The Download: growing threats to vulnerable languages, and fact-checking Trump’s medical claims
Plus: Huntington’s disease has been treated successfully for the first time.
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.
How AI and Wikipedia have sent vulnerable languages into a doom spiral
Wikipedia is the most ambitious multilingual project after the Bible: There are editions in over 340 languages, and a further 400 even more obscure ones are being developed. But many of these smaller editions are being swamped with AI-translated content. Volunteers working on four African languages, for instance, estimated to MIT Technology Review that between 40% and 60% of articles in their Wikipedia editions were uncorrected machine translations.
This is beginning to cause a wicked problem. AI systems learn new languages by scraping huge quantities of text from the internet. Wikipedia is sometimes the largest source of online linguistic data for languages with few speakers—so any errors on those pages can poison the wells that AI is expected to draw from. Volunteers are being forced to go to extreme lengths to fix the issue, even deleting certain languages from Wikipedia entirely. Read the full story.
—Jacob Judah
This story is part of our Big Story series: MIT Technology Review’s most important, ambitious reporting. These stories take a deep look at the technologies that are coming next and what they will mean for us and the world we live in. Check out the rest of the series here.
Trump is pushing leucovorin as a new treatment for autism. What is it?
On Monday, President Trump claimed that childhood vaccines and acetaminophen, the active ingredient in Tylenol, are to blame for the increasing prevalence of autism. He advised pregnant women against taking the medicine.
The administration also announced that the FDA would work to make a medication called leucovorin available as a treatment for children with autism. The president’s assertions left many dismayed. “The data cited do not support the claim that Tylenol causes autism and leucovorin is a cure, and only stoke fear and falsely suggest hope when there is no simple answer,” said the Coalition for Autism Researchers, a group of more than 250 scientists, in a statement. So what does the evidence say? Read our story to find out.
—Cassandra Willyard
This is part of our MIT Technology Review Explains series, where our writers untangle the complex, messy world of technology to help you understand what’s coming next. You can read more from the series here.
Fusion power plants don’t exist yet, but they’re making money anyway
This week, Commonwealth Fusion Systems announced it has another customer for its first commercial fusion power plant, in Virginia. Eni, one of the world’s largest oil and gas companies, signed a billion-dollar deal to buy electricity from the facility.
One small detail? That reactor doesn’t exist yet. This is a weird moment in fusion. Investors are pouring billions into the field to build power plants, and companies are even signing huge agreements to purchase power from those still-nonexistent plants.
But all this comes before companies have actually completed a working reactor that can produce electricity. It takes money to develop a new technology, but all this funding could lead to some twisted expectations. Read the full story.
—Casey Crownhart
This story is from The Spark, our weekly newsletter all about the latest in climate change and clean tech. Sign up to receive it in your inbox every Wednesday.
The AI Hype Index: Cracking the chatbot code
Millions of us use chatbots every day, even though we don’t really know how they work or how using them affects us. In a bid to address this, the FTC recently launched an inquiry into how chatbots affect children and teenagers. Elsewhere, OpenAI has started to shed more light on what people are actually using ChatGPT for, and why it thinks its LLMs are so prone to making stuff up.
There’s still plenty we don’t know—but that isn’t stopping governments from forging ahead with AI projects. In the US, RFK Jr. is pushing his staffers to use ChatGPT, while Albania is using a chatbot for public contract procurement. Check out the latest edition of our AI Hype Index to help you sort AI reality from hyped-up fiction.
The must-reads
I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.
1 Huntington’s disease has been treated successfully for the first time
Gene therapy managed to slow progress of the disease in patients by 75%. (The Economist $)
+ Here’s how the gene editing tool CRISPR is changing lives. (MIT Technology Review)
2 Google says 90% of tech workers are using AI
But most of them also say they don’t trust AI models’ outputs. (CNN)
+ Why does AI hallucinate? (MIT Technology Review)
3 A MAGA TikTok takeover is coming
Just as free speech protections in the US start to look worryingly fragile. (The Atlantic $)
4 Chinese tech workers are returning from the US
There’s a whole bunch of complex factors both driving them to leave, and luring them back. (Rest of World)
+ But it’s hard to say what the impact of the new $100,000 fee for H-1B visas will be on India’s tech sector. (WP $)
+ Europe is hoping to nab more tech talent too. (The Verge)
5 If AI can diagnose us, what are doctors for?
They need to prepare for the fact chatbot use is becoming more and more widespread among patients. (New Yorker $)
+ This medical startup uses LLMs to run appointments and make diagnoses. (MIT Technology Review)
6 Drones have been spotted at four more airports in Denmark
It looks like a coordinated attack, but officials still haven’t worked out who is behind it. (FT $)
7 TSMC has unveiled AI-designed chips that use less energy
The AI software found better solutions than TSMC’s own human engineers—and did so much faster. (South China Morning Post)
+ These four charts sum up the state of AI and energy. (MIT Technology Review)
8 How to find love on dating apps 💑
It’s not easy, but it is possible. (The Guardian)
9 AI models can’t cope with Persian social etiquette
It involves a lot of saying ‘no’ when you mean ‘yes’, which simply doesn’t wash with computers. (Ars Technica)
10 VR headsets are better than ever, but no one seems to care
The tech industry keeps overestimating how willing people are to strap computers to their faces. (Gizmodo)
Quote of the day
“We are living through the most destructive arms race in human history.”
—Ukrainian president Volodymyr Zelenskyy tells world leaders gathered at the UN that they need to intervene to stop the escalating development of drone technology and AI, The Guardian reports.
One more thing
STUART BRADFORD
The great AI consciousness conundrum
AI consciousness isn’t just a tricky intellectual puzzle; it’s a morally weighty problem. Fail to identify a conscious AI, and you might unintentionally subjugate a being whose interests ought to matter. Mistake an unconscious AI for a conscious one, and you risk compromising human safety and happiness for the sake of an unthinking, unfeeling hunk of silicon and code.
Over the past few decades, a small research community has doggedly attacked the question of what consciousness is and how it works. The effort has yielded real progress. And now, with the rapid advance of AI technology, these insights could offer our only guide to the untested, morally fraught waters of artificial consciousness. 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.)
+ It’s Fat Bear Week! Who gets your vote this year?
+ Learn about Lord Woodbine, the forgotten sixth Beatle.
+ There are some truly wild and wacky recipes in this Medieval Cookery collection. Venison porridge, anyone?
+ Pessimism about technology is as old as technology itself, as this archive shows.
