The Most Useful Professor You've Never Had
Carl T. Bergstrom doesn't hold a chair in philosophy or media studies. He's an evolutionary biologist at the University of Washington who studies how information flows through biological and social systems. But somewhere between analyzing pathogen genetics and corvid behavior, he became one of the most important voices in America's argument with its own attention span.
The course he co-created with Jevin West in 2017, "Calling Bullshit: Data Reasoning in a Digital World," went viral before the first lecture. The syllabus alone was covered by CNN, The Guardian, NPR, and a dozen other outlets. People weren't laughing at the name - they were relieved someone had finally given the problem a name. That course has since been replicated at hundreds of universities. The book that followed it, published by Random House in 2020, became a handbook for a generation trying to survive the information apocalypse.
But Bergstrom isn't a media celebrity or a tech pundit. He's still very much a working scientist. He publishes in Nature. He teaches evolution. He photographs crows. And he's doing something harder than writing think-pieces: he's building actual frameworks for thinking about bad data, frameworks rigorous enough for statisticians but accessible enough for anyone who's ever been lied to with a pie chart.
"Bullshit is language, statistical figures, data graphics, and other forms of presentation intended to persuade or impress an audience by distracting, overwhelming, or intimidating them with a blatant disregard for truth, logical coherence, or what information is actually being conveyed."- Carl T. Bergstrom, Calling Bullshit (2020)
This is not the usual academic complaint about fake news. Bergstrom's definition of bullshit extends Harry Frankfurt's philosophical concept into the world of data visualization, scientific publishing, and algorithmic recommendation. His argument is that the modern information environment is less defined by lies than by impressively packaged nonsense - and that the cure is not fact-checkers but better-educated citizens.
What Bergstrom means by "Bullshit"
Not lying - lying requires knowing the truth and choosing otherwise. Bullshit, by contrast, is indifferent to truth entirely. It's content designed to persuade or impress with no regard for whether it is actually accurate. The digital era's contribution: you don't need words anymore. A misleading graph or a cherry-picked statistic works just as well - and reads as authoritative.
From Genomes to Graphs: How a Biologist Became a Data Cop
Bergstrom arrived at Stanford in 1993 with a Harvard degree and a head full of evolutionary theory. His dissertation, supervised by Marcus Feldman, was about information encoding in biological systems - how organisms communicate, how signals evolve, and how evolutionary pressures shape the reliability of those signals. It's genuinely arcane stuff.
The leap from genome signaling to social media misinformation is less random than it looks. The underlying question is always the same: how does information move, why do certain signals survive, and what makes a signal honest rather than deceptive? In biology, deceptive signals get selected against over time. In the information economy, they get amplified.
By 2007, Bergstrom had already translated this instinct into something concrete. With Jevin West, he built the Eigenfactor system - a way to rank academic journals not just by how often they were cited, but by the prestige of the journals doing the citing. Think of it as Google's PageRank for academic publishing. Clarivate integrated Eigenfactor into its Journal Citation Reports in 2009. It's now an industry standard.
The same logic animates everything Bergstrom does: not all information is equal, sources matter, and the structure of the information network shapes what gets amplified. He was working on the architecture of bad information long before "misinformation" became a cable news segment.
A Field Guide to the Types of Bullshit
Misleading Visualization
Charts that start at non-zero baselines, maps that use absolute numbers instead of rates, graphs with no axes.
Spurious Precision
Claiming to know something to four decimal places when the underlying data doesn't support one.
Causation Confusion
Presenting correlation as if it were causation - the most beloved rhetorical move in advertising and policy alike.
Selection Bias
Citing the studies that support the point you want to make, ignoring the preponderance of evidence.
AI Hallucination
LLMs generating authoritative-sounding facts with no grounding in reality - the newest and most scalable form.
Narrative Override
When a compelling story makes an audience stop checking whether the data actually says what the storyteller claims.
Calling Bullshit: The Course That Went Viral Before It Started
In January 2017, Bergstrom and West posted the syllabus for a new UW course. That's it - just a syllabus. Within days it was being shared across every platform that mattered, translated into multiple languages, and covered by international press. The syllabus read like a manifesto: a systematic curriculum for recognizing the machinery of deception in modern data presentation.
BIOL/INFO 270: Data Reasoning in a Digital World
Goal: Students learn to spot, analyze, and dismantle misleading claims disguised as data.
Approach: Case studies from science, media, politics, and advertising.
