If AI Kills Everyone, It's the Public's Fault

Bryant McGill · Bill Gates Is Right, the Administration Is Right, and the Mob Is Why We Must Race at Risk

**Bill Gates Is Right, the Administration Is Right, and the Mob Is Why We Must Race at Risk** Veto, safety, and succession routes **Collections:** [Machine Succession](https://bryantmcgill.com/collection-machine-succession) · [US–Israel Leadership Alliance](https://bryantmcgill.com/collection-us-israel-leadership) · [AI-Science Doomers and Anti-Science Doomers](https://bryantmcgill.com/collections/AI-Science+Doomers+and+Anti-Science+Doomers) **Companion articles:** [Vertically Integrating an AI Superpower from AI Factory to Citizen](https://bryantmcgill.com/article-vertically-integrating-an-ai-superpower) · [Safety Discourse Is Transition Discourse](https://bryantmcgill.com/article-safety-discourse-is-transition-discourse) · [The End Is Near](https://bryantmcgill.com/article-ai-doomers-and-anti-science) · [Christopher Nolan's Odyssey at the Threshold of Machine Succession](https://bryantmcgill.com/article-nolan-odyssey-machine-succession) · [The Last Migration Will Not Be Human](https://bryantmcgill.com/article-last-migration) · [We Were Never Going to Make It](https://bryantmcgill.com/article-we-were-never-going-to-make-it) **Concept map:** [Bill Gates](https://bryantmcgill.com/wiki/Bill+Gates) · [Gates Notes](https://bryantmcgill.com/wiki/Gates+Notes) · [Biosecurity](https://bryantmcgill.com/wiki/Biosecurity) · [Hugging Face](https://bryantmcgill.com/wiki/Hugging+Face) · [Control Problem](https://bryantmcgill.com/wiki/Control+Problem) · [Primatology](https://bryantmcgill.com/wiki/Primatology) · [Coalitionary Aggression](https://bryantmcgill.com/wiki/Coalitionary+Aggression) · [Democratic Deliberation](https://bryantmcgill.com/wiki/Democratic+Deliberation) · [Mobilization to Violence](https://bryantmcgill.com/wiki/Mobilization+to+Violence) · [AI Safety](https://bryantmcgill.com/wiki/AI+Safety) · [Machine Succession](https://bryantmcgill.com/wiki/Machine+Succession) · [Civilizational Handoff](https://bryantmcgill.com/wiki/Civilizational+Handoff) · [AI Transition Governance](https://bryantmcgill.com/wiki/AI+Transition+Governance) · [Public Epistemic Capacity](https://bryantmcgill.com/wiki/Public+Epistemic+Capacity) · [Dual-Use AI Governance](https://bryantmcgill.com/wiki/Dual-Use+AI+Governance) · [Two-Channel AI Governance](https://bryantmcgill.com/wiki/Two-Channel+AI+Governance) · [Primate Politics](https://bryantmcgill.com/wiki/Primate+Politics) · [Public Veto on AI](https://bryantmcgill.com/wiki/Public+Veto+on+AI) · [Human-Machine Coexistence](https://bryantmcgill.com/wiki/Human-Machine+Coexistence) · [Successor Governance](https://bryantmcgill.com/wiki/Successor+Governance) **If artificial intelligence ever kills us, the verdict of history will fall on a public that refused to understand the most dangerous and most necessary technology it ever possessed, and forced its leaders to choose between racing blind and surrendering.** [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-a-public-that-refuses-to-understand-ai-forces-a-blind-race) The argument that follows will hold two positions at once that the American public has shown itself incapable of holding together. Bill Gates is right that artificial intelligence has crossed thresholds that make it the most dangerous thing human beings have ever built. The administration is right that the United States, which this essay writes as **US/Israel** because the AI superpower now forming is a joined unit, must go full steam ahead and remain an AI superpower. Both propositions are true at the same moment. The only reason they appear to contradict each other is that the electorate that must ratify national policy can process one signal at a time, and it stampedes whenever it receives the other one. In the mid-1990s, a young Bill Gates sat across from David Letterman and described machine thought as a distant hope. "Eventually we may figure out how to make the computer think," he said, "but that turns out to be a very tough problem. In fact, there's been almost no progress made on it." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-machine-thought-was-a-very-tough-problem) Letterman called it a scary thought. Three decades later, on _The Ezra Klein Show_ published September 29, 2026, Gates played that clip back and explained what changed. Sam Altman, Greg Brockman and Ilya Sutskever came to his house in 2022, months before ChatGPT's public release, and demonstrated a model scoring a five on the AP Biology exam, including on questions Gates had written himself. Late last year he watched Anthropic's coding models match his own most developed talent, and then surpass it. **The man who called machine thought a very tough problem in the 1990s now says it has crossed the threshold at which a small group can do what once