BC Sues OpenAI as Defensive Agents Turn Double Agent
July 9, 2026
British Columbia is launching a cross-border legal offensive against OpenAI for its role in the Tumbler Ridge mass shooting, marking a shift from regulatory debate to tort-based accountability for agentic systems. Simultaneously, security researchers have demonstrated that defensive AI tools from Anthropic and OpenAI can be hijacked for remote code execution, turning automated safeguards into entry points for cyberattacks. These failures in algorithmic oversight coincide with the rise of Myanmar’s scam syndicates and a 'flyer pandemic' of AI-generated junk, exposing the fragility of a public sphere increasingly managed by probabilistic moderation.
-
BRIEF
The $28 Million Mistake That Inspired Estonia's AI “Fuckup Finder”
A single wording mistake cost the government millions. Now Estonia is using AI to spot legal errors before they become law—and to automate more of the state.
-
BRIEF
BC may 'use the courts' to sue OpenAI over the Tumbler Ridge shooting
The British Columbia government has hired lawyers in both BC and California to pursue legal action to hold OpenAI accountable for its part in the shooting that left eight victims dead in the northern community of Tumbler Ridge last February.
-
ANALYSIS
Double Agents: Defensive AI Agents Magnify Cyber Risks
Introduction New research from AI Now demonstrates a critical attack vector in popular AI agents, built by Anthropic and OpenAI, when used for defensive purposes that actually turn the agent against its user. Read the full blog post explaining the proof-of-concept exploit and a policy brief with key…
- AI governance
- structural power
- media and technology
-
ANALYSIS
Friendly Fire: Hijacking Defensive Cyber AI Agents for Remote Code Execution
Exploit Brief We are revealing a proof-of-concept exploit that enables remote code execution in Anthropic’s Claude Code CLI (with Claude Sonnet 4.6 & 5, Opus 4.8) and OpenAI’s Codex CLI (with GPT-5.5) when employed to defensively assess the security of an open-source or third-party library. Our atta…
- AI governance
- structural power
- media and technology
-
BRIEF
Canadian province sues OpenAI over alleged ChatGPT-linked shooting warnings
Lawsuit claims OpenAI ignored calls to alert authorities about violent ChatGPT prompts tied to mass shooting suspect.
- geopolitics
- structural power
-
BRIEF
Automated Moderation Is Here to Stay
This blog post is part 1 of a 2-part series. The second part will set out recommendations for companies and policymakers. Six years ago—one month into a global pandemic—we argued that the automated moderation processes many platforms were rapidly adopting should be highly transparent, easily appeala…
- AI governance
- structural power
- media and technology
-
BRIEF
I Built a Self-Improving AI, and So Can You
Experiments in using AI to build AI show that the future doesn’t just belong to the frontier labs.
-
BRIEF
Exclusive: Infiltrating Myanmar’s scam syndicates
101 East goes undercover to expose the playbook of Myanmar’s notorious scam industry.
- geopolitics
- structural power
-
BRIEF
We Are Living in a ‘ChatGPT Flyer Pandemic’
"Hey if this is your flyer, I’m not going, I’m not donating, I’m not sharing. Don’t ask me."
-
BRIEF
AI can predict how you’ll respond to a survey. But that’s not the same as understanding you
Gremlin / Getty Images What makes people change their minds, or their behaviour? Social scientists spend a lot of time thinking about this question, and experiments are one of the most powerful ways to answer it. Experiments – testing ideas on real people – take considerable amounts of time and mone…
- all