AI Chatbots Will Never Stop Hallucinating
Some amount of chatbot hallucination is inevitable. But there are ways to minimize it
Stories by Lauren Leffer
Lauren Leffer is a contributing writer and former tech reporting fellow at Scientific American. She reports on many subjects including artificial intelligence, climate and weird biology because she's curious to a fault. When she's not writing, she's hopefully hiking. Follow her on X (formerly Twitter) @lauren_leffer and on Bluesky @laurenleffer.bsky.social
Inside the Race to Protect Artists from Artificial Intelligence
AI-generated art is creating new ethical issues—and competition—for digital artists. Nightshade and Glaze are two tools helping creators fight back.
Song Lyrics Really Are Getting Simpler and More Repetitive, Study Finds
An assessment of hundreds of thousands of songs confirms that choruses and hooks have taken over—but simpler isn’t necessarily worse
Banning TikTok Would Do Basically Nothing to Protect Your Data
Proposed restrictions on TikTok would be “security theater” in the face of the staggering amounts of data that foreign and domestic tech companies collect
Too Much Trust in AI Poses Unexpected Threats to the Scientific Process
It’s vital to “keep humans in the loop” to avoid humanizing machine-learning models in research
Everything to Know About OpenAI’s New Text-to-Video Generator, Sora
A machine-learning tool that transforms text prompts into detailed video has generated excitement—and skepticism
Artists Are Slipping Anti-AI ‘Poison’ into Their Art. Here’s How It Works
Digital cloaking tools such as Nightshade and Glaze help artists take back control from generative AI—but they’re no forever fix
What Do You Mean, Bisexual People Are ‘Risk-Taking’? Why Genetic Studies about Sexuality Can Be Fraught
A recent study on risk-taking and bisexuality made assumptions that some experts don’t agree with.
What Apple’s New Vision Pro Headset Might Do to Our Brain
The release of Apple’s mixed-reality headset raises questions about hours spent in a virtual replacement of our world
Is Bisexuality Genetic? It’s More Complex Than Some Studies Imply
The controversy over a recent paper on human bisexual behavior emphasizes how important it is not to overinterpret genetic studies of sexuality—and how easy it is to do so
A Camera-Wearing Baby Taught an AI to Learn Words
Most machine-learning models rely on mountains of data to replicate human text, but new research suggests the recipe for learning language might be simpler
AI Audio Deepfakes Are Quickly Outpacing Detection
An alleged voice recording of racist remarks exemplifies the challenges of our new AI normal
Animals Can See Colors We Can’t—And New Tech Offers Us a Glimpse
A colorful new video technique lets scientists see the world like birds and bees
Can AI Predict Your Death?
A new study used machine learning to “autocomplete” the life trajectories of six million Danish people—and forecast when they might kick the bucket.
Can this AI Tool Predict Your Death? Maybe, But Don’t Panic
Amid the machine-learning boom, model developers have built an all-purpose digital oracle from a trove of big data
AI Weather Forecasting Can’t Replace Humans—Yet
GraphCast and other artificial intelligence-based forecasting tools offer a whole new way to predict the weather, but they have limits
What’s Missing from the Emoji Animal Kingdom?
In the digital age, some scientists argue the emojisphere should better represent Earth’s biosphere—tardigrades, flatworms and all
How Search Engines Boost Misinformation
Data voids in search results can lead down rabbit holes that bolster belief in fake news
Cats Play Fetch, Too—But Only on Their Own Terms
Retrieving isn’t just for dogs, but the emerging science of cat play can’t fully explain the feline phenomenon
Inside the Satellite Tech Revealing Gaza’s Destruction
Amid restrictions on optical satellite images, researchers have developed a radar technique to gauge building damage in Gaza
When It Comes to AI Models, Bigger Isn’t Always Better
Artificial intelligence models are getting bigger, along with the data sets used to train them. But scaling down could solve some big AI problems
Robotics ‘Revives’ a Long-Extinct Starfish Ancestor
Engineers and paleontologists teamed up to reconstruct an ancestor of starfish from the Paleozoic era and figure out how it moved
It’s Not All in Your Head—You Do Focus Differently on Zoom
Virtual meetings and video calls don’t quite stack up to in-person interaction—and a new study proves it
Biden’s Executive Order on AI Is a Good Start, Experts Say, but Not Enough
A new executive order signed this week sets the stage for federal AI standards and requirements and moves beyond previous voluntary agreements with AI companies