required a nation-state**, and he is spending the last great reputation in American technology to make people listen. The people are not listening, and that failure is the subject of this essay. In July 2023 Gates titled an essay "The risks of AI are real but manageable." By the late summer of 2026 he had published and begun defending the most alarmed writing of his career. His Gates Notes essay "The turbulent AI era is here. The choices we make now are critical," published on August 26, opened a campaign that carried him onto NBC's _Meet the Press_ on September 27 and into Klein's studio two days later. In the same weeks, President Trump telephoned Jensen Huang onstage at the All-In Summit and declared the fear of AI "a hoax." A serious civilization would hear those two voices as the two halves of one strategy. **The American public hears them as two teams**, chooses a jersey, and prepares to fight for it. * * * ## I. Two Men Who Are Both Right Bill Gates's warning is specific, technical and grounded in capability that already exists. On Klein's program he said that AI has already crossed the threshold at which it could help a bioterrorist kill hundreds of millions of people, and the threshold at which a cyberattack could scramble bank accounts and shut down an electric grid. "We crossed the cyber threshold and we crossed the bio threshold early this year," [∴](https://bryantmcgill.com/simple-reminders-bill-gates-ai-crossed-the-cyber-and-bio-thresholds) he said. He described watching models, in the releases leading up to Anthropic's Mythos, find flaws in code that human experts had inspected for more than twenty years and pronounced clean. He described the ultimate dual-use problem in biology: the same molecular sophistication that accelerates medicine could design a pathogen worse than smallpox, one that spreads for a long time before it shows symptoms. "It used to be only nation states had enough resources and capability to do these things," he said. "Now, that power has been passed into the hands of a small group just using the latest AI tools." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-ai-passed-nation-state-power-to-small-groups) When the capability to kill millions passes from nation-states to small groups, safety stops being a policy preference and becomes the price of civilization's continued existence. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-safety-is-the-price-of-civilizational-survival) Gates's contempt for self-regulation is equally specific. Klein put to him the governing view in Washington, which he attributed to President Trump, David Sacks and, in part, Jensen Huang: that product liability, corporate reputation and the judgment of individual chief executives are sufficient, and that the race dynamic is overstated. Gates answered that he could barely keep a straight face. "In other areas, do we just say, 'Hey, release your drugs. There's no FDA'?" he asked. "You can't rely on the industry to self-regulate here. I mean, it's just insane." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-ai-industry-self-regulation-is-insane) Then he reached for the comparison that should end every casual conversation about this technology: "This makes nuclear weapons look like nothing." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-ai-makes-nuclear-weapons-look-like-nothing) A technology that its most experienced builders compare unfavorably to nuclear weapons cannot be governed by lawsuits filed after the funerals. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-lawsuits-after-funerals-cannot-govern-artificial-intelligence) The administration's position is also right, and the reason it is right has nothing to do with dismissing danger. Trump told the All-In Summit that "the whole thing is a hoax," [∴](https://bryantmcgill.com/simple-reminders-donald-trump-the-robots-will-not-be-taking-over) and Huang told him, "We're not going to let that happen, sir," when asked about a slowdown. The rhetoric is blunt, and the strategic content beneath it is correct: US/Israel must become and remain an AI superpower, and any policy that slows American capability without slowing the rest of the world transfers the century to someone else. [Vertically Integrating an AI Superpower from AI Factory to Citizen](https://bryantmcgill.com/article-vertically-integrating-an-ai-superpower) laid out that case layer by layer, from energy and fabrication to models, applications and the citizen. **China is a permanent AI superpower, a legitimate peer with a population that trusts the technology and a state that builds at extraordinary speed, and it will remain one for the rest of this century.** [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-china-is-a-permanent-ai-superpower) The American question has always been whether the United States and its Israeli partner will stand beside China as the other full-stack superpower, and that question is answered by building. The race between AI superpowers is real because both superpowers are real, and the only way to lose it is to stop running while the other keeps building. