This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.
Taiwan’s “silicon shield” could be weakening
Taiwanese politics increasingly revolves around one crucial question: Will China invade? China’s ruling party has wanted to seize Taiwan for more than half a century. But in recent years, China’s leader, Xi Jinping, has placed greater emphasis on the idea of “taking back” the island (which the Chinese Communist Party, or CCP, has never controlled).
Many in Taiwan and elsewhere think one major deterrent has to do with the island’s critical role in semiconductor manufacturing. Taiwan produces the majority of the world’s semiconductors and more than 90% of the most advanced chips needed for AI applications.
But now some Taiwan specialists and some of the island’s citizens are worried that this “silicon shield,” if it ever existed, is cracking. Read the full story.
—Johanna M. Costigan
This story is from our forthcoming print issue, which is all about security. If you haven’t already, subscribe now to receive future issues once they land.
Why there’s a big backlash against ChatGPT’s new ‘personality’
When OpenAI made the switch to its new GPT-5 model last week, a number of people reacted with shock, frustration, sadness, or anger to previous model 4o’s sudden disappearance from ChatGPT.
Despite its awareness that people are developing emotional bonds with the model, OpenAI appears to have been caught flat-footed by the fervor of users’ pleas for its return. Within a day, the company made 4o available again to its paying customers (free users are stuck with GPT-5).
MIT Technology Review spoke with several ChatGPT users who were deeply affected by the loss of 4o. All are women between the ages of 20 and 40, and all bar one considered 4o to be a romantic partner. Read the full story.
—Grace Huckins
Why US federal health agencies are abandoning mRNA vaccines
This time five years ago, we were in the throes of the covid-19 pandemic. Then came the vaccines. The first mRNA vaccines for covid were authorized for use in December 2020. The US government played an important role in the introduction of these vaccines, providing $18 billion to support their development.
But now, that government is turning its back on the technology. Funding is being withdrawn. Partnerships are being canceled. Leaders of US health agencies are casting doubt on the vaccines’ effectiveness and safety. And this week, the director of the National Institutes of Health implied that the reversal was due to a lack of public trust in the technology.
Plenty of claims are being thrown about. So let’s consider the evidence. Read the full story.
—Jessica Hamzelou
This article first appeared in The Checkup, MIT Technology Review’s weekly biotech newsletter. To receive it in your inbox every Thursday, and read articles like this first, sign up here.
The must-reads
I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.
1 The Trump administration is in talks to buy a stake in Intel
Just weeks after Trump called for the CEO to step down. (Bloomberg $)
+ It’s part of its plan to increase US market share in chip manufacturing. (WSJ $)
+ Intel is probably hoping such a deal could help its beleaguered Ohio factory. (TechCrunch)
2 Meta’s AI rules allowed its chatbots to flirt with children
And it only recently amended the guidelines after being questioned about it. (Reuters)
+ We don’t know how long the policies were in place. (The Verge)
+ An AI companion site is hosting sexually charged conversations with underage celebrity bots. (MIT Technology Review)
3 Erin is America’s first real test of hurricane readiness under Trump
It looks like it’ll become the season’s first hurricane. (Vox)
+ Trackers are uncertain about where the storm will head. (NYT $)
+ Here’s what we know about hurricanes and climate change. (MIT Technology Review)
4 xAI lost a major US government contract after Grok praised Hitler
Leaving the government to partner with OpenAI, Anthropic, and Gemini instead. (Wired $)
+ xAI’s ‘Grok for Government’ site doesn’t appear to reflect this. (Ars Technica)
5 Tech leaders are upping their security
As public hostility towards corporate executives deepens. (FT $)
6 These TikTokers are documenting their lives after deportation
They’re sharing their realities and creating new communities. (NY Mag $)
+ ICE added a random person to a highly sensitive group chat. (404 Media)
7 We may soon be able to hear some patients’ inner voices
New research has successfully guessed words imagined by people unable to speak. (NYT $)
+ Motor neuron diseases took their voices. AI is bringing them back. (MIT Technology Review)
8 China’s plug-in hybrids are everywhere
And they’re likely to dominate exports for the next three years at least. (Rest of World)
+ China’s EV giants are betting big on humanoid robots. (MIT Technology Review)
9 The UK is working with TikTok influencers to tackle medical tourism
It’s a bid to raise awareness of the risks of undertaking cosmetic surgery abroad. (BBC)
10 AI may experience the passage of time differently to us
What does this mean for our future? (IEEE Spectrum)
+ What is AI? (MIT Technology Review)
Quote of the day
“We’ve realized the best way to get them is when they’re scrolling social media.”
—Ryan Odendahl, president and CEO of construction company Kwest Group, tells the Washington Post how his company is getting young people interested in learning traditional trades.
One more thing
The next generation of neural networks could live in hardware
Networks programmed directly into computer chip hardware can identify images faster, and use much less energy, than the traditional neural networks that underpin most modern AI systems.
Neural networks, from GPT-4 to Stable Diffusion, are built by wiring together perceptrons, which are highly simplified simulations of the neurons in our brains. In very large numbers, perceptrons are powerful, but they also consume enormous volumes of energy.
Part of the trouble is that perceptrons are just software abstractions—running a perceptron network on a GPU requires translating that network into the language of hardware, which takes time and energy. Building a network directly from hardware components does away with a lot of those costs. And one day, they could even be built directly into chips used in smartphones and other devices. Read the full story.
—Grace Huckins
We can still have nice things
A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)
+ Ever wished you knew more about art? This YouTube channel is a fantastic resource.
+ A very happy birthday to Madonna Louise Ciccone, who turns 67 years young tomorrow.
+ What do dolphins and whales really think of each other? 
+ A fond farewell to thrash metal titans Megadeth, who are retiring next year.