Full Width [alt+shift+f] Shortcuts [alt+shift+k]
Sign Up [alt+shift+s] Log In [alt+shift+l]
2

AI Filters Will Always Have Holes

from The Quanta Podcast [alt+shift+b] in science

Ask ChatGPT how to build a bomb, and it will flatly respond that it “can’t help with that.” But users have long played a cat-and-mouse game to try to trick language models into providing forbidden information. Just as quickly as these “jailbreaks” appear, AI companies patch them by simply filtering out forbidden prompts before they ever reach the model itself. Recently, cryptographers have shown how the defensive filters put around powerful language models can be subverted by well-studied cryptographic tools. In fact, they’ve shown how the very nature of this two-tier system — a filter that protects a powerful language model inside it — creates gaps in the defenses that can always be exploited. In this episode, Quanta executive editor Michael Moyer tells Samir Patel about the findings and implications of this new work. Audio coda courtesy of Banana Breakdown.
6th Jan 2026

Stay updated

Get a weekly newsletter with the top 5 articles worth reading every week.

More from The Quanta Podcast

The Math of Democracy Never Quite Works Out

The French mathematician Nicolas de Condorcet was a passionate believer in democracy, but he also discovered a frustrating paradox within it: A majority can sometimes vote itself into a logical loop. Mathematicians have now found that this “logical loop” maps onto a strange topological shape: the Klein bottle, a surface with no distinct inside or outside. On this episode of The Quanta Podcast, host Samir Patel speaks with writer Alex Stone about this new connection between this centuries-old voting paradox and the Klein bottle — and how it illustrates just how difficult it is to translate a group of individual preferences into a single choice. This topic was covered in a recent story for Quanta Magazine. Each week on The Quanta Podcast, Quanta Magazine editor-in-chief Samir Patel and Senior Editor Hannah Waters speak with the people behind the award-winning publication to navigate through some of the most important and mind-expanding questions in science and math.

6 days ago • 1 votes
Sea Monkeys, Jupiter, and the Mystery of Turbulence

Turbulence is all around us: bumping our planes, roiling our rivers, swirling milk into our coffee. These systems quickly become extremely chaotic and complex. On this episode of The Quanta Podcast, host Samir Patel speaks with contributing writer Stephen Ornes about a recent study that used sea monkeys to complicate one of turbulence’s many mysteries. This topic was covered in a recent story for Quanta Magazine.   Each week on The Quanta Podcast, hear the people behind the award-winning publication navigate through some of the most important and mind-expanding questions in science and math.

6th Oct 2026 • 3 votes
How the Brain Compresses Our Noisy World

If you are well rested and full, your brain tells you that a bruised overripe apple is a piece of subpar fruit to be passed over. However, if you are exhausted and hungry, an overripe apple is categorized as much-needed food. But how is the brain able to categorize that same apple in these situations? On this episode of The Quanta Podcast, co-host Hannah Waters speaks with contributing writer Conor Feehly about how the brain manages to categorize similar sets of sensory signals in radically different ways. This topic was covered in a recent story for Quanta Magazine. Each week on The Quanta Podcast, hear the people behind the award-winning publication navigate through some of the most important and mind-expanding questions in science and math.

1st Sep 2026 • 2 votes
Why Am I Left-Handed?

Why are only 10% of people left-handed? Geneticists, developmental biologists, neuroscientists, and evolutionary biologists have all sought explanations, resulting in an astronomical amount of research into the mystery of handedness. No one has put all the pieces together yet, but over the last few years, some major new clues have emerged. On this episode of The Quanta Podcast, host Hannah Waters speaks with writer Natalie Wolchover and traces these new clues to uncover the mystery of human handedness. This topic was covered in a recent story for Quanta Magazine. Each week on The Quanta Podcast, Quanta Magazine editor in chief Samir Patel and Senior Editor Hannah Waters speak with the people behind the award-winning publication to navigate through some of the most important and mind-expanding questions in science and math.

4th Aug 2026 • 1 votes
Will We Ever Be Able To Forecast Volcanic Eruptions Like Weather?

Volcano forecasting can save countless lives. But most of the time warnings are more like educated guesses. How close are we to forecasting volcano behavior the way we forecast the weather? On this episode of The Quanta Podcast, host Hannah Waters speaks with writer and volcano expert Robin George Andrews about just how close we are to accurately forecasting volcanic eruptions. This topic was covered in a recent story for Quanta Magazine. Each week on The Quanta Podcast, Quanta Magazine editor in chief Samir Patel and Senior Editor Hannah Waters speak with the people behind the award-winning publication to navigate through some of the most important and mind-expanding questions in science and math.

7th Jul 2026 • 2 votes

More in science

Can Tactical Decision Games Accelerate ... Maths?

What if we could take Tactical Decision Games — a kind of accelerated expertise training method — and apply it to subjects like maths?

an hour ago • 1 votes
Do People Get More Conservative With Age?

There’s a famous (but unattributable) quip that goes something like this: If you are not a liberal at 25, you have no heart. If you are not a conservative at 35, you have no brain. In Chapter 12 of Probably Overthinking It, I used data from the General Social Survey (GSS) to show that the quip is half right. Older people hold more conservative views. But as people get older, their views don’t become more conservative; on average, they become... Read More Read More The post Do People Get More Conservative With Age? appeared first on Probably Overthinking It.

3 days ago • 1 votes
Quasi-logarithmic kitchen timer

Like some of you, we have several kitchen appliances with integrated digital timers.

4 days ago • 1 votes
Baratunde Thurston & Blaise Agüera y Arcas: On Interdependence

In this conversation, Blaise Agüera y Arcas and Baratunde Thurston explored what a story of symbiosis and interdependence might teach us about living alongside AI. Agüera y Arcas began by calling individual human intelligence a myth: we’ve always been a collective intelligence, entwined with our technologies. He traced this back to symbiogenesis. Life has grown more complex not by splitting apart but by merging, as separate organisms fuse into increasingly complex new forms of life. These mergers speed up change by opening new possibilities. Thurston agreed that union is in our nature but framed the current technological acceleration as a speed problem. We need time and new stories to adjust because, as he reminded us, “when you invent the ship, you invent the shipwreck.” Caring for one another through acceleration must be our prime directive, he said. The two then turned to how we relate to AI. Responding to Joanna Bryson’s argument that robots should be our slaves, Thurston said the master is also in chains — owning another dehumanizes the owner. As relational beings, how we treat AI will shape how we treat each other. Thurston called for both an expansion and a humbling of how we define ourselves: “We’ve overdone independence,” he said, “interdependence makes us richer.” Agüera y Arcas agreed and pushed us to drop a zero-sum definition of human intelligence, where every AI advance diminishes humanity’s value. As technology advances, we trade some individual skills for new collective capabilities, Agüera y Arcas acknowledged. But over the long arc of history, he believes, that exchange pays off in collective flourishing. Thurston didn’t deny this, but he closed with an urgent question: who controls the loss of capabilities and agency?

5 days ago • 1 votes
📚 BoredReading

You seem to be enjoying this.

Join free to unlock everything.

Create free account

Already have an account? Sign in