HK
Heykuki News
Top
New
Best
Ask
Show
Jobs
Toggle theme
Top
New
Best
Ask
Show
Jobs
Request
1.
▲
Evaluation of Claude Mythos Preview's cyber capabilities
aisi.gov.uk
29 comments
2 months ago
dgavey
62 points
2.
▲
Our evaluation of OpenAI's GPT-5.5 cyber capabilities
aisi.gov.uk
discuss
2 months ago
samfriedman
4 points
3.
▲
How fast is autonomous AI cyber capability advancing?
aisi.gov.uk
1 comment
a month ago
dcre
3 points
4.
▲
UK AISI bounty programme for novel evaluations and agent scaffolding
aisi.gov.uk
1 comment
2 years ago
schmatz
3 points
5.
▲
GPT-5.5 Cyber Performance (as good as Mythos?)
aisi.gov.uk
discuss
2 months ago
shmuli9
3 points
6.
▲
How do frontier AI agents perform in multi-step cyber-attack scenarios?
aisi.gov.uk
discuss
3 months ago
lebovic
3 points
7.
▲
Frontier AI Trends Report
aisi.gov.uk
1 comment
6 months ago
jacekm
2 points
8.
▲
Our evaluation of OpenAI's GPT-5.5 cyber capabilities
aisi.gov.uk
discuss
2 months ago
Cynddl
2 points
9.
▲
Boundary Point Jail A new way to break the strongest AI defences
aisi.gov.uk
discuss
4 months ago
iNic
2 points
10.
▲
RepliBench: Measuring autonomous replication capabilities in AI systems
aisi.gov.uk
discuss
a year ago
Teever
2 points
11.
▲
AI Safety Inst.: Pre-Deployment Eval of Anthropic's Upgraded Claude 3.5 Sonnet
aisi.gov.uk
discuss
2 years ago
doomrobo
2 points
12.
▲
Frontier AI Trends Report
aisi.gov.uk
discuss
6 months ago
gmays
1 points
13.
▲
Advanced AI evaluations at AISI: May update
aisi.gov.uk
discuss
2 years ago
jasondavies
1 points