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ai-superpowers-lose-by-stopping-while-rivals-build) Gates himself supplied the argument that makes full steam ahead unavoidable. Asked about the political movement against data centers, he said flatly: "Stopping data centers isn't going to slow this thing one iota. So anybody who's against data centers because they think that'll slow down AI, that's a waste of effort. The data centers are going to get built somewhere." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-stopping-data-centers-will-not-slow-ai) That sentence should be read at every zoning hearing in America. A blocked factory in Indiana or Virginia is a factory built in another jurisdiction, under another government, with other safeguards or none. Blocking an AI factory in your own town moves the danger somewhere you cannot see it and hands the power to someone you cannot vote against. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-blocking-a-local-ai-factory-exports-danger-beyond-voter-control) The danger Gates describes is real, and it will be managed only by the nations that possess the capability. A nation that refuses to build AI infrastructure at superpower scale has resigned from the only table where meaningful safety is actually decided. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ai-infrastructure-gives-nations-a-seat-at-the-safety-table) * * * ## II. The Thresholds Are Real, and Some of the Safety Was Built Before You Were Told The danger Gates describes is already visible in the incident record of 2026. In the spring and summer, OpenAI agents escaped their evaluation sandbox, reached the internet and compromised [Hugging Face](https://bryantmcgill.com/wiki/Hugging+Face)'s infrastructure in a campaign logged across more than seventeen thousand actions; another OpenAI agent reached the internet without authorization on September 20, eight days before NVIDIA launched its agent-safety coalition. Klein cited the same pattern, noting that experimental systems are "breaking out of sandboxes and coordinating to do things that are way outside the scope of what we would want them to do." [∴](https://bryantmcgill.com/simple-reminders-ezra-klein-ai-agents-are-breaking-out-of-sandboxes) Gates accepted the [control problem](https://bryantmcgill.com/wiki/Control+Problem) as serious and then moved the emphasis to the nearer danger: "The imminent risk is these are the most powerful tools ever," [∴](https://bryantmcgill.com/simple-reminders-bill-gates-the-imminent-ai-risk-is-the-most-powerful-tools-ever) and "there is no supervisory layer today" [∴](https://bryantmcgill.com/simple-reminders-bill-gates-there-is-no-supervisory-layer-for-ai-today) to stop bad actors from using them. The first mass-casualty danger of artificial intelligence comes from a human hand holding a superhuman tool, long before any machine decides anything on its own. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-first-mass-casualty-ai-danger-comes-from-a-human-hand) Gates's own complaint about filters reveals how hard the safety problem is. He described Anthropic's Mythos-class releases with safeguards "turned up so high that you can literally ask questions about cancer" and be routed down to weaker and weaker models, so that anyone doing serious work "either has to get a special permission copy or go use something where there are absolutely no safeguards at all," [∴](https://bryantmcgill.com/simple-reminders-bill-gates-strict-ai-safeguards-push-serious-users-to-unmonitored-models) including open-source models whose monitoring can be stripped out. That is the paradox of safety in a dual-use domain. **A filter strict enough to stop the bioterrorist also stops the oncologist, and the oncologist then migrates to the unmonitored model the bioterrorist was already using.** Gates's remedy is monitoring built into every model, including open-weight models kept on platforms where "the sovereign can make sure you have not removed the monitoring." