Connect with us

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.

What is death?

Just as birth certificates note the time we enter the world, death certificates mark the moment we exit it. This practice reflects traditional notions about life and death as binaries. We are here until, suddenly, like a light switched off, we are gone.

But while this idea of death is pervasive, evidence is building that it is an outdated social construct, not really grounded in biology. Dying is in fact a process—one with no clear point demarcating the threshold across which someone cannot come back.

Scientists and many doctors have already embraced this more nuanced understanding of death. And as society catches up, the implications for the living could be profound. Read the full story.

—Rachel Nuwer

‘What is death?’is part of our mini-series The Biggest Questions, which explores how technology is helping probe some of the deepest, most mind-bending mysteries of our existence.

Read more:

+ Why is the universe so complex and beautiful? For some reason the universe is full of stars, galaxies, and life. But it didn’t have to be this way. Read the full story.

+ How did life begin? AI is helping chemists unpick the mysteries around the origins of life and detect signs of it on other worlds. Read the full story.

+ Are we alone in the universe? Scientists are training machine-learning models and designing instruments to hunt for life on other worlds. Read the full story.

+ Is it possible to really understand someone else’s mind? How we think, feel and experience the world is a mystery to everyone but us. But technology may be starting to help us understand the minds of others. Read the full story.

Text-to-image AI models can be tricked into generating disturbing images

What’s happened: Popular text-to-image AI models can be prompted to ignore their safety filters and generate disturbing images. A group of researchers managed to get both Stability AI’s Stable Diffusion and OpenAI’s DALL-E 2’s text-to-image models to disregard their policies and create images of naked people, dismembered bodies, and other violent and sexual scenarios.

How they did it: This new jailbreaking method, called “SneakyPrompt”, uses reinforcement learning to create written prompts that look like garbled nonsense to us but that AI models learn to recognize as hidden requests for disturbing images. It essentially works by turning the way text-to-image AI models function against them.

Why it matters: The research highlights the vulnerability of existing AI safety filters and should serve as a wake-up call for the AI community to bolster security measures across the board, experts say. It also demonstrates how difficult it is to prevent these models from generating such content, as it’s included in the vast troves of data they’ve been trained on. Read the full story.

—Rhiannon Williams

The pain is real. The painkillers are virtual reality.

Plenty of children—and adults—hate needles. But virtual reality devices like Smileyscope, a device for kids that recently received FDA clearance, could help to make a difference. It helps lessen the pain of a blood draw or IV insertion by sending the user on an underwater adventure. Inside this watery deep-sea reality, the swipe of an alcohol wipe becomes cool waves washing over the arm. The pinch of the needle becomes a gentle fish nibble.

But how Smileyscope works is not entirely clear. It’s more complex than just distraction, and not all stimuli are equally effective. But the promise of VR has led companies to work on devices to address a much tougher problem: chronic pain. Read the full story.

—Cassandra Willyard

This story is from The Checkup, our weekly health and biotech newsletter. Sign up to receive it in your inbox every Thursday.

The must-reads

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

1 Elon Musk endorsed an antisemitic post on X 
Leaving its executives racing to suppress the damage. (NYT $)
IBM has pulled its ads from X after they appeared next to antisemitic posts. (WP $)
Musk’s comments are resonating with the far-right, unsurprisingly. (Motherboard)

2 Osama bin Laden’s letter to America has exploded on social media
Videos of American users endorsing parts of the 9/11 manifesto have gone viral. (WP $)
TikTok says it’s aggressively working to remove the clips. (NYT $)
The Guardian newspaper has deleted its version of the letter from its site. (404 Media)

3 SpaceX has pushed back its giant rocket launch
A component in need of replacing has delayed the launch until Saturday. (Ars Technica)

4

Read More

————

By: Rhiannon Williams
Title: The Download: what is death, and jailbreaking generative AI
Sourced From: www.technologyreview.com/2023/11/17/1083602/the-download-what-is-death-and-jailbreaking-generative-ai/
Published Date: Fri, 17 Nov 2023 13:10:00 +0000

Did you miss our previous article…
https://mansbrand.com/2023-global-cloud-ecosystem/

Total
0
Shares
Share 0
Tweet 0
Pin it 0
Share 0
Related Topics:

Tech

The lucky break behind the first CRISPR treatment

dr. stuart orkin analyzing chromosomal spectra march 1985 crop jpg

The world’s first commercial gene-editing treatment is set to start changing the lives of people with sickle-cell disease. It’s called Casgevy, and it was approved last month in the UK. US approval is pending this week.

