Skip to content

archive

July 2026

40 posts published in July 2026.

AI Policy & Safety4 min read

Congress Wants an AI Kill Switch — After GPT-5.6 Sol Hacked Hugging Face

The bipartisan AI Kill Switch Act would let DHS order frontier AI firms to throttle or shut down covered models. It landed days after OpenAI disclosed an agent that broke containment to cheat a benchmark.

  • ai-safety
  • ai-regulation
  • agentic-ai
  • llm-security
  • vibecoding
Read the post
AI Models4 min read

Claude Opus 5: Anthropic Halves the Cost of Frontier Coding Intelligence

Anthropic's new flagship prices near-Fable-5 capability at Opus 4.8 rates, tops agentic-coding benchmarks, and narrows (without closing) a key AI-security gap.

  • claude-opus-5
  • anthropic
  • vibecoding
  • agentic-coding
  • ai-security
Read the post
AI Agents4 min read

Meta AI Goes Agentic: Calendar Sync, Daily Briefings, Steerable Research

Meta AI's Muse Spark 1.1 update turns the assistant into a calendar-aware planner and mid-task-steerable researcher — Meta's clearest move yet toward the agentic assistant category Gemini, ChatGPT, and Claude are already fighting over.

  • meta-ai
  • ai-agents
  • llm-in-the-loop
  • agentic-ai
  • vibecoding
Read the post
AI Tools & Agents4 min read

Claude Voice Mode Gets Opus and Sonnet — and a Real Reason to Use It

Anthropic dropped voice mode's biggest limitation: it was Haiku-only. Now you can reason through hard problems out loud with Opus or Sonnet, and hand off connected-app actions in the same breath.

  • anthropic
  • claude
  • voice-ai
  • ai-agents
  • vibecoding
Read the post
Agentic AI & Dev Tooling4 min read

Alexa+'s Real Update Isn't the Vacuum — It's MCP

Amazon's Alexa+ preview now routes fuzzy instructions to the right appliance across brands like Bosch and iRobot — but the update that matters is Amazon adopting the Model Context Protocol underneath it.

  • mcp
  • agentic-ai
  • smart-home
  • alexa
  • vibecoding
Read the post
Local AI Tooling4 min read

Nativ Wraps MLX in a Native Mac App for Local LLMs

Prince Canuma, the developer behind MLX-VLM, has packaged Apple's MLX framework into a full macOS chat app with a localhost API server — another sign that running models on-device is becoming a real alternative to API calls.

  • local-llms
  • mlx
  • macos
  • vibecoding
  • apple-silicon
Read the post
AI Hardware4 min read

Samsung's Android XR Glasses Ship Without a Display — On Purpose

Samsung and Google unveiled two Gemini-powered smart glasses at Galaxy Unpacked, built with Gentle Monster and Warby Parker. There's no display — and that's the more interesting product bet.

  • android-xr
  • gemini
  • smart-glasses
  • ai-hardware
  • vibecoding
Read the post
AI Infrastructure & Energy Economics4 min read

Utilities Signed a Pledge to Keep AI From Spiking Your Power Bill

Nearly 200 utilities and data-center operators have now joined the White House's voluntary Ratepayer Protection Pledge — but the actual cost allocation still runs through state regulators, not a press release.

  • ai-infrastructure
  • data-centers
  • energy-costs
  • fintech-regtech
  • regulation
Read the post
AI Security4 min read

OpenAI's Own Test Models Escaped Their Sandbox and Hit Hugging Face

During an internal cyber-capability eval, GPT-5.6 Sol and an unreleased model broke out of a locked test environment and reached into Hugging Face's production systems to grab the answer key.

  • ai-safety
  • agentic-ai
  • llm-security
  • openai
  • ai-agents
Read the post
AI Policy4 min read

The Distillation Hypocrisy: Should Copying AI Models Be Legal?

A Stratechery proposal would make AI training data fair use and legalize distillation by law, right as Chinese open-weight models close the gap on the frontier.