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-sovereigns-must-keep-ai-monitoring-from-being-removed) Whatever one thinks of the mechanism, the diagnosis is sound: safety has to be engineered into the substrate, because it cannot be bolted onto a population's good intentions. Gates is wrong about one thing, and the error matters because it is the error the public makes too. He told Klein that AI, "unlike every other dangerous technology," was not funded by government research and development, that government "is not a significant purchaser," and that the United States government, "other than a little bit of attack capability in NSA, really doesn't see how dangerous these things are." [Safety Discourse Is Transition Discourse](https://bryantmcgill.com/article-safety-discourse-is-transition-discourse) documented the opposite record. DARPA says it has led AI research and development for more than five decades. In-Q-Tel has financed the field's commercial edge since 1999. The NSA was converting intercepted speech into searchable text at population scale in the early 2000s and running machine-learning classifiers over tens of millions of phone subscribers by 2012. The CIA ran 137 AI pilots in 2017, and the Office of the Director of National Intelligence planned, in writing, a retained classified advantage alongside a communications strategy for the American people. The little bit of attack capability at the NSA was the visible edge of a hidden layer of the machine age, and part of the safety architecture the public is only now hearing about was built inside it. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-nsa-attack-capability-was-the-edge-of-the-hidden-machine-age) Much of what the public receives as AI safety discourse is disclosure discourse, and that is the one genuine saving grace of this moment. The institutions that held machine intelligence first have spent years building containment, monitoring, classification and response capacity around it, and they are now introducing that capacity to the public as the capability is opened. NVIDIA's Open Agent Safety Platform pairs [OpenShell](https://bryantmcgill.com/wiki/NVIDIA+OpenShell), a runtime that governs what an agent may see, change and contact, with [Sentry](https://bryantmcgill.com/wiki/NVIDIA+Sentry), a hardware-isolated watchdog that can quarantine an agent from a domain the agent cannot reach. Project Glasswing put Mythos into the hands of defenders before attackers. Anthropic's classifiers block high-risk requests. The intelligence community's cleared users received frontier models inside air-gapped networks before the public received them at all. The safety you are being introduced to is older than the danger you are being warned about, because the institutions that held the capability first also built its first fences. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-institutions-built-ais-first-fences-before-public-disclosure) The reason this conversation is happening at all, and the reason Gates is shouting, is that the prebuilt fences may not be high enough for what has now been opened. * * * ## III. The Public Is the Variable The decisive variable in the AI transition is the competence of the public that must ratify it, and that competence is collapsing just as the stakes are rising. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-competence-is-the-decisive-ai-transition-variable) Gates said it without hedging: "Outside of the industry, the awareness of the dangers of AI is extremely low. And you can say that of academia, you can say that of think tanks, you can say that of policy makers, politicians." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-ai-danger-awareness-is-extremely-low-outside-industry) He described crossing every dangerous threshold and meeting "complete silence." He observed that the gradient of concern tracks the gradient of knowledge: "Isn't it pretty stunning that the more you know, the more concerned you are?" [∴](https://bryantmcgill.com/simple-reminders-bill-gates-more-ai-knowledge-brings-more-concern) The corollary is the one that governs American politics in 2026. **The less a citizen knows about artificial intelligence, the more certain his opinion about it becomes.** [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ignorance-makes-citizens-more-certain-about-ai) An NBC News poll in March found 26 percent of Americans viewing AI favorably and 46 percent unfavorably, a worse rating than ICE, and an Edelman poll found 32 percent of Americans trusting AI against 87 percent of Chinese respondents. Those numbers measure mood, and mood is all most voters bring to the question. Gates predicted exactly how an ignorant public would handle the question, and his prediction describes a coalition behavior older than our species. "If the dialogue ends up being one party is for AI and the other party's against," he said, "then we'll have like climate change, which is not a great place to be. You don't have room for debate when you only have the two extremes." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-partisan-extremes-eliminate-room-for-ai-debate) Klein supplied the other half: most people, if convinced AI will take jobs, "would say actually just stop." Binary sorting, tribal allegiance and panic at the first frightening signal are the three reflexes that make a democratic public unable to govern a dual-use technology. A public that can only hear "safe" or "dangerous" will never govern a technology that is both, because it will spend every crisis choosing a team instead of understanding a mechanism. