The Download: understanding deep matter, and AI jailbreak protection

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 the Rubin Observatory will help us understand dark matter and dark energy

We can put a good figure on how much we know about the universe: 5%. That’s how much of what’s floating about in the cosmos is ordinary matter—planets and stars and galaxies and the dust and gas between them. The other 95% is dark matter and dark energy, two mysterious entities aptly named for our inability to shed light on their true nature.

Previous work has begun pulling apart these dueling forces, but dark matter and dark energy remain shrouded in a blanket of questions—critically, what exactly are they?

Enter the Vera C. Rubin Observatory, one of our 10 breakthrough technologies for 2025. Boasting the largest digital camera ever created, Rubin is expected to study the cosmos in the highest resolution yet once it begins observations later this year. And with a better window on the cosmic battle between dark matter and dark energy, Rubin might narrow down existing theories on what they are made of. Here’s a look at how.

—Jenna Ahart

This story is part of MIT Technology Review Explains, our series untangling the complex, messy world of technology to help you understand what’s coming next. You can read more from the series here.

Anthropic has a new way to protect large language models against jailbreaks

What’s new? AI firm Anthropic has developed a new line of defense against a common kind of attack called a jailbreak. A jailbreak tricks large language models (LLMs) into doing something they have been trained not to, such as help somebody create a weapon. And Anthropic’s new approach could be the strongest shield against the attacks yet.

How they did it: Jailbreaks are a kind of adversarial attack: input passed to a model that makes it produce an unexpected output. Despite a decade of research there is still no way to build a model that isn’t vulnerable. But, instead of trying to fix its models, Anthropic has developed a barrier that stops attempted jailbreaks from getting through and unwanted responses from the model getting out. Read the full story.

—Will Douglas Heaven

Three things to know as the dust settles from DeepSeek

The launch of a single new AI model does not normally cause much of a stir outside tech circles, nor does it typically spook investors enough to wipe out $1 trillion in the stock market. Now, a couple of weeks since DeepSeek’s big moment, the dust has settled a bit.

Within AI, though, what impact is DeepSeek likely to have in the longer term? Here are three seeds DeepSeek has planted that will grow even as the initial hype fades.

—James O’Donnell

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here.

If you’re interested in learning more about what DeepSeek’s breakout success means for the future of AI, watch this conversation between our news editor Charlotte Jee, senior AI editor Will Douglas Heaven, and China reporter Caiwei Chen. It was held at noon ET yesterday as part of our subscriber-only Roundtables series—check it out!

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 Elon Musk’s government allies are weighing up using AI to cut costs  
As part of Musk’s plans to gut federal contracts across the board. (NYT $)
+ A 25-year old engineer now has access to the US’s top secret systems. (Wired $)
+ Staffers for the US agency that sends aid to the world’s neediest have been locked out of their email accounts. (NY Mag $)
+ Such measures would have been unthinkable just a few short years ago. (Vox)
+ Palantir CEO Alex Karp is a fan of ‘DOGE’. (Insider $)

2 China has announced its own tariffs on US imports
Sparking new fears of a full-blown trade war. (FT $)
+ The days of cheap Chinese shopping in the US could be coming to an end. (NY Mag $)
+ Here’s what Trump’s tariffs mean for the likes of Temu and Shein. (The Information $)

3 US senators blame Silicon Valley for DeepSeek’s runaway success
Big Tech’s lobbying for softer export controls created corporate loopholes, they claim. (WP $)
+ The rise of DeepSeek doesn’t mean the controls have failed, according to ASML. (WSJ $)
+ How a top Chinese AI model overcame US sanctions. (MIT Technology Review)

4 Meta says it won’t release AI systems it deems too risky
But how that risk is measured is up to Meta. (TechCrunch)
+ A new public database lists all the ways AI could go wrong. (MIT Technology Review)

5 Gender affirming care is under major threat in the US
Advocates fear Trump’s executive order will prevent many people from accessing lifesaving treatments. (Undark)
+ Many hospitals are continuing to offer their services, though. (Axios)+ New York’s Attorney General says pausing such care could violate state law. (The Hill)

6 The App Store is now hosting its first porn app
And Apple is not happy about it. (Reuters)
+ The company has an EU antitrust law to thank. (WP $)

7 The Doomsday Clock has been given a makeover
We are now 89 seconds away from the end of the world. (Fast Company $)

8 Meet the UK’s AI grandmother wasting scammers’ time
Fraudsters have been left frustrated by the bot’s dithering. (The Guardian)
+ The people using humour to troll their spam texts. (MIT Technology Review)

9 We still don’t know much about Mars’ moons
But a new mission could change that. (New Scientist $)

10 Mark Zuckerberg’s famous hoodie is up for auction
If you’re so inclined to want to own a piece of nerd history. (Insider $)

Quote of the day

“It’ll scare people, it’ll make people think that the industry is a scam.”

—Anthony Scaramucci, Donald Trump’s former communications director, doesn’t think much of his former boss’s memecoin, he tells the Financial Times.

The big story

The open-source AI boom is built on Big Tech’s handouts. How long will it last?

May 2023

In May 2023 a leaked memo reported to have been written by Luke Sernau, a senior engineer at Google, said out loud what many in Silicon Valley must have been whispering for weeks: an open-source free-for-all is threatening Big Tech’s grip on AI.

New open-source large language models—alternatives to Google’s Bard or OpenAI’s ChatGPT that researchers and app developers can study, build on, and modify—are dropping like candy from a piñata. These are smaller, cheaper versions of the best-in-class AI models created by the big firms that (almost) match them in performance—and they’re shared for free.

In many ways, that’s a good thing. AI won’t thrive if just a few mega-rich companies get to gatekeep this technology or decide how it is used. But this open-source boom is precarious, and if Big Tech decides to shut up shop, a boomtown could become a backwater. Read the full story.

—Will Douglas Heaven

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.)

+ Today would have been the 112th birthday of Rosa Parks, the civil activist who changed the course of history.
+ If you’re planning a spring break, consider this well-timed inspiration.
+ A Buffy the Vampire Slayer reboot is reportedly in the works.
+ Rise up, daughters of grunge!