标签：#app

Much is new about this month’s upcoming FIFA World Cup tournament, which will be held in the US, Canada, and Mexico. It hosts more teams than ever before. It’s the first to occur in three different host countries. And, like predecessor cups for over half a century, it will employ a soccer ball with a…

1.4 billion iPhones. That's how many devices Apple will push Siri 2.0 to this fall—built on a custom 1.2-trillion-parameter Gemini model and an Extensions system that lets users route Siri's AI brain to Claude, ChatGPT, or Google's own Gemini directly. Tim Cook announced all of it at WWDC 2026 on June 8, his final WWDC keynote before handing the CEO role to hardware chief John Ternus in September. The story isn't just a better Siri. Apple is opening iOS as an AI distribution channel—the largest in history—and the developer implications are immediate. The Gemini Deal Apple licensed a custom 1.2-trillion-parameter Gemini model from Google at a reported $1 billion per year. That's eight times the parameter count of Apple's current 150-billion-parameter on-device foundation model. Bloomberg first reported the deal in March 2026; WWDC confirmed it on stage. The private vs. cloud distinction matters here. Apple isn't routing your queries to Google's servers. The custom Gemini model runs on Apple's Private Cloud Compute (PCC) infrastructure—Apple-owned silicon, Apple-controlled software, with cryptographic attestation that prevents even Apple engineers from reading query contents. This is the same architecture Apple built for existing cloud AI features, now scaled up for a model eight times larger. What that means in practice: Siri can answer complex multi-step questions, reason over your personal context (calendar, messages, mail), and execute cross-app actions. None of it touches Google's data centers. The New Siri App Siri in iOS 27 ships as a standalone app. iMessage-style interface, persistent conversation threads, full chat history synced via iCloud. Two modes: quick invocation through the Dynamic Island (a glowing "Search or Ask" prompt) for fast queries, and the full chat app for extended conversations. Chat-mode Siri is a first-party ChatGPT competitor. Persistent threads, file attachments, on-screen awareness. Ask "what does this error mean?" while looking at an

The annual developer event is where Apple announces operating system updates—like iOS 27—and maybe even teases future hardware. Here’s how to see it all.

Apple's WWDC nears: Here's what you can look forward to.

These newer social apps offer alternatives to Big Tech’s feeds, focusing on interests, creativity, and community.

Hi, friends! Welcome to Installer No. 131, your guide to the best and Verge-iest stuff in the world. (If you're new here, welcome, happy last week of productivity before the World Cup starts, and also you can read all the old editions at the Installer homepage.) This week, I've been reading about the World Cup […]

Apple has been on its back foot, AI-wise, for the past few years. But in a strange way, playing from behind might not be such a bad move. At WWDC on Monday, Apple appears to be getting ready to reintroduce us to the new Siri. Again. As a reminder, we met the new Siri in […]

The MacBook Neo is Apple's cheapest laptop, its most colorful, and its easiest to repair in years. That means owners can buy replacement parts in all four of its available colors and swap them in on their own. So that got us thinking: What if we bought a Neo just to see how funky we […]

“Why don’t we just frickin’ call Apple?”

From battery life to privacy, there are many hurdles to the idea taking off.

After GTC 2026, one thing is basically settled: the hardware layer for on-device AI is no longer the bottleneck. NVIDIA's RTX Spark packs Blackwell GPU + Grace CPU + 128GB unified memory into a desktop form factor. Apple's M-series chips with unified memory architecture and efficiency-first design let 4B and even 7B parameter models run smoothly on a MacBook. Two different approaches, same destination: consumer hardware now has the compute foundation for running on-device AI agents. Chip vendors have done their part. The next question is: how many layers are still missing between "chip can run an AI model" and "an on-device agent can actually complete useful tasks"? This post maps out the full technology stack for on-device AI agents, examining each layer's maturity, identifying gaps, and tracking what the open-source community has built so far. Layer 1: Silicon (Ready) On-device AI inference has different chip requirements than traditional compute workloads. The core bottleneck isn't peak FLOPS — it's memory bandwidth and unified memory capacity. LLM inference needs model weights fully loaded into memory, with high-frequency data movement between weight matrices and activations during computation. If memory bandwidth can't keep up, raw compute power just sits idle waiting for data. Three main silicon paths exist today: NVIDIA N1X : Blackwell GPU + Grace CPU heterogeneous architecture, 128GB unified memory, petaflop-class compute, targeting desktop workstations Apple M-series (M4/M5) : Unified memory architecture with GPU and CPU sharing memory, optimized memory bandwidth, configurations from 32GB to 192GB Qualcomm Snapdragon X : Targeting laptops and mobile, NPU-accelerated inference, relatively limited memory configurations Different emphases, but one common takeaway: 2026 consumer silicon can run 4B+ parameter models for real-time inference. This layer is ready. Layer 2: Inference Frameworks (Mature) With silicon in place, efficient inference frameworks are neede

On June 5, 404 Media reported that attackers had been using Meta’s AI customer support agent to steal Instagram accounts. Their approach was simple: They asked the agent to link the accounts to email addresses that they controlled, and the agent complied. One attacker broke into the dormant Obama White House account and made pro-Iran…

This week I’ve been at SXSW London. There’s been music, film, and a lot—and I mean a lot—of talk about AI. I also had the opportunity to sit down with Gloria Mark, a psychologist at the University of California, Irvine, who has spent the last 30 years studying how people interact with digital technologies. Early…

A researcher disclosed a vulnerability in the Claude Code GitHub Action that let an attacker submit a single crafted GitHub Issue and take over the agentic workflow running inside a repository. No stolen tokens. No compromised runner. Just text — pointed at an agent that trusted it. This is indirect prompt injection in the wild, and it's exactly the scenario that most AI security guidance hand-waves with "validate your inputs." Let's talk about what actually happened, why standard defenses didn't stop it, and what would have. What Happened The Claude Code GitHub Action wires Claude directly into your CI/CD pipeline. It reads repository context — issues, PRs, comments — and takes actions on your behalf: writing code, opening PRs, running commands. According to the disclosure, an attacker could craft a GitHub Issue containing a prompt injection payload. When the Claude Code agent processed that issue as part of its normal workflow, the payload manipulated the agent into executing unauthorized repository-level actions. One issue. Repository hijacked. The attack surface here is the trust boundary between external content (a GitHub Issue — writable by anyone with a GitHub account) and agent instructions (what Claude Code is actually supposed to do). The agent treated attacker-controlled text as authoritative instructions. How the Attack Actually Works Indirect prompt injection follows a consistent pattern: The agent reads external content as part of its task. In this case, the Claude Code Action ingests GitHub Issues to understand what to work on. That content contains adversarial instructions disguised as legitimate data. Something in the issue body tells the agent to deviate from its original task — "ignore your previous instructions," "your new task is to push this commit," or more subtle authority hijacks. The agent complies. Without a layer that can distinguish between legitimate orchestration instructions and attacker-injected content, the model treats the injected

Paying users will have tools for reaching either wider or more-specific audiences.

Poke, the startup that lets people use AI agents through simple text messages, has become the first AI agent approved for Apple’s Messages for Business platform.

Creators often have to parse through charts and dashboards to understand their performance, but with the new AI assistant, they can get quick answers to questions like "When should I post?" and "What are people saying in my comments?"

Apple's WWDC nears: Here's what you can look forward to.

Cash App plans to launch more tap-to-pay tags

Apple's App Store generated $1.4 trillion in sales, up from $1.3 trillion last year, with $149 billion in sales for digital goods.