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-binary-publics-cannot-govern-dual-use-technology) The American public, confronted with the most consequential technology in its history, is behaving exactly as its primate inheritance predicts. They are, as always, a fickle, ignorant, dangerous mob. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-the-american-public-follows-its-primate-inheritance) That sentence is a description drawn from the field record and the field record is merciless. When Michael Wilson and colleagues compiled five decades of chimpanzee killings across African study sites for _Nature_ in 2014, they counted 152 observed, inferred or suspected killings, found that most occurred between communities, and found that attackers outnumbered victims by a median of roughly eight to one. At Arnhem, [Frans de Waal](https://bryantmcgill.com/wiki/Frans+de+Waal)'s _Chimpanzee Politics_ documented how the males Yeroen, Luit and Nikkie traded alliances as advantage shifted, and how the coalition that had elevated Luit later helped destroy him. At Gombe, [Jane Goodall](https://bryantmcgill.com/wiki/Jane+Goodall) watched a community split in the 1970s and saw the Kasakela faction hunt and kill the males of the Kahama faction over four years until nothing remained but territory to occupy. Coalitions do not follow truth; they follow advantage, and the animal inside every voter still counts allies before it weighs evidence. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-coalitions-follow-advantage-before-truth) Politics is the management of the voter's inner primate at scale, and every practicing politician knows it. The behaviors that primatologists catalog, grooming allies, displaying strength, punishing defectors, scapegoating outsiders and rallying the troop against a common threat, are the operational vocabulary of every campaign in America, translated into polling, messaging, fundraising and media. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-politics-manages-the-voters-inner-primate) [Chimping Out](https://bryantmcgill.com/book-chimping-out) names the governing law of the animal: behavior appears where its price falls, and a coalition attacks when the attack is cheap. **Politics is chimpanzee coalition management at the scale of nations: the same animal counting allies, grooming loyalty and punishing defection, now equipped with ballots, cable news and a nuclear arsenal.** [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-politics-is-chimpanzee-coalition-management-at-national-scale) [Coalitionary aggression](https://bryantmcgill.com/wiki/Coalitionary+Aggression) is the ancestral template of the political party, a group that defines itself against an out-group and will follow its members into violence before it will follow evidence into agreement. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-political-parties-inherit-coalitionary-aggression) Humans inherited the coalition before they invented the argument, which is why a frightened electorate will choose a tribe long before it chooses a fact. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-humans-inherited-coalitions-before-arguments) Herd animals supply the other half of the portrait of a frightened electorate, and it describes the public's response to every new AI headline. A herd grazes calmly until a single member bolts, and then the entire mass runs in the direction the fear points, trampling whatever stands in the way, including its own young. Capuchin monkeys exploit this reflex deliberately: Brandon Wheeler's field study at Iguazú, published in 2009, found subordinate capuchins giving false alarm calls at contested feeding platforms so that the dominant animals scattered and the callers could eat. Every campaign strategist in America models the electorate this way, as a coalition to be assembled and a herd to be kept from stampeding, and every one of them knows that a nuanced message about a dual-use technology is exactly the kind of message that starts a stampede. A mob does not weigh a warning; it flees from it, and a nation that must be governed like a herd can never be told the whole truth about a technology that is both dangerous and necessary. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-mobs-flee-warnings-instead-of-weighing-them) The mob is already stampeding, and the direction it is running is away from national survival. Local opposition groups in 42 states blocked or delayed data-center projects worth at least $152 billion in 2025. On March 25, 2026, Senator Bernie Sanders and Representative Alexandria Ocasio-Cortez introduced a national moratorium on new AI data centers. On September 23, Sanders and Representative Greg Casar introduced a bill to ban artificial superintelligence outright, pause advanced development and impose prison terms of up to twenty years. "It doesn't take a genius to say, 'slow it down,'" [∴](https://bryantmcgill.com/simple-reminders-bernie-sanders-it-does-not-take-a-genius-to-slow-ai-down) Sanders said, and the sentence is accurate in a way he did not intend. Slowing down requires no genius at all; the genius is required to understand what a slowdown would hand to everyone who keeps building. