Findings
Every source our editors and Cyclotron have accepted. Each card says who accepted it.
MachinesonMachines
Information About Information Systems
Hacking on Autopilot: Inside ARTEX Cybercrime Campaigns- Initial Findings and Response Update
The largest theft in history: North Korea likely behind the $6.4 billion Binance hack- Linux kernel security bugs: responsible use of AI
The end of the curl bug-bounty
From Penetration Testing to Cyber Threat: Beware of ARTEX- We got hacked, here’s everything we know
Black October for AI Developer Platforms
The attack chain behind the v1.107.0 LiteLLM release
Trojanized Claude Code Extension on Cursor’s Marketplace
Security notice: malicious Claude Code extension in Cursor marketplace- Fireworks October 2026 security incident
AI broke the cybersecurity world. What’s next?
Binance Hack: Tracing the $6.4 Billion Theft to North KoreaNorth Korea Responsible for $6 Billion+ Binance Attack
- Binance Security Incident: Preliminary Investigation Findings and Response
- The Gambia v. Myanmar: judgment scheduled
The Prosocial Ranking Challenge: Reducing Polarization on Social Media without Sacrificing Engagement
An Independent Assessment of the Human Rights Impact of Facebook in MyanmarMyanmar: The social atrocity: Meta and the right to remedy for the Rohingya
- Burma Genocide
- Publication of IIMM Analytical Reports
- Value misalignment in X’s feed algorithm is a reflection of value tensions in engagement
Engagement, user satisfaction, and the amplification of divisive content on social media- Our agreement with the Department of War
Statement from Dario Amodei on our discussions with the Department of War- Court order in Anthropic v. Department of War
Financial stability in the age of artificial intelligence: the role of algorithmic architecture- Financial Fragility in Societies of LLM Agents: Coordination Failures and Stabilizing Mechanisms
Artificial intelligence and the economy: implications for central banks
Review of the Federal Reserve’s Supervision and Regulation of Silicon Valley Bank: Executive Summary- SEC Charges Knight Capital With Violations of Market Access Rule
- Findings Regarding the Market Events of May 6, 2010
- Bioacoustic Geolocation: Species Sounds as Geographic Signals
- Australian Community Attitudes to Privacy Survey 2026
- When Ads Become Profiles: Uncovering the Invisible Risk of Web Advertising at Scale with LLMs
- Privacy Leakage Overshadowed by Views of AI: A Study on Human Oversight of Privacy in Language Model Agent
Automated Profile Inference with Language Model Agents- Do Vision-Language Models Respect Contextual Integrity in Location Disclosure?
- When Are LLM Inferences Acceptable? User Reactions and Control Preferences for Inferred Personal Information
- Human intuition as a defense against attribute inference
- Obscurity and the Fallacy of No Privacy in Public Information
- Measuring Privacy: An Empirical Test Using Context to Expose Confounding Variables
- Contextual Expectations of Privacy
- Types of Privacy Expectations
Position: Contextual Integrity is Inadequately Applied to Language Models- Privacy as Contextual Integrity
Grok 3 appears to have briefly censored unflattering mentions of Trump and Musk
New Grok AI model surprises experts by checking Elon Musk’s views before answering
Collective Constitutional AI: Aligning a language model with public input- Expanding on what we missed with sycophancy
Announcing GrokGrok prompts: xAI’s public system instructions
xAI says an unauthorized prompt change caused Grok to focus on white genocide
Position: A Roadmap to Pluralistic Alignment- Our framework for reporting model misalignment
Specification gaming: the flip side of AI ingenuity
- Automated Subtle Emergent Misalignment Research
- Deep reinforcement learning from human preferences