Dead Internet Theory: Fiction, or Future?
summary
Makes a concrete claim: much of online content risks being auto‑generated and consumed by bots, backed by striking examples — 1000 SIM cards sending millions of messages a day and alleged dataset theft of 82 terabytes of books — and concrete proposals: use DevOps-style four security gates (raw‑data review, PII/toxicity removal, adversarial/supply‑chain testing, red‑teaming/evaluation), mandatory AI tagging metadata, and time‑based penalties; argues regulators already push tagging but companies sidestep it, no full solution yet.
AI-written summary · about
chapters (25)
- 0:00Dead Internet Theory intro
- 1:31Origins of Dead Internet Theory
- 2:35Attention economy & ads
- 4:44AI as 2025’s engine
- 6:09Corporate responsibility & openness
- 7:35Trust collapse in AI era
- 9:24Legal double standards in AI
- 11:33Solutions to Dead Internet
- 14:00Humans in the AI loop
- 16:10Four security gates for AI
- 17:14Why fines fail, time penalties work
- 18:00Need for AI content tagging
- 19:24EU AI Act & corporate avoidance
- 21:50AI’s fork in the road
- 22:54Q&A
- 23:21Hybrid content & AI tagging
- 24:05Enforcing AI penalties
- 25:57AI in 2027 & global duty
- 26:22Penguin slide meaning
- 27:07AI’s impact on cognition
- 29:54Forced AI adoption & layoffs
- 31:19Gargoyles vs grotesques
- 32:02AI malware threats
- 34:06Privacy & internet future
- 36:14Closing remarks
related talks
49mGenAI and Cybersecurity: Safeguarding the Digital Age
18mDefending the unknown using AI and cybersecurity
47mSecuring Sexuality: Rewiring the Most Intimate Connections
25mThe Ethics of the Internet of Things
01h 24mOffensive Cybersecurity by AI Promises and Pitfalls
45mSecuring The Future
58mSecuring a World of Physically Capable Computers
01h 07mUnpoisoned Fruit: Seeding Trust into a Growing World of Algorithmic Warfare