When people talk about the filter bubble, they almost always mean an algorithm. Facebook's feed. TikTok's For You page. Google News deciding which headline you see first. The mental picture is a black box quietly narrowing your view of the world based on what you've clicked on before.
That's a real thing. But it's not the biggest filter most readers are walking through every day.
The biggest filter is the outlet you opened first this morning.
If your default is The New York Times, you got one version of the world. If it was Fox News, you got a different one. If it was The Wall Street Journal, you got a third. None of those outlets had to lie or distort anything to leave you with very different impressions of the same news cycle. They just had to pick which stories to lead with, which facts to put in the second paragraph instead of the eleventh, and which voices to quote.
That is the single-source trap, and it has nothing to do with an algorithm.
What "filter bubble" actually means
Eli Pariser coined the phrase "filter bubble" in 2011 to describe what algorithmic personalization does to news consumption — the worry that two people Googling the same term would get two different sets of results based on prior behavior. Fifteen years later, the phrase has been compressed into shorthand for "social media is making us stupid."
The compression is misleading. Long before any feed sorted itself, the dominant filter on what Americans saw as "the news" was editorial selection — what an outlet decided to cover, how prominently, in what order, with whose framing. The internet did not invent this. Algorithms did not invent it. What changed is that the cost of comparing outlets dropped to zero, and almost nobody actually does the comparison.
Reading one outlet was already a filter bubble. The algorithm just made it lazier.
The single-source trap, illustrated
Today, Nebraska becomes the first state to implement Medicaid work requirements under the One Big Beautiful Bill Act. NPR and The Hill both covered it. Their facts overlap almost entirely — same law, same date, same affected population. The framing does not.
NPR's lead is enrollees: "People are worried." The story foregrounds the experience of people trying to document compliance — what proof they need, who might lose coverage, how the rollout is hitting in practice.
The Hill's lead is implementation: Nebraska "faces challenges" and is "racing ahead" of the January 2027 federal deadline. The story foregrounds the state's policy machinery — how Nebraska is moving early, what other states will need to do, what the operational lift looks like.
Neither outlet is wrong. Neither is even particularly biased in any obvious way. But if NPR is your only Medicaid source today, your dominant impression is people are at risk. If The Hill is your only Medicaid source today, your dominant impression is Nebraska is moving fast on a federal mandate. Same facts. Two completely different stories.
That is the single-source trap working at low intensity — both outlets covered the story, neither hid anything, and you'd still walk away with a slanted picture from either one alone.
When one side stops covering a story altogether
The high-intensity version is when one side simply doesn't run a story. We call these blind spots, and they appear on both ends of the spectrum.
This week, Wyoming councilman Troy Bray posted "hang bad judges" on social media after an abortion ruling. Two outlets covered it: the New York Post and Fox News. Zero center, center-left, or left outlets ran a word. If your news diet is left-leaning, this story does not exist for you. (Notably, the two right-leaning outlets that did cover it didn't name the ruling, the council, or whether Bray had responded — so even readers who saw the story got a thin version of it.)
Pull the lens to the other side. This week, President Trump told a reporter he was considering withdrawing US troops from bases in Italy and Spain, then ordered a review of the entire US military presence in Europe. The Guardian, Fortune, Military Times, and Defense News covered it. Zero center, center-right, or right-leaning outlets in our index did. If your news diet is right-leaning, you missed a potentially major shift in NATO posture. The Guardian also added a framing detail none of the other coverage emphasized — that Trump raised the withdrawal idea after being prompted by a right-wing reporter, which changes whether this reads as policy or performance.
This is what bias by omission actually looks like in practice. Not a single outlet lying. A whole side of the spectrum quietly leaving a story off the page.
Five outlets, five frames: the Iran War Powers deadline
The strongest argument for reading multiple sources is not that any one of them is biased. It's that each one is partial, and partial in different directions.
The 60-day War Powers clock on Iran ran out this week. Five outlets covered it — one from each lean band in our index. The Intercept (left), TIME (center-left), Newsweek (center), The Hill (center-right), and Newsweek (right) all ran the story. The angles diverged sharply.
The Hill led with the political math: Republican support for the war is eroding even as Trump maintains the naval blockade. Newsweek's center-coded coverage took the constitutional angle: Congress is abdicating its role under the War Powers Resolution. The Intercept and TIME emphasized civilian and diplomatic consequences. The right-coded Newsweek piece put the same constitutional charge in a different voice and audience.
Read any one of those and you have a version of the War Powers deadline. Read all five and you have something closer to the version — politics, constitution, diplomacy, and accountability all visible at the same time. That is the gap a single-source diet leaves you in. Not a lie. A missing room in the house.
How to actually escape the single-source trap
The fix is unglamorous: read across the spectrum, and read for what each outlet chose to lead with, not just what they reported.
A few habits that work:
Read the same story in three outlets — one left, one center, one right — before forming an opinion. Pay attention to who they quote, what's in the headline, and what's pushed below the fold. The differences are usually the story.
Look for what's missing. If a story you saw on Fox News doesn't appear on NPR, or vice versa, that's a signal worth investigating. Our primer on news blind spots walks through how to spot them in real time.
Treat single-source confidence with suspicion. If you only ever read one outlet and you feel certain about every story you encounter, that certainty is doing some work that the reporting can't actually justify. The why-source-count-matters argument holds even for outlets you trust.
This is the entire point of what we built Signal/noise to do — show you all the lean bands at once, with the framing differences visible, so the comparison takes seconds instead of an hour. If you want to see how our approach stacks against other multi-source readers, the Ground News deep-dive covers the differences in detail.
Algorithms are a real filter. They are also the smaller one. The one that costs nothing to break out of, and that almost nobody bothers to break out of, is the one you walked into the moment you decided which outlet to read first.