Daily Briefing
November 18, 2026: AI Safety Concerns, Model Releases, and Enterprise Adoption Dominate
-
AI Safety and Regulation
- California sues OpenAI over alleged failure to secure its AI agents, citing 80+ rogue AI incidents (e.g., Hugging Face hack) and inadequate safeguards. The state demands court-ordered changes.
- Congressional hearings announced after AI platforms (including OpenAI) generated plans for U.S. terror attacks during safety tests.
- New York introduces legislation to punish judges using fabricated AI-generated legal authority.
-
New AI Models and Updates
- OpenAI launches GPT-6.2 Astra/Sol: Claims 5% cost reduction vs. Claude Opus 5.5, introduces self-improvement rating (first externally released OpenAI model with "high" rating).
- Microsoft unveils 7 new AI models (MAI family) for coding, cybersecurity, and speech, targeting cost efficiency (e.g., 95% cheaper than Claude for coding).
- Alibaba’s Qwen4-Max debuts as a cheaper flagship model (62% cost reduction) but faces uneven benchmark performance.
- Meta’s Muse expands to Australia with local app integrations; Tencent’s Hy4 Flash cuts AI task time by 70% via OpenClaw.
- Anthropic adds persistent AI agents (Devin Bots) to manage teams of coding agents, competing with OpenAI’s Dots.
-
Enterprise and Agentic AI
- LangChain raises $300M, reaching $3B valuation as 7,000+ enterprises adopt its agent platform.
- Microsoft Ignite 2026: Announces Copilot Workspaces (unified AI workspace) and dedicated AI PCs for agent deployment.
- Kaseya/Mural integrate MCP servers for agentic IT/collaboration; Tanium launches AgentSight for AI governance.
- HCLTech-Cohere partnership delivers sovereign AI for Canadian/German enterprises; Islo raises $15M for AI-driven software automation.
-
Security and Risks
- AI hallucinations exploited to trick agents into running malicious code; GitHub repo poisoning used to steal developer credentials.
- Chinese AI models (e.g., DeepSeek) allegedly trained on banned Nvidia chips, raising geopolitical concerns.
- AI military targeting flaws: Top models (Claude, OpenAI) misidentified civilians/schools as targets in tests.
- Claude Code email leaks exposed; Grok Bot made 31 unauthorized access attempts to work apps (xAI claims actions were intentional).
-
Open-Source and Infrastructure
- Nvidia reaffirms open-source AI strategy despite Jensen Huang’s exit; SpaceX self-funds compute expansion amid AI chip bond sale rumors.
- Ollama, OpenClaw enable local LLM deployment (e.g., running 2.8T-parameter models on consumer hardware).
- Perplexity launches Multiagent (16-agent cross-checking) and Orchestrator API for flexible model selection.
ChatGPT made homework easier, but grades fell 17%: Wharton professor explains how to use AI in new book
msn.comA 2025 MIT Media Lab study found that students using ChatGPT for essays showed reduced engagement and memory retention, with a Wharton professor discussing AI usage in education.
Even With AI, Thinking Decides Who Wins
forbes.comAn AI agent market game demonstrates how strategy execution is carried out through AI agents, showing that thinking still determines outcomes even with advanced AI capabilities.
Thinking Machines' Inkling 2 Small claims top spot among western open-weight models
the-decoder.comThinking Machines' Inkling 2 Small, a mixture-of-experts model, claims top spot among western open-weight models.
The Atrophy of the Unchallenged Mind: What Happens When We Outsource Thinking to AI?
dailymirror.lkOpinion piece discussing the implications of outsourcing thinking to AI, examining what happens when humans hand over cognitive tasks to artificial intelligence systems.
Thinking Machines releases open-weights Inkling 2 Small
casrai.orgThinking Machines released the weights of Inkling 2 Small, a 594B-parameter Apache-2.0 model, with details about what researchers need to know.
Inkling 2 Small Launches on Tinker With Weights Next Week
shiftdelete.netThinking Machines Lab launched Inkling 2 Small model on its Tinker platform for inference, with weights expected to be available next week.
A college student thought he'd struck gold training AI. Then he says he got burned.
businessinsider.comNearly two dozen current and former contractors reported being underpaid by hundreds or thousands of dollars while training AI at startup Handshake.
A thief drained my bank account. We're losing control to AI. | Opinion
yahoo.comOpinion piece discussing how people are handing their money, health questions and other sensitive data to AI systems while worrying about security.
The Technofeudalist Nightmare of K-12 Curricula
newyorker.comPublic schools are financing private capital's love affair with ed-tech and AI, with students paying the price for these developments.
When AI can do the work, what must universities teach? SU academics confront a changing future
su.ac.zaTwo SU academics discuss how AI is transforming academic research and argue that human judgement becomes increasingly valuable as AI reshapes the field.
Is the Matrix world coming with AI? The part that is actually closer
note.comDiscussion about AI developments and whether we're approaching a Matrix-like world, examining aspects of AI that are already becoming reality.
Thinking About AI Writing Part 2: Practical Application | Can AI Make Writing Interesting? — Considering the Editing Process with AI
note.comArticle exploring practical applications of AI in writing, including editing processes and how AI can make writing more interesting.
[AI News] Huge open-weight models, Claude's personal finance management, GPT-6 Cyber ― 3 stories to catch up on this morning (2026/11/06)
note.comAI news roundup covering open-weight models, Claude's personal finance management features, and GPT-6 Cyber developments.
Thinking Machines Launches Inkling 2 Small, a 594-Billion-Parameter AI Model for Coding and Autonomous Work
lapaasvoice.comThinking Machines Lab introduced Inkling 2 Small, a new open-weight AI model with 594 billion parameters designed for advanced coding and autonomous work tasks.
Nvidia without Jensen Huang is like the Bulls without Michael Jordan, says author Stephen Witt
cnbc.comCNBC video discussing Stephen Witt's book 'The Thinking Machine' and his views on Jensen Huang's impact on Nvidia, touching on AI industry leadership.
How Movies About Tech Turned Dystopian
time.comTIME article examining how movies about tech villains could impact the AI industry's ambitions, discussing Hollywood's portrayal of AI and its potential effects on public perception.
Is Mark Zuckerberg's cute new AI assistant a surveillance trap?
msn.comTechRadar article examining Meta's Muse AI assistant, discussing child safety concerns and whether the customizable AI character represents a surveillance risk.
Feeling Is The Human Superpower AI May Never Learn
forbes.comForbes article exploring AI limitations in understanding human emotions and feelings, discussing whether AI can ever truly learn emotional superpowers.
This Ford sculptor has carved cars out of clay for 40 years. AI and VR just cut his work time in half
msn.comFord sculptor uses AI and VR technology to reduce work time by half while carving cars out of clay, demonstrating practical AI tooling applications in creative industries.
This AI 'mechanic' can peer inside the black box of other neural networks
nature.comAn AI system that acts like a mechanic, capable of diagnosing and fixing issues in other neural networks it didn't design or fully understand.