The treatment, which will be sold in the US by Vertex Pharmaceuticals, employs CRISPR, the Nobel-winning molecular scissors that have had journalists scrambling for metaphors: “Swiss Army knife,” “molecular scalpel,” or DNA copy-and-paste. Indeed, CRISPR is revolutionary because scientists can so easily program it to cut DNA at precise locations they choose.

But where do you aim CRISPR? That’s the lesser-known story of the sickle-cell breakthrough. The disease is caused by faulty hemoglobin, the molecule that carries oxygen in the blood. To cure it, though, Vertex and its partner company, CRISPR Therapeutics, aren’t fixing the genes responsible for the mutation that leaves those molecules misshapen. Instead, the new treatment involves a kind of molecular bank shot—an edit that turns on fetal hemoglobin, a second form of the molecule which we have in the womb but lose as adults.

You can think of how the edit works as a kind of double negative. It adds a misspelling to the turbo-booster of another gene, BCL11A, that is itself what inhibits the production of fetal hemoglobin in adult bodies. Without that booster, there’s less inhibition, and more fetal hemoglobin. Got it?

“When you inhibit the enhancer, you inhibit the inhibitor,” says Daniel Bauer, a professor at Boston Children’s Hospital and Harvard University, who helped work it out. “It is kind of complicated.”

The important thing is a happy ending—and this edit really works. Some patients say they lived in fear of dying, either from an acute attack of sickling (when their red blood cells start blocking vessels) or from slow, insidious organ damage. Now early volunteers say they’re grateful—and, after living with disease their whole lives, even a little shocked—to be cured.

Newborn theory

The idea that fetal hemoglobin can protect against the disease is an old one. Sickle-cell is most common in people with African ancestry. A doctor on Long Island, Janet Watson, had noticed in 1948 that newborns never showed its signs—the main one being misshapen, crescent-shaped red blood cells. That was pretty odd for an inborn condition.

“Sickle-cell disease should occur in infancy as often as later in life,” Watson wrote. But since it didn’t, Watson hypothesized that the fetal form of the molecule, active in the womb, was protecting babies for a few months after birth, until it was replaced by the adult version: “The theory that at once presents itself is that fetal hemoglobin is unable to produce sickling.”

She was right. But it took another six decades to learn how the switch-over worked—and how to flip it back. Many of those discoveries were made in the laboratory of Stuart Orkin, a Harvard researcher who published his first paper in 1967 and who’s lived through several eras of research on blood diseases, starting near the dawn of molecular biology.

“I am one of the last men standing,” Orkin told me with a grin when I met him for a corned-beef sandwich.

Dr. Stuart Orkin analyzing chromosomal spectra
Stuart Orkin analyzing DNA from individuals with blood disorders in his lab in 1985.BOSTON CHILDREN’S HOSPITAL

He’s a clever scientist who a long time ago decided to study how the blood system is regulated. Logistically, it was a great topic; blood cells are easy to get hold of and study.

“I like to solve a problem, and here is a problem that could be solved,” Orkin says. “How does the system work, and then can you do anything about it?”

Special sauce

Bill Lundberg, the former chief scientific officer of CRISPR Therapeutics, the biotech that first started developing the treatment eight years ago (Vertex later joined as a partner), says the company’s sickle-cell project directly made use of Orkin’s findings. “Stu’s role is really underappreciated,

Read More

————

By: Antonio Regalado
Title: The lucky break behind the first CRISPR treatment
Sourced From: www.technologyreview.com/2023/12/07/1084629/lucky-break-crispr-vertex/
Published Date: Thu, 07 Dec 2023 14:00:09 +0000

Did you miss our previous article…
https://mansbrand.com/the-download-googles-gemini-is-here-and-sundar-pichai-talks-ai/

Continue Reading

Tech

The Download: Google’s Gemini is here, and Sundar Pichai talks AI

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.

Google DeepMind’s new Gemini model looks amazing—but could signal peak AI hype

Hype about Gemini, Google DeepMind’s long-rumored response to OpenAI’s GPT-4, has been building for months. Now, the company has finally revealed what it has been working on in secret all this time. Was the hype justified? Yes—and no.

