Lissin

Podcast listening just hit an all-time high — and AI users are leading it

A record year for podcast listening surfaced a striking pattern — AI-tool users report much higher audio consumption than non-users — and two major 2026 surveys help explain what that gap does and doesn't mean.

Lissin Editorial9 min read

Edison Research and SSRS published their Infinite Dial 2026 report this year, and the headline audio numbers were the highest the study has recorded since it began tracking online audio and podcast use in 2000. That part is a straightforward story about a medium that keeps growing.

A stranger number sits inside the same dataset. People who use AI tools reported listening to online audio and to podcasts at meaningfully higher rates than people who don't use AI tools at all. That gap comes from one national survey, taken at one point in time, and a gap inside a single survey does not tell you that one habit causes the other, or that anything causes either one.

This piece works through what the Infinite Dial actually found this year, what the AI-user pattern does and doesn't show, and why a separate report on AI and news — the Reuters Institute's Digital News Report 2026 — argues against reading too much into that pattern. AI chatbot use for news is growing, but it is still a small share of how people get their news, and the data says so directly.

What the Infinite Dial actually recorded

The Infinite Dial has run since 2000, which gives it one of the longer trend lines in US media research. The 2026 wave found that 81 percent of Americans age 12 and older listened to some form of online audio in the past month, and 76 percent listened in the past week. Both figures are records for the study, according to coverage from National Public Media, which has reported on the Infinite Dial's findings for several years.

Podcast listening moved by similar margins. Fifty-eight percent of Americans 12 and older listened to a podcast in the past month, and 45 percent listened in the past week. The share of Americans who say they have ever listened to or watched a podcast rose seven percentage points in a single year — an increase National Public Media's coverage describes as the largest one-year jump the study has recorded, equivalent to roughly 20 million additional people. Consumption isn't even across age groups: people between 35 and 54 posted 68 percent monthly podcast reach, above the population-wide 58 percent figure. Listeners 55 and older, historically the slowest group to adopt online audio, saw their monthly listening rate climb from 52 percent in 2024 to 70 percent in 2026.

None of this distinguishes between formats the way a listener actually experiences them. "Podcast" in the Infinite Dial's framing covers a true-crime series, a daily news briefing, a comedy show, and a limited-run documentary under one umbrella. If you want a sharper cut between "a briefing you check most mornings" and "a show you subscribe to and work through episode by episode," that distinction — and why it changes what growth numbers actually mean for a listener's day — is covered in Audio news briefing vs. news podcast. The record-setting numbers above describe a broad category, not a single kind of listening habit.

The AI-user gap inside the same survey

Edison Research and SSRS also broke out results by whether a respondent uses AI tools, and the contrast was larger than most of the demographic splits in the report. Eighty-seven percent of AI users said they had listened to online audio in the past week, compared with 61 percent of people who don't use AI tools. On podcasts specifically, more than half of AI users were weekly podcast listeners, versus roughly a third of non-users.

Both of those AI-user figures are worth sitting with for a second, because they're not small. A 26-point gap in weekly online-audio listening, and a gap of "more than half" versus "roughly a third" on weekly podcasts, is a bigger split than most single demographic variables produce in a survey like this one. It's also worth naming what the finding is not. This is a single cross-sectional survey of 2,050 people age 12 and older, fielded in January 2026 using probability-based sampling and weighted to the US population. It is not a controlled study, it does not track the same people over time, and it does not ask which habit came first. A survey that finds two behaviors clustered in the same respondents has found a correlation, and a correlation is the start of a question, not the end of one.

Three reasons this correlation might exist — none of them proven

It's tempting to read the AI-user gap as "AI recommends podcasts to people" or "podcast fans are early AI adopters," but the Infinite Dial data doesn't support either claim specifically. What follows are plausible explanations, offered as speculation, not findings.

One possibility is a general adoption-comfort effect. People who have already picked up one new digital habit — trying a chatbot for the first time, working it into how they search or write — may simply be more willing to try other new digital habits, podcasts and streaming audio included. On this account, AI use isn't causing podcast listening; both are downstream of the same trait, something like openness to new tools.

