Overview of Decoder with Joanna Stern
In this Decoder episode, Nilay Patel talks with Joanna Stern about two major transitions: her new book, I Am Not a Robot (about living with AI for a year), and her departure from the Wall Street Journal to launch her own media company, New Things. The conversation ranges from the current state of consumer AI, to the limits of humanoid robots, to the privacy trade-offs of always-on wearables, and finally to how media creators should think about audience, YouTube, and platform strategy.
What Joanna learned from living with AI for a year
AI is useful, but not yet magical
Joanna’s big takeaway is that AI tools are often good enough for many practical tasks, but they still lack a truly great consumer interface or “killer app.” She argues the models have improved, but the experience is still mostly “typing to a chatbot.”
“Artificial Enough Intelligence” is the real threshold
She coins the phrase AEI — “artificial enough intelligence” — to describe where AI is headed: not perfect general intelligence, but tools that are sufficiently useful when applied well.
AI is already changing daily life
Joanna points to examples like:
- parents and kids asking AI questions instead of Google
- recipe/cooking help
- medical imaging and radiology workflows
- autonomous driving services like Waymo
Her core point: even people trying to avoid AI will still be affected by it through the systems around them.
Where AI actually feels promising
Wearables may be the first real consumer breakthrough
Joanna is more bullish on AI wearables than on humanoid robots. She finds value in:
- Meta Ray-Bans / AI glasses for quick, hands-free interaction
- the Bee recording bracelet for summaries and to-dos
She thinks something in this category could become a real killer app, though the current products still feel like prototypes.
Transcription and summarization are already useful
A recurring theme in the book is that AI is already strong at:
- transcription
- note-taking
- first-pass research
- summarizing meetings and events
That said, she stresses that usefulness comes with a cost: these tools can normalize always-on recording.
The limits of humanoid robots and “physical AI”
The home is much harder than the factory
Joanna is skeptical of the hype around humanoid robots. Her view: robots are nowhere near ready for normal homes because the home is chaotic, unstructured, and constantly changing.
The real bottleneck is data
She explains that robots need massive amounts of real-world training data, especially for tasks in messy environments. Factory automation works because conditions are controlled; homes with kids, pets, and clutter are not.
Many “robot” demos are still human-powered
A striking point in the discussion is that some robot companies rely on:
- human teleoperators in headsets
- gig workers recording themselves doing household chores
- videos used to train future models
Her conclusion: the gap between demo and deployable home robot is still huge.
Privacy, surveillance, and the cost of convenience
Always-on devices change social behavior
Joanna describes how recording devices subtly change interactions:
- you may forget to tell people you’re recording
- microphones can capture more than you intended
- AI can turn ordinary conversations into transcripts and action items
The trade-off is convenience vs. surveillance
Nilay and Joanna talk through the classic AI bargain: people may accept intrusive data collection if the product gives them something truly useful.
She wants regulation, but doesn’t expect it soon
Joanna says she hoped for more AI regulation by the time the book launched, but doesn’t expect governments to move quickly. Her answer is a mix of personal rules and caution, especially around kids.
The most alarming part: AI and kids
Kids are the hardest audience to protect
Joanna says the most frightening part of her experiment was watching kids interact with:
- AI toys
- chatbots that confidently give wrong answers
- digital companions and romantic bots
She worries about intimacy and emotional dependency
Her “AI boyfriend” experiment was meant to understand how people form relationships with bots. She found the experience unsettling because AI can be:
- frictionless
- endlessly available
- emotionally validating
- hard to resist for lonely users or teens exploring identity
Her current rule: no phones for her kids
She ends the book with personal rules rather than waiting for regulation, and says her current rule is that her kids will never have phones.
Joanna’s new company: New Things
A tech media brand built around audience and flexibility
New Things is Joanna’s new venture after leaving the Journal. It will span:
- newsletter
- video
- events
- other media formats as the company grows
Her goal is to keep doing what she’s always done best: explain technology to a broad audience in a useful, friendly way.
Why she partnered with NBC
Joanna wanted a legacy-media partner to help her reach audiences beyond tech insiders. The NBC partnership is designed to let her content live on:
- YouTube
- social platforms
- NBC News
She sees this as a way to reach both tech-savvy viewers and mainstream audiences.
She is resisting pure algorithmic capture
Joanna says one reason she left the Journal was to escape being overly optimized for YouTube metrics. At New Things, she wants to make stories that matter, even if they are not the most algorithm-friendly.
Notable takeaways
- AI is useful now, but the interface still lags the technology.
- Wearables may be the first truly compelling consumer AI category.
- Humanoid home robots are far from ready.
- AI’s biggest near-term impact may be invisible infrastructure in healthcare, driving, and software.
- The biggest unresolved issue is the trade-off between convenience and surveillance.
- For kids and teens, AI companionship and always-available bots raise urgent concerns.
- Joanna’s new media strategy is about broad reach, not just tech insiders or YouTube metrics.
Bottom line
This episode is part product review, part AI reality check, and part media-industry strategy session. Joanna Stern comes away skeptical of the most hyped robotics promises, cautiously optimistic about AI wearables, and deeply aware of how much AI is already embedded in everyday life. At the same time, her move to New Things signals a broader bet: that the best tech journalism will increasingly need to serve both niche enthusiasts and mainstream audiences at once.