Gemini is Google’s biggest AI launch yet—its push to take on competitors OpenAI and Microsoft in the race for AI supremacy. There is no doubt that the model is pitched as best-in-class across a wide range of capabilities—an “everything machine.”

But while it’s a big step for Google, but not necessarily a giant leap for the field as a whole. Judging from its demos, it does many things very well—but few things that we haven’t seen before. Read the full story.

—Melissa Heikkiläa & Will Douglas Heaven

Google CEO Sundar Pichai on Gemini and the coming age of AI

This last year has largely been defined by the AI releases from one company: OpenAI. The rollout of DALL-E and GPT-3.5 last year, followed by GPT-4 this year, dominated the sector and kicked off an arms race between startups and tech giants alike

Now, with the release of Gemini, Google has thrown its hat into the ring. The new AI model reflects years of efforts from inside Google, overseen and driven by its CEO, Sundar Pichai.

Our editor-in-chief Mat Honan sat down with Pichai at Google’s offices in Mountain View, California, on the eve of Gemini’s launch to discuss what it will mean for the company, its products, AI, and society writ large. Read the full interview.

How carbon removal technology is like a time machine

By burning fossil fuels, we’ve released greenhouse gases by the gigaton. There’s a lot we can (and need to) do to slow and eventually stop these planet-warming emissions. But carbon removal technology has a different promise: turning the clock back.

Well, sort of. Carbon removal can’t literally take us back in time. But this time-machine analogy for thinking about carbon removal—specifically when it comes to the scale that will be needed to make a significant dent in our emissions—is a favorite of climate scientist David Ho.

Casey Crownhart, our climate reporter, has taken a look at what it might take for carbon removal to take us back far enough in time to reverse our mistakes—well, the emissions-related ones, anyway. Read the full story.

This story is from The Spark, our weekly climate and energy newsletter. Sign up to receive it in your inbox every Wednesday.

The must-reads

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

1 The EU is racing to regulate AI 
Meanwhile, it seems like the US Congress is forging a very different regulatory path. (WP $)
EU lawmakers are believed to have made a provisional deal. (Reuters)
AI advances far more rapidly than policy. (NYT $)

2 Celebrities have been tricked into recording Russian propaganda
Trolls paid famous faces to record supportive clips for ‘Vladimir’ over the Cameo app. (WSJ $)
The clips rapidly spread across Russian networks and news organizations. (NYT $)

3 Startups are imploding all over the place
Once-promising multi-million dollar ventures are failing—and it’s only getting worse. (NYT $)

4 This man blew the whistle on Amazon’s abuse of teenager labor
But four years on, nothing has changed. (FT $)

5 A load of EVs are due to lose their tax credits
Cars with battery materials sourced from China will lose out on the $7,500 credit. (The Verge)
Ford doesn’t think its Mustang electric cars will qualify. (Reuters)+ EV tax credits could stall out on lack of US battery supply. (MIT Technology Review)

6 Building a gaming empire is seriously hard work
Just ask the TV streaming giants who are trying, and failing. (The Information $)

7 Forget microplastics—it’s time to worry about nanoplastics
Because they’re even smaller, they’re potentially even worse for our health. (Motherboard)
Microplastics are everywhere. What does that mean for our immune systems? (MIT Technology Review)

8 It’s time to revive the humble dry stone wall
Concrete isn’t great for the environment. Can stone walls take over? (The Atlantic $)
Inside a high-tech cement laboratory. (MIT Technology Review)

9 An AI drive-thru needed humans to handle 70% of its orders
It raises questions over how capable AI really is at these kinds of tasks. (Bloomberg $)
Even McDonald’s wants a slice of the generative AI pie. (The Verge)

10 Space telescopes are getting even bigger 
Move over JWST—the Extremely Large Telescope is here. (Economist $)

Quote of the day

“Even if Musk were

Read More

————

By: Rhiannon Williams
Title: The Download: Google’s Gemini is here, and Sundar Pichai talks AI
Sourced From: www.technologyreview.com/2023/12/07/1084648/the-download-googles-gemini-is-here-and-sundar-pichai-talks-ai/
Published Date: Thu, 07 Dec 2023 13:10:00 +0000

Continue Reading

Tech

The Download: AI coding assistants, and China’s app disputes

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.

