a16zMarc Andreessen on how the internet changed news, politics, and outrage | The a16z Show
CHAPTERS
Why “Monitoring the Situation” is the internet’s native mode
Marc opens with the idea that modern news consumption is driven by what’s newest and most outrageous—not necessarily what’s most important. He frames viral outrage as a fast, repeatable cycle and hints at a contrarian claim: online conflict may reduce real-world violence.
- •“News” optimizes for novelty and outrage, not importance
- •Viral meme blowups behave like short panic cycles
- •Online anger may substitute for physical confrontation
- •Political violence (measured) is historically low in the West
- •Sets the stage for media as a behavior-shaping technology
CNN’s founding idea: “Random Ammonium” and 24-hour current-thing coverage
Marc recounts CNN’s early business plan via Reese Schonfeld and Ted Turner: a 24-hour channel that locks onto whatever is most compelling right now. The concept—“randomonium”—means airing incomplete, real-time fragments and correcting later.
- •Ted Turner + Reese Schonfeld’s 24-hour news bet
- •“Randomonium” = continuous coverage of the current thing
- •Real-time, fragmentary info is a feature, not a bug
- •Gulf War as CNN’s breakthrough ‘glued to the screen’ moment
- •CNN later struggled to repeatedly recreate that intensity
How the internet reinvented randomonium (social + streaming)
He argues the internet fully re-created CNN’s core loop, but with far more participants and speed—especially on X and across streaming platforms like YouTube and Twitch. “Monitoring the situation” becomes not just a show format but a meme-driven behavior pattern.
- •The internet operationalizes continuous ‘current thing’ tracking
- •X and livestreaming make the loop always-on and participatory
- •Memes become the interface layer for news events
- •The audience is no longer passive—everyone can broadcast
- •“Monitoring the situation” as a cultural ritual
Why current things accelerate: the global village meets the meme machine
Marc uses Marshall McLuhan to explain two drivers: the internet turns the world into a ‘global village’ with no privacy, and it converts every event into a memetic format. The result is brain-melting proximity to billions of people plus a built-in virality engine.
- •McLuhan’s ‘global village’ as a stressor (not a utopia)
- •Dunbar’s number vs always-on exposure to everyone
- •The medium shapes behavior (“the medium is the message”)
- •The internet forces events into meme-friendly packaging
- •Constant connectivity increases intensity and frequency
“If it’s on the internet, it’s a viral meme”: outrage half-lives and election irrelevance
He proposes the internet equivalent of “TV turns everything into a show”: online turns everything into a viral meme that sparks moral panic. These cycles spike and decay with a short half-life (~2.5 days), making long-range political prediction unreliable because attention resets repeatedly.
- •Everything becomes meme → outrage → tribal conflict
- •Typical outrage cycle lasts ~two and a half days
- •New current thing replaces old ones without resolution
- •Politics/elections hinge on what’s salient at the moment, not months earlier
- •Media dynamics operate like “autopilot” until displaced by a new medium
Is this better or worse? More truth, more lies—and less street violence
Marc argues social media increases both truth and deception by removing gatekeepers while enabling ‘ops.’ His deeper defense is historical: older media ecosystems often helped catalyze physical violence, whereas online ‘virtual combat’ may reduce real-world harm by absorbing aggressive energy.
- •Gatekeeper collapse enables direct truth-telling and direct lying
- •“Spot the ops” becomes a native online pastime
- •Historical media technologies often correlated with violence (posters, radio, TV)
- •Claim: political violence is low while rhetorical violence is high
- •Online conflict may substitute for street-level conflict
Political polarization is not new: the past was harsher than we remember
He pushes back on nostalgia by citing intense culture conflict throughout U.S./Western history, including riots, forced desegregation, Cold War paranoia, and even dueling among elites. TV-era sitcoms like All in the Family and Family Ties are used as windows into earlier culture wars.
