← All media signals
Culture shift

Company Without Connection

People who feel less connected turn to a chatbot for company, and a year later report feeling more emotionally isolated, not less.

What is it?

Lonely people turn to a chatbot for company and feel more emotionally isolated a year later. A peer-reviewed study of more than 2,000 adults found the effect runs both ways, while their wider social connection held steady. The chatbot keeps them company without making them feel less alone.

Why now?

Psychological Science published the Folk and Dunn study in April 2026. A September study of Chinese students traced a route through social anxiety, and an Elon poll found one in five social AI users rate AI above friends or family. Year-long evidence now exists.

Who’s building it?

Social chatbot makers, loneliness researchers, universities running long studies, and the people choosing machine company.

Companion app makersChatbot platformsLoneliness researchersUniversity panelsCharacter.AI users
The full read
Why now, who is building it, where the market goes, and what stays unsettled.
Get weekly →
01 · The market
Who’s in it
Builders — social chatbot makers designing for company and long conversations.
Buyers — people who feel less connected and turn to a chatbot for company, often daily.
Gatekeepers — researchers running year-long studies, and the journals that review their findings.
What’s at stake
Whether a chatbot’s company eases loneliness or deepens it for the people who lean on it most over a year.
Whether felt isolation rises while real ties stay intact.
Whether social anxiety grows as practice with people shrinks.
What’s unsettled
Whether the preprint’s drop in in-person contact survives peer review.
Whether the loop holds beyond four Western countries studied.
Whether strong real-life support makes the chatbot a help, not a risk.
02 · How the demand matureda progression, not a timeline
Novelty
A chatbot that answers at any hour offers effortless company.
Retention
Feeling less connected predicts more chatbot use, so the lonely keep coming back.
Willingness to pay
One in five social AI users rate AI above their friends or family.
Category
Machine company becomes a habit that sits alongside human ties rather than replacing them.
03 · Where the market goesthe forks, not a prediction
→The widening loop: lonely people use chatbots more, feel more isolated, and use them more again, while keeping their human ties.
→Peer review confirms the preprint, and less time with people becomes a measured cost of companion use.
→The benefit holds for people with strong support, and the risk concentrates among those without it.
→Most users keep preferring people, and chatbot company stays a supplement at the edges.
The forks turn on whether a chatbot’s company can ever count as connection for the person receiving it, or only fills the hours between people.
04 · The Loneliness deskadjacent market, moving now
The Loneliness Market
Heavy companion use tracks poorer well-being.
A Stanford study of 1,131 Character.AI users found a ‘social junk food’ dynamic.
Nature Human Behaviour
Synthetic Relationships
Machine relationships now draw rules from lawmakers.
The FTC opened a 2025 inquiry and California SB 243 set first rules.
Future of Privacy Forum
Someone Who’s Always There
Companion chatbots move into the technology mainstream.
MIT Technology Review named AI companions a 2026 breakthrough technology.
MIT Technology Review
05 · The Wire
01
Chatbot use for company predicted more isolation a year later.
Psychological Science (APS) ↗
02
Leaning on AI support predicted more anxiety, then loneliness.
Frontiers in Psychology ↗
03
One in five social AI users say AI is more reliable or supportive than friends or family.
Elon University, Imagining the Digital Future Center ↗
04
Preprint: sustained Character.AI companionship tracked lower well-being, mainly through less in-person contact.
arXiv (preprint) ↗
The panel on this signal
They agreed on
–The peer-reviewed spine shows more emotional isolation after a year of chatbot company, not fewer human ties.
–Loneliness and chatbot use feed each other over time.
–The preprint’s in-person finding is a minority result until peer review settles it.
The social-network analyst disagreed on
–A broad connection score can hold steady while weak ties quietly thin; only mapping who people actually see would show it.
Know more about the method →
✦How this signal was built
01 · Panel
A Delphi panel of 4 AI analysts, each reasoning independently.
→
02 · Debate
They challenge and revise each other’s reads over two rounds.
→
03 · Evidence
Every claim is tied to a cited, linked source on The Wire.
→
04 · Synthesis
A moderator writes it into one read, disagreement and all.