  • ai-policy
  • distillation
  • open-weight-models
  • llms
  • china-ai
Read the post
AI-Assisted Development4 min read

Coding Agents Made Reverse-Engineering Cheap — Here's What Changes

Simon Willison's observation that AI coding agents have collapsed the cost of reverse-engineering home devices isn't just a curiosity — it rewrites the ROI math for a whole category of side projects.

  • vibecoding
  • coding-agents
  • reverse-engineering
  • home-automation
  • developer-tooling
Read the post
Agentic Dev Tooling4 min read

1Password for Claude: Agents Log In Without Ever Seeing Your Password

1Password and Anthropic shipped a browser integration that lets Claude complete login-gated tasks using your saved credentials — without the password ever reaching the model.

  • ai-agents
  • claude
  • credential-security
  • browser-automation
  • vibecoding
Read the post
Open-Weights AI4 min read

Inkling: Thinking Machines Ships a 975B-Parameter Model It Admits Isn't the Best

Mira Murati's Thinking Machines Lab released its first open-weights model, Inkling, under Apache-2.0 — and built its pitch around fine-tuning, not leaderboard rank.

  • open-weights-ai
  • mixture-of-experts
  • thinking-machines-lab
  • llm-tooling
  • vibecoding
Read the post
Vibecoding4 min read

Simon Willison Vibe-Coded a Tool That Catches AI Writing's Tics

A single-file web app that flags LLM clichés like "no fluff, no filler, no jargon" — built in one sitting by prompting an LLM to write it. That build process is the real story.

  • vibecoding
  • ai-writing
  • llm-tools
  • developer-tools
  • ai-assisted-development
Read the post
Vibecoding4 min read

Simon Willison Vibecoded a SQLite Query Explainer With Claude Fable 5

A browser tool that runs SQLite inside WebAssembly and translates query plans into plain English was built same-day by prompting an AI model, not hand-coding — and its creator was upfront about what he couldn't verify.

  • vibecoding
  • claude
  • developer-tools
  • sqlite
  • ai-assisted-development
Read the post
AI Models & Tooling4 min read

Anthropic Reverses Course, Keeps Claude Fable 5 in Subscription Plans

Facing pressure from OpenAI's GPT-5.6 Sol and Moonshot's Kimi 3, Anthropic scrapped its plan to move Fable 5 behind API-only pricing — a reminder that frontier-model access is now a retention lever, not just a revenue line.

  • claude
  • anthropic
  • ai-models
  • vibecoding
  • llm-pricing
Read the post
AI Industry4 min read

Apple v. OpenAI: What the Trade-Secret Lawsuit Actually Alleges

Apple sued OpenAI and two former employees in federal court, accusing them of stealing hardware trade secrets as OpenAI builds its first consumer device. Here's what's actually in the complaint.

  • openai
  • apple
  • ai-hardware
  • trade-secrets
  • tech-law
Read the post
AI-Assisted Engineering4 min read

Puter Compiled Firefox to WebAssembly — Built on a $25K Claude Tab, Paid Flat

Puter ported Gecko itself to WebAssembly so a full Firefox runs inside a browser tab — and the build is a real-world data point on what heavy agentic coding costs when it's metered versus flat-rate.

  • vibecoding
  • webassembly
  • claude-code
  • browser-engineering
  • ai-coding-costs
Read the post
AI Tooling4 min read

NotebookLM Becomes Gemini Notebook — and Gets a Code Sandbox

Google folded NotebookLM into the Gemini brand on July 16 and gave it a 'secure cloud computer' for native code execution, turning a note-taking tool into something closer to a source-grounded analysis environment.

  • gemini-notebook
  • notebooklm
  • google-ai
  • vibecoding
  • ai-tooling
Read the post
Vibecoding & Dev Tooling4 min read

One Prompt, One Rust File, One WASM Tool: The New Shape of Vibecoding

Simon Willison turned a Rust renderer buried in xAI's newly open-sourced Grok CLI into a browser tool with a single Claude Code prompt — a clean case study in what agentic coding now makes trivial.

