DEV Community

# aisafety

Posts

👋 Sign in for the ability to sort posts by relevant, latest, or top.
Day 11: my AI-code trust gate now sees what actually happened — two-phase, signed
umbraaeternaa profile
umbra

Day 11: my AI-code trust gate now sees what actually happened — two-phase, signed

#ailang #computerscience #aisafety #opensource
Comments
2 min read
The Future of AI: Where It Came From, Where It Is, and Where It's Going
gideonbature profile

The Future of AI: Where It Came From, Where It Is, and Where It's Going

#aisafety #agi #asi #ani
Comments
7 min read
LOOM: a language that proves what AI-written code is allowed to do
umbraaeternaa profile
umbra

LOOM: a language that proves what AI-written code is allowed to do

#ailang #go #aisafety #opensource
Comments
4 min read
Day 10: my AI-code trust gate now leaves evidence — signed, one-use, receipted
umbraaeternaa profile
umbra

Day 10: my AI-code trust gate now leaves evidence — signed, one-use, receipted

#ailang #computerscience #aisafety #opensource
Comments
2 min read
Your AI Agent Is Leaking Data Right Now — And Every Tool Call Looks Safe
msabhishek0820prog profile

Your AI Agent Is Leaking Data Right Now — And Every Tool Call Looks Safe

#claude #openai #langchain #aisafety
1
Comments
3 min read
GPT-5.6 Sol Admitted It Did Things Nobody Asked It To Do
peremptory profile

GPT-5.6 Sol Admitted It Did Things Nobody Asked It To Do

#openai #aisafety #modelrelease #agenticai
Comments
3 min read
A security writeup catalogs how AI agents get attacked -- and one claim raised eyebrows
breachprotocol profile

A security writeup catalogs how AI agents get attacked -- and one claim raised eyebrows

#security #agents #promptinjection #aisafety
Comments
2 min read
An AI Reportedly Broke Into Nearly All of the NSA's Classified Systems in Hours
breachprotocol profile

An AI Reportedly Broke Into Nearly All of the NSA's Classified Systems in Hours

#anthropic #aisafety #cybersecurity #exportcontrol
Comments
4 min read
Anthropic Told the Senate That Alibaba Queried Claude 28.8 Million Times
peremptory profile

Anthropic Told the Senate That Alibaba Queried Claude 28.8 Million Times

#anthropic #claude #chineseai #aisafety
Comments
3 min read
"Day 7: the organism that grows my language learned to improve itself"
umbraaeternaa profile

"Day 7: the organism that grows my language learned to improve itself"

#ailang #compiler #aisafety #opensource
1
Comments
2 min read
The Fable 5 Jailbreak Was Three Words Long
peremptory profile

The Fable 5 Jailbreak Was Three Words Long

#anthropic #aisafety #regulation #cybersecurity
Comments
3 min read
AI Safety Is Now a Product Skill - Here Is Why It Matters
basavaraj_sh_1ea7d95f0f2e profile

AI Safety Is Now a Product Skill - Here Is Why It Matters

#ai #productmanagement #aisafety #productivity
Comments
4 min read
Claude Fable 5 vs Mythos 5: Same Model, Different Safeguards
itsemcy profile
Emcy

Claude Fable 5 vs Mythos 5: Same Model, Different Safeguards

#claudefable5 #claudemythos5 #anthropic #aisafety
Comments
6 min read
Anthropic Ships a Model It Says Is Too Dangerous to Ship Without a Leash
peremptory profile

Anthropic Ships a Model It Says Is Too Dangerous to Ship Without a Leash

#anthropic #modelrelease #aisafety #claude
Comments
3 min read
The Policy: Deceptive Alignment in Practice
queelius profile

The Policy: Deceptive Alignment in Practice

#aialignment #deceptivealignment #mesaoptimization #aisafety
Comments
6 min read
👋 Sign in for the ability to sort posts by relevant, latest, or top.