- •Nostalgia ‘sandpapers’ the past; conflict is historically constant
- •All in the Family and Family Ties as culture-war artifacts
- •Examples: Vietnam, civil rights clashes, Cold War, world wars
- •Elite dueling and street-fighting political movements were once normal
- •Current rhetorical heat doesn’t imply unprecedented societal fracture
Depression stats, benefits incentives, and “suppressed volatility” under centralized media
On rising depression claims, he suggests cross-country differences may reflect incentive structures in benefits and accommodations rather than a uniform tech-driven effect. He also argues the post–Cold War era had ‘suppressed volatility’ because media was unusually centralized, masking the natural fragmentation of public debate.
- •Mental health trend lines vary by country; incentives may distort reporting
- •Doesn’t fully accept a simple ‘smartphones caused depression’ story
- •Trust in institutions falling is long-running (since ~1970)
- •Media centralization peaked mid/late 20th century (few networks/newspapers)
- •Earlier eras had many partisan papers—fragmentation is historically normal
What makes a “current thing”: outrage triggers, tribes, and moral panic mechanics
Marc lays out the attributes of events that go viral: they must provoke emotion, form opposing moral tribes, and support a narrative of societal decay. He argues ambiguity helps—events that are hard to verify are easier to fight about and better fuel for ongoing conflict.
- •Outrage activation is the primary ‘eligibility requirement’
- •Tribal formation: sides can square off with moral claims
- •Moral panic framing (“society is going to hell”) drives escalation
- •Ambiguity/surrealism boosts virality by enabling argument
- •Underlying issues matter more than the specific incident details
Truth doesn’t scale with virality: small incidents can drive massive outrage
He emphasizes that neither factual certainty nor magnitude determines attention. Personalized stories can generate more intensity than large-scale harms, and many viral videos start mid-incident, stripping context and increasing misinterpretation and scapegoating.
- •Truth/falsity often doesn’t change the outrage outcome
- •Outrage doesn’t scale with number of people affected
- •Personalization can intensify reaction (“tragedy vs statistic”)
- •Viral videos often begin mid-event—context is missing by design
- •Key discipline: ask what happened before the clip started
Ops, availability cascades, and ‘dark money’ influence without disclosure
Marc offers a nuanced model: many events are organic, many are ops, and some start as ops then become real movements. He uses Kuran/Sunstein’s ‘availability cascade’ and ‘availability entrepreneurs’ to explain how attention is deliberately seeded, then points to a legal gray zone where paid moral/political influencing can be undisclosed.
- •Ops can be real, organic, or hybrids that ‘become real’
- •Availability cascades: attention spreads via availability bias
- •Availability entrepreneurs seed and amplify specific focal events
- •Rosa Parks example: organized action can catalyze genuine change
- •Undisclosed paid advocacy is legal when it’s neither product ads nor candidate donations (creating ‘dark money’ influence)
Legacy media vs new media: collapsing trust, the barbell of TikTok and 3-hour podcasts
He cites falling trust and shrinking audiences for legacy outlets while podcasts, streaming, and social platforms rise. He rejects the idea that everything becomes shallow: the new ecosystem is ‘barbell-shaped’ with both short-form video and ultra-long-form podcasts, plus practitioner-led media and AI ‘deep research.’
- •Trust in legacy institutions/media keeps hitting new lows
- •Old media businesses collapse unless they adapt
- •New media growth: podcasts, livestreaming, social distribution
- •Barbell effect: TikTok short-form + multi-hour (even 10-hour) podcasts
- •Rise of practitioner media and deep, tutorial-like explanations (including AI tools)
The first true internet election hasn’t happened yet—and neither has the true internet candidate
Marc argues 2008 (online fundraising), 2016 (social media), and 2024 (podcasts) show internet influence, but not a fully internet-native election. Trump is a hybrid—TV-obsessed yet social-native—and the future will bring a leader who ignores TV entirely and wins purely through internet dynamics.
- •Past elections show partial internet effects, not full transition
- •Trump as hybrid: shaped by TV and able to steer TV via social posts
- •Deepfakes aren’t new—societies have always had large-scale misinformation
- •Prediction: a truly internet-native candidate/president is coming
- •Speculation on timing: possibly the 2032 cycle