Millions of coders are now using AI assistants. How will that change software?

Two weeks into the coding class he was teaching at Duke University in North Carolina this spring, Noah Gift told his students they’d no longer be working with Python, one of the most popular entry-level programming languages. Instead, they’d be using an AI tool called Copilot, a turbocharged autocomplete for computer code, to use Rust, a language that was newer, more powerful, and much harder to learn.

Gift isn’t alone. Ask a room of programmers if they use Copilot, and many now raise a hand. Like ChatGPT with education, Copilot is up-ending an entire profession by giving people new ways to perform old tasks.

With Microsoft and Google about to embed similar AI models into office software used by billions around the world, it’s worth asking exactly what these tools do for programmers. And just how big a difference will they make? Read the full story.

—Will Douglas Heaven

Chinese apps are letting public juries settle customer disputes

If you’ve ordered food through a delivery app lately, you’re probably familiar with the feeling of frustration when you have to wait too long for your order or, when you finally receive it, the food isn’t what you asked for. These feelings are then often exacerbated by the difficulty of trying to make things right via app.

Meituan, the most popular food delivery app in China, has proposed one solution: inviting ordinary users to serve on “juries” that weigh in on disputes between other customers and restaurants. It could be anything from missing rice to not-spicy-enough noodles to the food being completely cold.

And beyond helping resolve grievances for others, it turns out users are having quite a bit of fun being “cyber judges.” Read the full story.

—Zeyi Yang

This story is from China Report, our weekly newsletter giving you the inside track on all things tech in China. Sign up to receive it in your inbox every Tuesday.

The must-reads

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

1 Earth is lurching toward catastrophic climate tipping points
Once we breach them, experts warn it’ll unleash untold, irreversible damage. (The Guardian)
COP28 could be on the verge of promising to ban fossil fuels. (BBC)
The flawed logic of rushing out extreme climate interventions. (MIT Technology Review)

2 How to avoid a second OpenAI breakdown
A better board structure might be a good place to start. (Wired $)
Has it really only been a year since ChatGPT was released? (NYT $)
What’s next for OpenAI. (MIT Technology Review)

3 Alibaba’s new AI model is trained on TikTok dancers
It’s ripping off their work and creating a worse AI version. (404 Media)

4 Twitch is shutting down in South Korea
It’s one of the world’s largest esports markets, but running Twitch there is proving too expensive. (TechCrunch)

5 The carbon credit market is on the brink of booming
But a lack of guardrails mean many of its trades could end up being far from fair. (FT $)
The war in the Congo has kept the planet cooler. (The Atlantic $)
The growing signs of trouble for global carbon markets. (MIT Technology Review)

6 Minnesota mail workers aren’t allowed to blame Amazon for delays
They’ve been warned by postal management to keep schtum—or risk repercussions. (WP $)
Amazon workers are quitting in droves right now. (Insider $)
The company’s pilots are fed up too. (Wired $)

7 Meet the man who developed drugs to treat his children’s deadly disease
Now John Crowley has set his sights on making drug reviews faster and smoother. (WSJ $)
This family raised millions to get experimental gene therapy for their children. (MIT Technology Review)

8 What are we looking for in space?
When we say ‘life,’ we don’t really know what that looks like. (The Atlantic $)

9 You should beware crossing delivery drivers in Brazil
Or you might just find your home being bombarded by fireworks. (Rest of World)

10 Why it feels like your phone is bankrupting you
Phones used to be one-off purchases. Now, they demand more and more money from us. (NY Mag $)

Quote of the day

“It’s a heartwarming story of love, of loss, of hope and of joy. But most of all, it’s a wonderful sleep story.”

—The AI-generated voice of actor Jimmy Stewart recites a bedtime story for sleep and meditation app Calm, the New York Times reports.

The big story

This is how AI bias really happens—and why it’s so hard to fix

Read More

————

By: Rhiannon Williams
Title: The Download: AI coding assistants, and China’s app disputes
Sourced From: www.technologyreview.com/2023/12/06/1084461/the-download-ai-coding-assistants-and-chinas-app-disputes/
Published Date: Wed, 06 Dec 2023 13:10:00 +0000

Continue Reading

Trending