unintended goals? AI control is a promising research agenda seeking to address this critical question. Today, we’re excited to launch ControlArena – our library for running secure and reproducible AI control experiments🧵
12K