A second possibility is that both habits fit the same moments in a day. Podcasts are built for listening while doing something else — commuting, cooking, exercising — and a lot of AI tool use is also opportunistic, squeezed into small windows between other tasks. People who have organized their day around fitting in audio, or fitting in quick digital tasks, may simply have more of both kinds of moments available, or may be more inclined to fill them.

A third possibility is demographic overlap rather than any direct connection at all. Heavier AI tool use skews toward younger, more online-native audiences, and those same cohorts already show higher baseline podcast consumption in the Infinite Dial's own numbers, independent of anything to do with AI. If that's most of what's driving the gap, then age and general digital engagement — not AI tool use as such — would be doing the real work, and the AI-user split would be picking up a proxy rather than a cause.

The honest summary is that the Infinite Dial gives us a real, sizable gap and no way to adjudicate between these explanations, or others, from this data alone. Anyone citing this pattern as proof that AI tools drive podcast adoption is saying more than the report supports.

What the Reuters Institute found about AI and news specifically

A separate study puts a useful boundary around all of this. The Reuters Institute's Digital News Report 2026 found that 10 percent of people surveyed used AI chatbots for news in the past week, up from 7 percent the year before — real growth, but growth from a small base, and a smaller number than the Infinite Dial's audio figures by a wide margin. The report's own segmentation also found that AI chatbot news users skew toward people who were already highly engaged with news: 38 percent of them fall into the report's "news lover" segment, compared with 22 percent of respondents overall.

More important for keeping this piece honest: the same report finds that only around 1 percent of respondents name AI chatbots as their main source of news. That is nowhere near television, social media, or news websites and apps as a primary channel. Ten percent trying AI chatbots for news at least once a week, and 1 percent treating one as their main source, describes a habit that is growing and still marginal at the same time — both things are true, and neither cancels the other out. Nothing in this year's Reuters Institute data supports a claim that AI has replaced other ways people get their news; if anything, the 1 percent figure argues the opposite.

That gap between "trying" and "relying on" is also the space where earlier personalized-news-audio products ran into trouble — plenty of past attempts got people to sample something new without giving them a reason to keep it as a habit. The graveyard of personalized news audio covers what typically went wrong in that category before this current wave of AI-audio tools.

The advertising market backs up the audio trend independently

Self-reported survey data is useful, but it isn't the only signal that audio is growing. Money tends to follow attention with a lag, and the digital-audio advertising market moved in the same direction as the listening figures. According to IAB and PwC data reported by Radio Ink, US digital audio advertising revenue reached $8.4 billion in 2025, up 10.2 percent from $7.6 billion in 2024. Podcast advertising specifically brought in $2.862 billion in 2025, a 17.6 percent increase, putting the category close to the $3 billion mark.

That podcast growth rate has slowed from its earlier pace — the same reporting notes a 72 percent spike in 2021 and a 26 percent gain in 2022 — but the category has still posted double-digit growth in four of the last five years. Advertisers don't fund a shrinking audience at that scale for long, so the ad-revenue trend is a reasonable, independent check on what the listening surveys are reporting. It doesn't touch the AI-user question at all, but it does support the more basic claim that podcast listening's record highs are a real market shift and not a one-year statistical blip.

Sitting between two curves that are both climbing

Two separate trend lines are both pointing up this year: general audio and podcast consumption, and general comfort with AI tools. The Infinite Dial's AI-user split shows those two curves overlapping in the same people more than chance alone would predict, without settling why. The Reuters Institute data adds a necessary caveat to the AI side of that overlap — chatbots are gaining ground in news specifically, but from a small base, and nowhere close to displacing the outlets and formats people already rely on.

A personalized audio app sits at the intersection of those two growing habits without needing either one to have caused the other. People are listening to more on-demand audio than the Infinite Dial has ever recorded, and more people than before are at least willing to try an AI tool for some part of their day. Lissin turns a topic, a question, a mood, or a source into a narrated show built around what you actually asked for, which is one way of meeting people where both of those habits already are, rather than betting that either trend will keep climbing on its own. We don't have usage numbers of our own to add to this data — what we have is a product built for exactly the overlap these two reports describe. Explore audio on Lissin to see what a show built around your own topics sounds like.

Sources

Related reading