Large language models (LLMs) are becoming increasingly more impressive at creating human-like text and answering questions, but whether they can understand the meaning of the words they generate is a hotly debated issue. A big challenge is that LLMs are black boxes; they can make predictions and decisions on the order of words, but they cannot communicate the reasons for doing so.

Ellie Pavlick at Brown University is building models that could help understand how LLMs process language compared with humans. In this episode of The Joy of Why, Pavlick discusses what we know and don’t know about LLM language processing, how their processes differ from humans, and how understanding LLMs better could also help us better appreciate our own capacity for knowledge and creativity.

Quantum gravity is one of the biggest unresolved and challenging problems in physics, as it seeks to reconcile quantum mechanics, which governs the microscopic world, and general relativity, which describes the macroscopic world of gravity and space-time.

Efforts to understand quantum gravity have been focused almost entirely at the theoretical level, but Monika Schleier-Smith at Stanford University has been exploring a novel experimental approach — trying to create quantum gravity from scratch. Using laser-cooled clouds of atoms, she is testing the idea that gravity might be an emergent phenomenon arising from quantum entanglement.

In this episode of the Joy of Why podcast, Schleier-Smith discusses the thinking behind what she admits is a high-risk, high-reward approach, and how her experiments could provide important insights about entanglement and quantum mechanical systems even if the end goal of simulating quantum gravity is never achieved.

Quantum computing promises unprecedented speed, but in practice, it’s proven remarkably difficult to find important questions that quantum machines can solve faster than classical ones. One of the most notable demonstrations of this came from Ewin Tang, who rose to prominence in the field as a teenager. When quantum algorithms had in principle cracked the so-called recommendation problem, Tang designed classical algorithms that could match them.

So began the approach of “dequantizing,” in which computer scientists look at quantum algorithms and try to achieve the same speeds with classical counterparts. To understand the ongoing contest between classical and quantum computing, co-host Janna Levin spoke to Tang on The Joy of Why podcast. The wide-ranging conversation covered what it was like for Tang to challenge the prevailing wisdom at such a young age, the role of failure in scientific progress, and whether quantum computing will ultimately fulfill its grand ambitions.

At first, life on Earth was simple. Cells existed, functioned and reproduced as free-living individuals. But then, something remarkable happened. Some cells joined forces, working together instead of being alone. This transition, known as multicellularity, was a pivotal event in the history of life on Earth. Multicellularity enabled greater biological complexity, which sparked an extraordinary diversity of organisms and structures.

How life evolved from unicellular to multicellular organisms remains a mystery, though evidence indicates that this may have occurred multiple times independently. To understand what could have happened, Will Ratcliff at Georgia Tech has been conducting long-term evolution experiments on yeast in which multicellularity develops and emerges spontaneously.

In this episode of The Joy of Why podcast, Ratcliff discusses what his “snowflake yeast” model could reveal about the origins of multicellularity, the surprising discoveries his team has made, and how he responds to skeptics who question his approach.

How did complex life evolve? Where did space-time come from? Will computers ever understand language like we do? How did geometry create modern physics? These are just a few of the big and bold questions that we’ll be exploring in the latest season of Quanta’s interview podcast, “The Joy of Why,” starting March 20, and released every other Thursday.

We have identified thousands of planets just in our neighborhood in the Milky Way, mostly from the way they impact their host stars. Basic calculations suggest that there are countless more across the galaxy, and that billions of them could potentially support life. But what kind of life they host, and how we would be able to detect the presence of those biological processes from Earth, remain big questions in the world of exoplanets and astrobiology. What technologies might lie ahead to help us answer the question of whether we are alone in the universe? Lisa Kaltenegger, an astrophysicist and astrobiologist at Cornell University, talks to Janna Levin about that search, the atmospheric fingerprints of life, and why an advanced alien civilization might decide not to talk to us.

Death might seem like a pure loss, the disappearance of what makes a living thing distinct from everything else on our planet. But zoom in closer, to the cellular level, and it takes on a different, more nuanced meaning. There is a challenge in simply defining what makes an individual cell alive or dead. Scientists today are working to understand the various ways and reasons that cells disappear, and what these processes mean to biological systems. In this episode, cellular biologist Shai Shaham talks to Steven Strogatz about the different forms of cell death, their roles in evolution and disease, and why the right kinds and patterns of cell death are essential to our development and well-being.

It’s fair to say that enjoyment of a podcast would be severely limited without the human capacity to create and understand speech. That capacity has often been cited as a defining characteristic of our species, and one that sets us apart in the long history of life on Earth. Yet we know that other species communicate in complex ways. Studies of the neurological foundations of language suggest that birdsong, or communication among bats or elephants, originates with brain structures similar to our own. So why do some species vocalize while others don’t? In this episode, Erich Jarvis, who studies behavior and neurogenetics at the Rockefeller University, chats with Janna Levin about the surprising connections between human speech, birdsong and dance.