ChatGPT gets physical

Plus: Elon goes to court

January 21, 2026

Welcome back! While tech companies race to embed AI into every corner of daily life, a quiet rebellion is underway. People are buying landline phones, knitting, writing letters, and swapping smartphones for film cameras. They're calling it “going analog.” In fact, searches for “analog hobbies” jumped 136% in the past six months, according to craft store Michael’s. 

The irony? Many are discovering the analog lifestyle through…TikTok. 🙃

OpenAI's First Device Arrives This Year

Photo: Dani Ammann Photography for Axios

OpenAI is “on track” to unveil its first physical device in the second half of this year, according to Chief Global Affairs Officer Chris Lehane. The announcement, made at Axios House Davos this week, marks the first concrete timeline for a product CEO Sam Altman has teased since acquiring former Apple design chief Jony Ive's company last May.

What we know (and don't) ➡️ OpenAI hasn't revealed what the device will look like or what it will do. Given the involvement of Ive—the designer behind the iPhone, iPad, and Apple Watch—expectations are high for something that reimagines how people interact with AI beyond screens and keyboards. The acquisition of Ive's firm signaled OpenAI's ambition to move beyond software, but details have been scarce.

The hardware shift: This marks a clear pivot for AI companies that have spent years building models to run on other people's infrastructure—phones, laptops, and browsers controlled by Apple, Microsoft, and Samsung. A dedicated AI device would give OpenAI direct control over the user experience.

Musk Demands $134B From OpenAI

Elon Musk is seeking up to $134 billion in damages from OpenAI and Microsoft, claiming the AI company defrauded him by abandoning its nonprofit mission. The figure comes from an expert witness who calculated that Musk deserves a hefty portion of OpenAI's $500 billion valuation based on his $38 million seed donation in 2015—a 3,500-fold return on investment.

The lawsuit's core claims:

  • The case centers on OpenAI's shift from its original nonprofit structure to a for-profit model, which Musk claims betrayed the company's founding mission.

  • Musk's legal team argues he should be compensated as an early startup investor, factoring in both his financial contributions and the technical and business expertise he provided to OpenAI's early team.

  • The expert analysis calculates wrongful gains of $65.5 billion to $109.4 billion for OpenAI and $13.3 billion to $25.1 billion for Microsoft, which owns 27% of the company.

Just saying: Musk's personal fortune currently sits around $700 billion, making him the world's richest person by a $500 billion margin over Google co-founder Larry Page.

The bigger picture: This is the latest chapter in Musk's ongoing feud with OpenAI, the company he co-founded and later left. OpenAI has characterized the lawsuit as part of an “ongoing pattern of harassment” rather than a legitimate financial grievance. It seems that, for Musk, this isn't about the money.

A Three-Month-Old Startup Just Raised $480M at a $4.5B Valuation

Humans&, a startup that's barely three months old, just closed a $480 million seed round at a $4.48 billion valuation. The company, founded by alumni from Anthropic, xAI, and Google, is betting that AI should empower people rather than replace them. And investors—including Nvidia, Jeff Bezos, Google Ventures, and Laurene Powell Jobs's Emerson Collective—are buying in.

What they're building: Humans& is developing software to help people collaborate with each other—think an AI-powered instant messaging app, but designed to strengthen teams and communities rather than automate them away. The company plans to train AI in new ways, like programming chatbots to request information from users and store it for later use, with a focus on long-term memory and multi-agent coordination.

Who's behind it: The founding team includes Andi Peng, a former Anthropic researcher who worked on Claude 3.5 through 4.5; Georges Harik, Google's seventh employee who helped build its first advertising systems; and Eric Zelikman and Yuchen He, two former xAI researchers who developed the Grok chatbot. The 20-person team also includes talent from OpenAI, Meta, MIT, and Stanford.

Why it matters: The pitch is always slightly different—more human-centric, more collaborative, more ethical—but the playbook is the same: assemble a team of recognizable names from top labs, raise an eye-watering valuation, and pledge to build AI differently. Whether Humans& can deliver on that promise remains to be seen.

The internet, on alert

Ayewatch

AyeWatch uses a multi-agent system to scan more than a billion sources across news, social, research, and the web. It filters out noise and duplicates to deliver real-time, rule-based alerts when meaningful changes or matches occur. 

How you can use it:

  • Track competitors, funding rounds, or product launches the moment they surface

  • Monitor specific people, companies, or topics across media and social platforms

  • Set research alerts for academic papers or regulatory updates

  • Build custom signals for investment, PR, or market intelligence workflows

Pricing: Free and paid plans

Pick the right model

WhatLLM.org

WhatLLM.org benchmarks 100+ large language models using a standardized AI index that ranks them by price, performance, speed, and context window, updated daily.

How you can use it:

  • Compare models for coding, reasoning, or long-context workloads

  • Optimize cost vs. performance before shipping a feature or product

  • Track pricing and performance changes across providers in real time

  • Shortlist models for production without manual testing

Pricing: Free

Jobs, announcements, and big ideas

  • Google’s AI chief says it has no plans for Gemini with ads.

  • OpenAI adds age prediction to ChatGPT as a new safety layer for teen users.

  • Adobe brings AI directly into Premiere and After Effects with new video and motion design tools.

  • Anthropic launches a global AI training program for educators with Teach For All.

  • Groq’s CEO says AI will drive deflation and create more jobs than it replaces.

  • China blocks Nvidia’s H200 AI chips despite US clearance, forcing suppliers to halt production.

  • Google adds its Flow AI video tool to Workspace with expanded admin controls.

Claude takes the keyboard: I break down Anthropic’s new Cowork feature that lets Claude work directly inside your files.

That’s a wrap! See you Friday for more.

—Matt (FutureTools.io)