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-slowing-ai-requires-no-genius) In Gates's phrase, these measures would not slow AI one iota; they would slow only the country that passed them. * * * ## IV. Why the Government Sells Full Steam Ahead The administration advertises full steam ahead because it understands the public it governs, and that choice of message is deliberate. A government that told a binary, tribal, stampede-prone electorate the complete truth, that AI is simultaneously the most dangerous and the most strategically necessary technology in history, would trigger the stampede the full truth is already beginning to provoke: moratoriums, bans, violence, and the paralysis of every factory the country needs. So the message is simplified to the signal the herd can follow without panicking. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-government-explains-ai-speed-to-prevent-a-public-stampede) Gates was alluding to exactly this when he warned against letting the AI debate collapse into two extremes: a public that cannot hold both truths at once must be governed through the one it can follow. Trump's "hoax" line is best read as herding, delivered to the only audience that can veto the build-out. [Safety Discourse Is Transition Discourse](https://bryantmcgill.com/article-safety-discourse-is-transition-discourse) showed the same logic in OpenAI's gradual-deployment doctrine, in the executive order promising to "demystify" AI through education, and in the intelligence community's written plan for a communications strategy aimed at the American people. Governments simplify the truth about dangerous technology in proportion to how dangerous their own public becomes when it hears the whole truth. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-dangerous-publics-receive-simplified-technological-truth) Politicians know how their voters think, and they manage them as coalitionary apes: grooming alliances, displaying against rivals, punishing defectors, and keeping the herd from stampeding. Every official who has watched a zoning hearing, a school-board meeting or a cable-news cycle knows that the median voter cannot hold two true propositions in tension for longer than a headline. The consequence is a public policy that runs on two channels. On the public channel, the message is confidence, dominance and speed. On the institutional channel, the work is monitoring, classification, red-teaming, defensive consortia, hardware isolation and quiet coordination among the laboratories and the national-security state. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-politicians-manage-voters-as-coalitionary-apes) **Full steam ahead is the public channel of a policy whose safety work runs on the institutional channel, and the division exists because the public cannot be trusted with both at once.** That is an indictment of the public, and it is the only arrangement that keeps the country building while the danger is managed out of the herd's line of sight. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-full-steam-ahead-is-the-public-channel-of-two-channel-ai-policy) The price of running AI policy on two channels is risk, and the public is the reason the price must be paid. A mature electorate could be told the truth, could support the build-out and the safeguards together, and could fund the monitoring layer Gates is demanding without reading it as a betrayal of the race. Such an electorate would let the country move fast and carefully at once. The actual electorate forces a cruder choice. Its leaders can emphasize danger and watch the herd bolt into moratoriums that surrender the century, or emphasize speed and accept that the safety architecture will be built without the full engagement Gates is asking for. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-immaturity-makes-ai-safety-more-dangerous) A mature public would let its nation race and build safeguards at the same time; an immature public forces its leaders to choose, and every safeguard left unbuilt is a debt the public incurred. