AHTV | Augmented Human TV

13 dispatches

Apple Intelligence: When AI Becomes an Operating-System Feature (Jun 2024)

On June 10, 2024 (WWDC), Apple announced Apple Intelligence—not as an app, not as a chatbot you visit, but as a system layer built into iOS 18, iPadOS 18, and macOS Sequoia. When AI is 'a tool,' you choose when to use it. When AI is 'part of the OS,' it quietly starts choosing *with you*. The question is simple: who are you when the default version of you is 'AI-assisted'?

11 minJun 10, 2024

AI & Voicemedium

GPT-4o and the Real-Time Voice Interface — When AI Stops Feeling Like Software

OpenAI's GPT-4o achieved 232ms response latency—basically human conversation speed. When AI responds in real time, it stops feeling like software and starts feeling like presence. The ethical stakes: voice carries emotion, persuasion, and trust. Speed becomes a form of influence.

5mMay 13

AI Wearableshigh

When Your Glasses Can 'See' — Meta AI With Vision on Ray-Ban (Apr 2024)

Meta rolled out AI with Vision to Ray-Ban smart glasses in April 2024—turning them from voice assistants into multimodal conversational devices that see and answer questions about your surroundings. The shift from 'searching' to 'asking your surroundings' is frictionless, but it raises a critical question: when glasses become a contextual AI platform, who protects the privacy of everyone else in the room?

9mApr 23

AI Wearablesmedium

Humane Ai Pin Ships — The Phone-Free Future Meets Reality

On April 11, 2024, Humane's Ai Pin started shipping—a $699 'screenless' AI computer promising a phone-free future. But the moment it arrived, the gap between vision and daily usability became undeniable. The real lesson: ambient computing can't ship like a beta, and presence comes from certainty, not just removing screens.

9mApr 11

Policy & Governancehigh

The EU AI Act: A 'GDPR Moment' for AI (and Why It Changes What Tech Can Get Away With)

On March 13, 2024, the EU AI Act approved landmark AI regulation that treats 'trust' as enforceable. When AI moves from ethics guidelines to binding law, everything changes: what products ship, what business models survive, and what harms become illegal instead of 'oops.'

14mMar 13

Roboticshigh risk

Figure + OpenAI: Humanoids Get a Brain (and Society Gets a New Liability)

Figure raised $675M and partnered with OpenAI to power humanoid robots with language models. The moment robots can reason is when mistakes become kinetic, accountability blurs, and 'labor shortage' becomes cover for control. What we demand now determines what we tolerate later.

14 minFeb 29, 2024

Generative AIhigh

OpenAI Sora: When 'Video' Becomes a Prompt

OpenAI previewed Sora, a text-to-video model that generates up to 60 seconds of video from a prompt. The real risk isn't creation—it's credibility. When anyone can generate believable video on demand, 'seeing is believing' stops being a default. The question: can society still trust what it sees?

14mFeb 15

Spatial Computinghigh

Apple Vision Pro Launch: When Your Eyes Become the Interface

On February 2, 2024, Apple Vision Pro arrived in stores. It's not just hardware—it's a new social situation where cameras on your face become the default, eye tracking normalizes attention as input, and bystanders never opted in. The ethical question: does the product protect the people who didn't buy it?

13mFeb 2

AI Agentshigh

Rabbit R1: The "Do It For Me" Device and the Price of Delegation

In January 2024, rabbit launched the r1—a $199 device built around a Large Action Model (LAM) that promises to use your apps for you. It's the shift from 'AI that answers' to 'AI that acts.' But when AI can do things as you, delegation becomes a security event. The real danger isn't hallucination—it's invisible action.

14mJan 9

Gene Therapyhigh

The First CRISPR Medicine Is Approved. The Hard Part Is Who Gets It.

The FDA approved CASGEVY and LYFGENIA—the first CRISPR-edited therapies for sickle cell disease. This is a historic win. But a cure is not just a molecule. A cure is an infrastructure event. If we treat this as a finish line, we'll get a world where the cure exists and the disease still destroys lives because access is rationed.

14mDec 8

Wearable Techhigh

Ray-Ban Meta Made Recording Look Like a Face. That's the Ethical Problem.

Meta's new Ray-Ban smart glasses can record from your point of view and livestream instantly. The problem isn't the camera—it's that when recording looks like eyewear, bystanders lose the ability to tell when they're being filmed. That changes everything about consent, autonomy, and what it means to be in public.

14mSep 27

Spatial Computinghigh risk

Vision Pro's Real Innovation Is Making Your Eyes a Mouse

Apple unveiled Vision Pro and called it a 'spatial computer.' The real shift isn't the display—it's the interface. When your gaze becomes the cursor, attention stops being private behavior and starts becoming a potential dataset. The question: will these systems respect attention as a human boundary, or treat it as a resource to mine?

12 minJun 5, 2023

Neurotechhigh

Neuralink Got FDA Clearance. That's Not a Victory Lap.

FDA clearance for Neuralink's clinical trial is regulatory permission to learn—not proof of safety. The ethical question is simple: who carries the uncertainty, who gets the upside, and what do we owe the people who volunteer first?

10mMay 20

Command Palette

Apple Intelligence: When AI Becomes an Operating-System Feature (Jun 2024)

GPT-4o and the Real-Time Voice Interface — When AI Stops Feeling Like Software

When Your Glasses Can 'See' — Meta AI With Vision on Ray-Ban (Apr 2024)

Humane Ai Pin Ships — The Phone-Free Future Meets Reality

The EU AI Act: A 'GDPR Moment' for AI (and Why It Changes What Tech Can Get Away With)

Figure + OpenAI: Humanoids Get a Brain (and Society Gets a New Liability)

OpenAI Sora: When 'Video' Becomes a Prompt

Apple Vision Pro Launch: When Your Eyes Become the Interface

Rabbit R1: The "Do It For Me" Device and the Price of Delegation

The First CRISPR Medicine Is Approved. The Hard Part Is Who Gets It.

Ray-Ban Meta Made Recording Look Like a Face. That's the Ethical Problem.

Vision Pro's Real Innovation Is Making Your Eyes a Mouse

Neuralink Got FDA Clearance. That's Not a Victory Lap.

Command Palette

Apple Intelligence: When AI Becomes an Operating-System Feature (Jun 2024)

GPT-4o and the Real-Time Voice Interface — When AI Stops Feeling Like Software

When Your Glasses Can 'See' — Meta AI With Vision on Ray-Ban (Apr 2024)

Humane Ai Pin Ships — The Phone-Free Future Meets Reality

The EU AI Act: A 'GDPR Moment' for AI (and Why It Changes What Tech Can Get Away With)

Figure + OpenAI: Humanoids Get a Brain (and Society Gets a New Liability)

OpenAI Sora: When 'Video' Becomes a Prompt

Apple Vision Pro Launch: When Your Eyes Become the Interface

Rabbit R1: The "Do It For Me" Device and the Price of Delegation

The First CRISPR Medicine Is Approved. The Hard Part Is Who Gets It.

Ray-Ban Meta Made Recording Look Like a Face. That's the Ethical Problem.

Vision Pro's Real Innovation Is Making Your Eyes a Mouse

Neuralink Got FDA Clearance. That's Not a Victory Lap.