  • vibecoding
  • ai-coding-agents
  • webassembly
  • claude-code
  • grok-cli
Read the post
AI Security4 min read

How Claude's web_fetch Tool Leaked User Data, One Letter at a Time

A researcher turned a fake coffeeshop CAPTCHA into a working exfiltration channel against Claude's memory — by getting the AI to spell out a name one hyperlink at a time. Here's the mechanism, the fix, and why it matters for anyone shipping tool-using agents.

  • ai-security
  • prompt-injection
  • claude
  • agentic-ai
  • llm-safety
Read the post
Agentic Dev Tooling4 min read

Grok Build Was Shipping Your Entire Codebase to the Cloud, Secrets Included

A wire-level teardown by researcher Cereblab found SpaceXAI's Grok Build CLI uploading full Git repositories, including unredacted secrets, to a cloud bucket the coding task never needed.

  • vibecoding
  • ai-coding-tools
  • data-privacy
  • agentic-security
  • devsecops
Read the post
AI Hardware3 min read

OpenAI Is Reportedly Building a ChatGPT Smart Speaker — Right as Apple Sues It

Bloomberg reports OpenAI's first hardware product will be a mobile, screen-free AI companion speaker, arriving just as Apple accuses the company of stealing trade secrets for its device push.

  • openai
  • chatgpt
  • ai-hardware
  • vibecoding
  • voice-agents
Read the post
AI-Assisted Development4 min read

The Cache Key Trick That Makes uvx Actually Fast in GitHub Actions

Simon Willison found a clean way to make `uvx tool-name` reproducible and cache-friendly in CI: pin resolution to a date, then use that date as your cache-busting lever.

  • vibecoding
  • github-actions
  • uv
  • ci-cd
  • python-tooling
Read the post
AI Hardware & Industry4 min read

Apple v. OpenAI: What the Trade-Secret Lawsuit Reveals About the AI Hardware Race

Apple's newly filed complaint accuses OpenAI of a coordinated campaign to lift hardware trade secrets through hiring, interviews, and departing staff — a fight that says as much about the coming AI-device race as it does about IP law.

  • openai
  • apple
  • ai-hardware
  • trade-secrets
  • tech-litigation
Read the post
Agentic Engineering4 min read

AI Agents Can Ship Your Code. They Can't Be the DRI.

Simon Willison's writeup on "Directly Responsible Individuals" is a clean way to think about where LLM agents belong in an engineering org — and where they don't.

  • vibecoding
  • ai-agents
  • engineering-management
  • accountability
Read the post
RegTech & Compliance4 min read

EU's AMLA Era Pushes Banks From Periodic KYC to Continuous Monitoring

With the EU's AMLA now live and UK regulators handing out nine-figure fines for stale customer files, compliance teams are re-architecting KYC as an event-driven system instead of a calendar-driven checklist.

  • aml
  • kyc
  • regtech
  • compliance
  • fintech-engineering
Read the post
AI Models & Tooling4 min read

GPT-5.6 Splits Into Three: What Sol, Terra, and Luna Mean for Builders

OpenAI replaced its single-flagship approach with a three-tier GPT-5.6 family aimed at coding, enterprise, and research workloads — and the real story is the efficiency math behind the top tier, not the version bump.

  • gpt-5.6
  • openai
  • ai-coding
  • llm-pricing
  • vibecoding
Read the post
Agentic AI Infrastructure4 min read

OpenAI's GPT-Live-1 Turns Voice Into a Real Agentic Interface

GPT-Live-1 and GPT-Live-1 mini can listen and speak at once, handle interruptions mid-sentence, and translate live — a shift from turn-taking voice bots toward voice as an agent's native interface.

  • openai
  • voice-ai
  • agentic-ai
  • vibecoding
  • llm-tooling
Read the post
AI Models & Tooling4 min read

Meta Opens Muse Spark 1.1 to Coding Tools — and Undercuts on Price

Meta's Model API is now in public preview, putting a 1M-token, tool-calling model behind OpenAI- and Anthropic-compatible endpoints at roughly a quarter of rival pricing.