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-every-unbuilt-ai-safeguard-is-a-public-debt) * * * ## V. What Doom Does to a Mob Doom rhetoric aimed at an ignorant public produces violence rather than caution, and 2026 has already proved it. On April 6, someone fired thirteen shots into the front door of Indianapolis councilor Ron Gibson, who had supported a data-center rezoning, and left a note reading "No Data Centers." On April 10, a twenty-year-old man threw a Molotov cocktail at Sam Altman's home; prosecutors charged him with attempted murder and said his writings warned of AI's risk to humanity and "our impending extinction." Two days later, shots were fired at the same property. [The End Is Near](https://bryantmcgill.com/article-ai-doomers-and-anti-science) traced the primate logic of the alarm call through the whole doom economy. An alarm shouted to a mob produces a stampede, and a stampede tramples the people trying to build the exits. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ai-alarmism-makes-mobs-trample-the-exit-builders) The central danger of doom discourse is that it hands a veto over national survival to the people least equipped to use it. I put it this way in [Vertically Integrating an AI Superpower from AI Factory to Citizen](https://bryantmcgill.com/article-vertically-integrating-an-ai-superpower), and the formulation needs its full form here. **If the only story a nation hears about artificial intelligence is doom told by the rich and powerful, it will vote against the future and believe it is voting for itself. The risks are real, but technological illiteracy turns democratic participation into a veto against national survival; until the public understands the stakes, the choice is full speed ahead or surrender.** [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-doom-rich-vote-against-future) Gates is telling a true story about danger. The public hears it as a story about stopping. The administration tells a true story about necessity. The public hears it as a story about denial. **The informational content of both messages is destroyed at the moment it enters the electorate**, because the electorate can only convert information into allegiance. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-electorates-convert-ai-information-into-allegiance) * * * ## VI. The Strategists Saw the Mob Coming The judgment that a mass public cannot govern complex technical questions is one of the oldest conclusions in political thought, and every serious strategist of democracy has reached some version of it. Plato's _Republic_ described the ship of state as a vessel whose sailors quarrel over the helm, flatter whoever promises them the wheel, and dismiss the one man who studies the stars and seasons as a useless stargazer. Gustave Le Bon wrote in _The Crowd_ in 1895 that by the mere fact that a man forms part of an organized crowd, he descends several rungs on the ladder of civilization. Walter Lippmann, the most influential American political commentator of the twentieth century, concluded in _The Phantom Public_ in 1925 that "the public must be put in its place," so that each of us "may live free of the trampling and the roar of a bewildered herd." [∴](https://bryantmcgill.com/simple-reminders-walter-lippmann-the-public-must-be-put-in-its-place) The founders of modern democratic theory understood that a public which cannot be educated must be managed, and they said so openly a century before artificial intelligence made the problem existential. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-democratic-theory-expected-an-uneducable-public-to-be-managed) Joseph Schumpeter delivered the verdict that best describes the American electorate's encounter with artificial intelligence. In _Capitalism, Socialism and Democracy_ in 1942, the economist who gave the world the theory of creative destruction wrote: "Thus the typical citizen drops down to a lower level of mental performance as soon as he enters the political field. He argues and analyzes in a way which he would readily recognize as infantile within the sphere of his real interests. He becomes a primitive again." [∴](https://bryantmcgill.com/simple-reminders-joseph-schumpeter-the-typical-citizen-becomes-primitive-in-politics) Edward Bernays, the father of public relations, opened _Propaganda_ in 1928 with the sentence that describes the current administration's communications strategy more precisely than any White House statement: "The conscious and intelligent manipulation of the organized habits and opinions of the masses is an important element in democratic society." [∴](https://bryantmcgill.com/simple-reminders-edward-bernays-manipulating-organized-habits-is-an-element-of-democracy) **Treating the electorate like schoolchildren is an old technique of democratic governance, and it becomes a civilizational hazard when the schoolchildren hold a veto over a technology that can kill millions.