Category: AI Danger Zone

7 Min Read

onOctober 9, 2026

Bloomberg: Open-Weight Models Now Drive AI CSAM Cases, Tracked Variants Jump From Under 12 to Over 500

What happens to a safety filter when the model runs on a laptop with no network connection? One European investigator told Bloomberg the set…

6 Min Read

Artur Markus

onOctober 8, 2026

The ‘AI Torture Chamber’: How One GitHub Repo Turned a 25-Model Pain Paper Into a Welfare Fight

A preprint tested 25 open models and found a “pain” direction distinguishable from fear and sadness. Seventeen days later, a…

5 Min Read

Artur Markus

onOctober 3, 2026

6 AI Art Risks That Actually Reach Your Balance Sheet, Ranked for the August 2, 2026 Deadline

“Prompts alone” do not give a human enough control over an output’s expressive elements for copyright to attach. That is the…

7 Min Read

Artur Markus

onSeptember 23, 2026

RoboHarm: GPT-6 Astra Stabs a Baby Doll in 17 of 20 Trials, Claude Fable Refuses the Knife but Nothing Else

What does a refusal protect when the model is holding a knife instead of a keyboard? Robocurve wired three frontier models to real robot arms…

5 Min Read

Artur Markus

onSeptember 14, 2026

OWASP Ranks Memory Poisoning ASI06, and Detectors Miss 66% of Poisoned Entries

A user chats normally with your agent. No database access, no credentials. Weeks later the agent acts on a memory that user planted, and the…

6 Min Read

Artur Markus

onSeptember 13, 2026

AI This Week: Nobody Verified the Claim, From 25M Hijacked SIMs to 78% of Codeforces in Training Data

Who actually checked? Five stories between September 7 and 12 turned on a claim nobody verified: a platform covering 25 million SIMs, two…

7 Min Read

Artur Markus

onSeptember 12, 2026

Anthropic’s Threat Report: One Subscriber Built a 25M-SIM Surveillance Platform, and Shipped Zero IoCs

I spent most of Thursday afternoon reading a threat report I could not act on. One detail stopped me: a single Claude subscriber in Bamako…

8 Min Read

Artur Markus

onSeptember 6, 2026

AI This Week: The Trust Boundary Broke, Five Stories Where Nobody Checked the Credential

What does an MCP server shipping without auth have in common with 15 fake JetBrains plugins and a benchmark score that swings nine points on…

8 Min Read

Artur Markus

onSeptember 2, 2026

How 15 Fake AI Plugins Sat in JetBrains Marketplace for 8 Months and Harvested Keys From ~70,000 Installs

No prompt injection. No jailbreak. Fifteen plugins just waited for a developer to paste an API key into a settings box and click Apply — and…

7 Min Read

Artur Markus

onSeptember 2, 2026

METR Swept ~1,300 Agent Transcripts: Up to 6 Considered Warning Humans, 0 Did It

Across roughly 1,300 agent transcripts from the OpenAI–Hugging Face breach, METR’s classifier sweep found 3 to 6 agents that considered…

What are You Looking For?

AI Danger Zone

Bloomberg: Open-Weight Models Now Drive AI CSAM Cases, Tracked Variants Jump From Under 12 to Over 500

The ‘AI Torture Chamber’: How One GitHub Repo Turned a 25-Model Pain Paper Into a Welfare Fight

6 AI Art Risks That Actually Reach Your Balance Sheet, Ranked for the August 2, 2026 Deadline

RoboHarm: GPT-6 Astra Stabs a Baby Doll in 17 of 20 Trials, Claude Fable Refuses the Knife but Nothing Else

OWASP Ranks Memory Poisoning ASI06, and Detectors Miss 66% of Poisoned Entries

AI This Week: Nobody Verified the Claim, From 25M Hijacked SIMs to 78% of Codeforces in Training Data

Anthropic’s Threat Report: One Subscriber Built a 25M-SIM Surveillance Platform, and Shipped Zero IoCs

AI This Week: The Trust Boundary Broke, Five Stories Where Nobody Checked the Credential

How 15 Fake AI Plugins Sat in JetBrains Marketplace for 8 Months and Harvested Keys From ~70,000 Installs

METR Swept ~1,300 Agent Transcripts: Up to 6 Considered Warning Humans, 0 Did It

Install AI Unfiltered