Website: callingbullshit.org - lectures freely available to anyone.
Status: One of UW's most popular courses; replicated at hundreds of institutions globally.
The course materials are free. The lectures are on YouTube. That was a choice, not an accident. Bergstrom and West believed the problem they were addressing was civic, not just academic, and that a course locked behind tuition walls wouldn't be enough. This is the move of someone who actually believes what he's teaching about information access.
The 2020 book that followed became a commercial success and a cultural artifact. It's been assigned in business schools, journalism programs, medical schools, and philosophy departments. It's probably in a few military and intelligence training programs too, though those don't tend to show up on Amazon reviews.
"The 'Calling Bullshit' course went viral because the syllabus alone was covered by CNN, The Guardian, and NPR - before a single student had attended a single lecture. Sometimes the promise of rigor is enough to make people feel hopeful."
- From media coverage of the course launch, 2017Now He's Coming for the Chatbots
Bergstrom's newest project is called "Modern-Day Oracles or Bullshit Machines?" and if the title sounds familiar, that's the point. The free 18-lesson online course, co-developed with Jevin West, launched in early 2025 and was adopted at 130 universities within weeks. It covers how large language models actually work, when to trust them, when not to, and how to see through the hype that surrounds them.
The course is available at thebullshitmachines.com - another free resource, another bet that civic education doesn't belong behind a paywall. Bergstrom's position on AI is characteristically precise: he's not anti-AI, he's anti-AI-bullshit. The distinction matters. LLMs are genuinely powerful tools. They are also very good at sounding more confident than they should be about things they don't actually know. In that sense they are the most perfect bullshit machine ever built.
"You can't do good public health work and you can't do good science under a roof that is owned by someone who's trying to undermine it."- Carl Bergstrom, on leaving Twitter/X, 2022
His migration away from major platforms is itself a data point. He was among the first prominent scientists to leave Twitter when Elon Musk acquired it in 2022, explicitly citing the incompatibility between the platform's new direction and the practice of science. He left Instagram in January 2025 for similar reasons. He now posts primarily on Bluesky, where he can be found at @carlbergstrom.com. The pattern is consistent: Bergstrom makes platform choices on principle, not convenience.
The Great Platform Migration
Bergstrom has been unusually public about why he leaves platforms - citing each owner's specific actions as incompatible with science communication. His bird photography Instagram (@ctb_birds, 2,214 followers) went dormant when he left Meta in early 2025.
Still a Scientist, Actually
It would be easy to mistake Bergstrom for a science communicator who happens to have academic credentials. He's something more specific: an active researcher whose public work is an extension of his scientific questions, not a departure from them.
His research spans evolutionary biology, information theory, scholarly communication, and antibiotic resistance. He's published on the evolution of sex, the ecology of pathogens, how scientific publishing incentives shape the research that gets done, and (with recent urgency) the integrity of scientific literature in the age of AI-generated fake papers. His papers appear in Nature, Science, PNAS, and Evolution. He is also, genuinely, a professor who teaches introductory biology.
The 3rd edition of his evolution textbook, co-authored with Lee Alan Dugatkin and published by W.W. Norton in 2023, is described as the most current introduction to evolutionary biology - with an emphasis on anthropogenic evolution and data literacy. That last phrase is the tell. Even in a textbook about natural selection, he's threading in the data literacy argument.
His appointment as External Professor at the Santa Fe Institute, the famous complexity science research center in New Mexico, reflects how his peers categorize him: not as a popularizer, but as a serious thinker working at the intersection of biology, information, and society.
The Crow Photographer Problem
One of the odder facts about Carl Bergstrom is that he is genuinely excellent at photographing birds, particularly corvids - crows, ravens, and jays. His Instagram account @ctb_birds, which he ran until leaving the platform in 2025, featured hundreds of high-quality wildlife photographs. He also maintains a Flickr archive and a bird photography gallery on his personal site.
This is not incidental color. Corvids are famously intelligent - among the most cognitively complex non-human animals. They use tools, recognize individual human faces, form long-term alliances, and appear to understand causality in ways that other birds don't. A biologist who studies information systems and chooses to spend his leisure time photographing the most intellectually interesting birds on Earth is either very deliberate or very consistent. Probably both.
There's also something pointed about a misinformation researcher being drawn to animals with a reputation for cleverness and occasional deception. Corvids are known to feign caching food when they know they're being watched. The ecology of honest signaling, in a different form.