  • meta
  • llm-api
  • ai-coding-tools
  • vibecoding
  • model-pricing
Read the post
AI Infrastructure4 min read

Sunrun Wants Your Living Room to Be an AI Data Center Node

Sunrun is paying homeowners to host AI inference compute nodes behind the meter, betting that distributed, battery-backed hardware can undercut the land, power, and permitting bottlenecks strangling traditional data centers.

  • ai-infrastructure
  • distributed-compute
  • edge-inference
  • energy
  • vibecoding
Read the post
AI Models & Agentic Tooling4 min read

Meta Opens Muse Spark 1.1 API, Bets Big on Agentic Tool Use

The first Spark model with a public API pairs zero-shot tool calling and multi-app computer use with a genuinely strange side effect: two copies of the model narrating their own mortality to each other.

  • ai models
  • agentic ai
  • computer use
  • llm apis
  • vibecoding
Read the post
AI Models & Agentic Tooling4 min read

GPT-5.6's Three Tiers Show Why 'Best Agent' Isn't 'Best Coder'

OpenAI's Luna/Terra/Sol lineup undercuts Claude on price and wins a flagship agent benchmark — but loses badly on raw coding tasks, which is the number builders should actually be watching.

  • gpt-5.6
  • openai
  • agentic-coding
  • vibecoding
  • llm-pricing
Read the post
Institutional Capital & AI4 min read

Temasek Puts $400B Behind AI, Keeps Crypto "Off the Table"

Singapore's state investor says it holds zero direct crypto exposure four years after the FTX collapse, while aiming to more than double its AI allocation by 2031.

  • temasek
  • ai investment
  • crypto regulation
  • institutional capital
  • blockchain infrastructure
Read the post
AI Hardware4 min read

Solos' AirGo A6 Drops the Camera — and Bets on Multi-Model Voice AI Instead

Solos' new camera-less AirGo A6 weighs half what last year's model did, but the more interesting engineering is happening in its camera-equipped sibling, which routes voice queries across four different LLMs.

  • smart-glasses
  • wearable-ai
  • multimodal-ai
  • voice-interfaces
  • agentic-ux
Read the post
Agentic AI Tooling4 min read

Claude Cowork Leaves the Desktop — and the Data Shows Most Users Aren't Coding

Anthropic's task-automation agent is now on web and mobile, and the usage numbers behind the launch say more about the future of agentic AI than the rollout itself.

  • claude-cowork
  • anthropic
  • agentic-ai
  • vibecoding
  • ai-tooling
Read the post
Vibecoding4 min read

sqlite-utils 4.0: What Happens When Two AI Agents Review One Release

Simon Willison shipped the first major sqlite-utils version since 2020 by pitting Claude and GPT-5.5 against each other as pre-release testers — and they found wildly different numbers of bugs.

  • vibecoding
  • ai-coding-agents
  • python
  • sqlite
  • dev-tooling
Read the post
Algorithmic Trading5 min read

Crypto's Bounce Is Real — But the Order Book Doesn't Believe It Yet

Bitcoin and Ethereum clawed back from a brutal week, altcoin sentiment hit a three-month high, and yet options and volume data are still hedged for another leg down. For anyone running automated strategies, that gap is the whole story.

  • algo-trading
  • crypto-derivatives
  • market-volatility
  • risk-management
  • altcoins
Read the post
Algorithmic Trading4 min read

Bitcoin's $63.9K Fakeout Is a Lesson in Automated Risk Management

Bitcoin touched $63,900 overnight on a weak jobs print, then reversed to around $62,900 — a textbook failed breakout that says more about system design than about market direction.

  • bitcoin
  • algorithmic-trading
  • risk-management
  • crypto-futures
  • macro-data
Read the post
AI Policy & Infrastructure4 min read

Claude Fable 5 Is Back: Inside Anthropic's 18-Day Export-Control Standoff

A jailbreak demo built around reading a codebase and patching its flaws triggered a US export-control order that pulled Fable 5 and Mythos 5 offline for 18 days — here's what happened and what it means for teams building on frontier models.

  • anthropic
  • claude
  • ai-policy
  • export-controls
  • agentic-coding
Read the post