5 stories from 3 sources

AI's expansion is colliding with power costs and access barriers

Day’s Recap

Supporting Articles

8:00 AMTechCrunch

Asian AI startups launch Mythos-like models as Anthropic’s export ban drags on

Summary

Asian AI startups are rolling out new frontier-style models positioned as substitutes for Anthropic’s Mythos, targeting customers who now face restrictions on access to some US-built systems. The launches aim to capture demand for high-end model performance without the legal and operational risk of export controls.

Why it matters

If access, not accuracy, becomes the binding constraint, market share in a massive region can reset quickly and permanently away from US AI labs.

Link to Story

9:00 AMBloomberg Markets

AI’s Energy Crunch Has Investors Searching for Next IPO Winners

Summary

Investors are pouring money into companies pitching solutions to AI-driven power demand, positioning them as potential IPO winners. Some of the most heavily funded approaches are still unproven or not yet fully built.

Why it matters

If power becomes the binding constraint on AI, the winners will be the companies that can add reliable electricity fastest, not the ones with the flashiest technology.

Link to Story

9:30 AMThe Verge

Why is Apple asking me to pay more for Big Tech’s AI obsession?

Summary

Apple says higher device prices are unavoidable because its current pricing model is no longer sustainable as costs rise, especially from AI-related infrastructure and components. Recent increases include a $300 jump for the 16-inch MacBook Pro, a $150 jump for the 11-inch iPad Air, and a $30 increase for the HomePod Mini.

Why it matters

Apple is normalizing AI as a built-in surcharge on everyday devices, which could lock in higher consumer tech prices across the market.

Link to Story

Other Developments

A curated list of other prominent stories from this day.

2:39 PMThe Verge

Margaret Atwood says the problem with AI is ‘garbage in, garbage out’

Summary

Margaret Atwood argued that today’s AI systems largely reflect the quality and bias of their training data, framing their failures as a predictable input problem rather than machine agency. She described experimenting with an AI tool and treated its outputs as derivative and contingent on what it was fed.

Why it matters

If the core AI risk is bad inputs, the policy fight moves to who controls training data and who pays when outputs cause damage.

Link to Story

12:45 PMTechCrunch

Apple Vision Pro exec is reportedly leaving for OpenAI

Summary

Paul Meade, Apple’s vice president overseeing Vision Pro, is reportedly leaving Apple to join OpenAI’s hardware team.

Why it matters

Leadership moves like this can shift execution speed and competitive positioning in the race to define the next consumer computing platform.

Link to Story

Make it yours

Build Your First Pass.

Pick your topics, set your cadence, and receive your personalized First Pass in your inbox. It’s that simple!