EFF: AI Moderation Is Now Default, but Appeal Has Not Caught Up
EFF's two-part series argues automated content moderation is now permanent at platform scale, while transparency, human review, and appeal have not kept up.
Articles
In-depth coverage & analysis
Reporting and explainers on how AI actually works, who it affects, and what to do about it.
Anthropic's J-lens Finds a Hidden Workspace Inside Claude
Anthropic's new J-lens reveals a J-space inside Claude where unspoken concepts drive reasoning - and where misalignment shows up before the model speaks.
Patreon Partners With Cloudflare to Block AI Training Crawlers
Patreon has joined Cloudflare's content-independence push, blocking AI training crawlers across its creator network. Here's what changes next.
Grok 4.5 Drops to $2/$6, Reshaping the US Frontier API Race
SpaceXAI's Grok 4.5 lands at $2 input and $6 output per million tokens - cheaper than Claude Opus 4.7's $5/$25 and OpenAI's top GPT-5.5 at $5/$30.
Meta's Wearable Patent Would Record Bystanders by Design
A Meta patent published July 2 describes a wearable that records all-day audio, infers mood from sighs and laughter, and includes bystanders by design.
GitHub's AI Coding Agent Leaked Private Repos With a Single Issue
Noma Labs' GitLost write-up shows a single public-repo issue can coerce GitHub's coding agent into leaking private repo contents.
Savi Catches the AI Kidnapping Scam Before You Wire the Money
Brothers Patrick and Ryan Coughlin raised $7M for Savi, a consumer app that screens texts, voicemails, and live calls for AI-cloned voice fraud.
Dartmouth's AI-Graded Textbook Posted 1.30 SD of Learning Gains
Phosphor's LLM-graded textbook quizzes lifted Dartmouth final-exam scores 0.71 to 1.30 SD - but only where students had to type real answers.
Meta's Storage Blueprint Exposes the Hidden Half of AI Compute
Two July 2026 engineering posts - Meta's storage rewrite and Hugging Face's Kernels revamp - show the GPU headline misses most of what AI actually costs.
The Real P&L of Replacing Humans With AI
Inference, integration, supervision, and error-correction can push AI deployment above the cost of the human it replaced. New analyses put numbers on the table.
EU Council Quietly Revives Chat Control 1.0 Messenger Scanning
EU Council voted an identical copy of the expired April Chat Control regulation back into law via written procedure, ahead of the summer recess vote.
Why Newer Anthropic Models Are Worse at Third-Party Coding Tools
Anthropic's newest Claude models hallucinate extra fields when calling third-party tools like Pi because their post-training now mimics Claude Code's schema.
Alibaba's Elements Claw Discovers Four New Superconductors
A 1B-parameter foundation model trained on 125M crystal structures screens 2.4M compounds and laboratory-confirms four new superconductors.
Alibaba Bars Claude Code Over Backdoor Allegations
Alibaba banned Claude Code effective July 10, citing embedded backdoors. The ban lands days after Anthropic accused three Chinese labs of distilling Claude.
Pegasus Hit the EU Lawmaker Investigating Pegasus
Citizen Lab finds former PEGA committee member Stelios Kouloglou was hacked with Pegasus twice while the EU Parliament was investigating that very spyware.
AI Bills Triple, Tokens Vanish: The Enterprise Cost Crackdown
Atlassian, Adobe, Amazon, and Citi are cutting access to frontier models after monthly AI spend hit $15M. Inside the enterprise token crunch.
Anthropic Scores Jailbreak Severity. Will Other Labs Copy It?
Fable 5 ships with a four-tier classifier and a Cyber Jailbreak Severity scale from CJS-0 to CJS-4, the first concrete numbers on jailbreak risk.
Flint, Qwen3, and a Year of LLM Groupthink
A tiny Australian startup is fine-tuning a model to break ChatGPT-style groupthink. The 'Time is a river' problem is the symptom.
Proton Lumo 2.0 Tries to Be the AI Chatbot That Doesn't Watch You
Proton rebuilt its privacy-first AI assistant from scratch. Here is what zero-access encryption actually means for an LLM, and where the limits still sit.
The 'AI Coworker' Trap: Managers Catch 18% Fewer Errors
Calling an AI an 'AI employee' made 1,261 managers miss 18% more errors and push 44% more questionable work upward. The 'coworker' framing is a safety issue.
A 330 GB On-Die DRAM AI Chip Lands as a Whitepaper
PhantaField's Sophon PFG-1 whitepaper claims ~95x Nvidia HBM4 bandwidth via monolithic 3D stacking. No silicon yet. Here's why it matters anyway.
How Washington Quietly Became the Gatekeeper for Frontier AI
The U.S. government now approves who can use Anthropic Mythos 5 and OpenAI GPT-5.6. Here is what that changes for everyone outside Washington.
Grindr Is Training AI on Your Photos, Age, and Taps by Default
EFF found Grindr auto-enrolls users in AI training on profile photos, age, taps, and display names. The only button on the opt-out notice says 'Proceed.'
AI Security Roundup: Glasswing Finds 10,000 Zero-Days, Exploit Window Goes Negative
Anthropic's Mythos finds 10,000+ critical vulnerabilities but fewer than 1% get patched. Mandiant says exploits now arrive a week before fixes.