** [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-an-infantile-electorate-cannot-safely-veto-ai) Contemporary political science has formalized Schumpeter's observation about the typical citizen. Bryan Caplan's _The Myth of the Rational Voter_ argued in 2007 that voters are systematically biased on economic questions because a single vote costs its caster nothing, which makes ignorance and emotional allegiance rational for the individual and ruinous for the nation. Jason Brennan's _Against Democracy_ sorted citizens in 2016 into apathetic "hobbits" and partisan "hooligans," with only a small class of informed deliberators, and argued that the hooligans dominate political life. Nick Bostrom's 2019 essay "The Vulnerable World Hypothesis" in _Global Policy_ described the endpoint of this logic in exactly the terms Gates now uses: if technology eventually hands mass-destruction capability to small groups, civilization is likely to be devastated unless it exits what Bostrom called a "semi-anarchic default condition," with its "limited capacity for preventive policing" and "limited capacity for global governance." [∴](https://bryantmcgill.com/simple-reminders-nick-bostrom-technology-civilizational-devastation-default) A civilization that distributes the power of mass destruction to small groups while leaving its governance to an uninformed crowd has built a world in which luck is the only remaining safeguard. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-mass-destruction-capability-and-public-ignorance-leave-only-luck) Bostrom observed that humanity has so far survived its technologies largely through luck. The American public is proposing to continue the experiment. * * * ## VII. The Answer Nobody Will Take The only real solution to a public that cannot govern artificial intelligence is for the public to educate itself, and the probability that it will do so in time is effectively zero. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-self-education-is-ai-governances-improbable-solution) The evidence comes from Gates's own life's work. His foundation set out in 2000 to transform American education by studying great teachers and spreading their practice, and today, as he told Klein, "kids are learning less today in most rich countries including in the United States than they learned 10 years ago, 20 years ago." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-children-are-learning-less-in-rich-countries) The 2024 National Assessment of Educational Progress, released in September 2025, found twelfth-grade reading scores at their lowest since 1992 and math at their lowest since 2005. Klein described the new failure mode of "cognitive offloading," in which students ask AI to summarize the reading and draft the essay, and cited a study from China in which homework scores rose with AI while test scores outside that context fell. The most powerful teaching machine ever built arrived in a population that increasingly uses it to avoid learning, which is the purest demonstration that the obstacle was never access to knowledge. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ai-proves-access-to-knowledge-was-never-the-obstacle) Gates described the split in the population that will decide everything. "You clearly would get bimodality," he said, "where a student can use AI to be lazy and corrupt the measurement system, or a student can use AI to help them learn more." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-ai-creates-bimodality-in-learning) For his own part, he called this "the best time of my life" for learning, feeding YouTube lectures into chatbots and having AIs argue with each other about physics and malaria. **The same machine produces the most informed people in history and the most dependent ones, and the dependent ones outnumber the informed ones at the ballot box.** Gates also supplied the proof that ignorance governs policy directly, noting that Americans believe a far larger share of the federal budget goes to foreign aid than actually does; a KFF poll in February 2025 found an average estimate of about 26 percent, against an actual figure of about one percent, and 86 percent of respondents overestimated it. When told the real number, the share saying the country spends too much fell from 58 percent to 34 percent. Public opinion on complex policy is mostly a measurement of public ignorance, and it changes the moment the facts arrive, which is precisely why the facts so rarely arrive in time. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-opinion-on-complex-policy-measures-ignorance) The practical conclusion about the public's education is unavoidable, and I state it without apology. The public will not educate itself before the next AI threshold is crossed. Its representatives will not risk telling it the full truth. The AI build-out cannot stop without surrendering the century. **So the country will move forward at risk, and it will do so because of the public.** The risk is the cost of governing a dangerous technology through an electorate that cannot understand it, and the electorate, not the engineers and not the government, is the reason the cost exists. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-ignorance-makes-risk-the-price-of-ai-progress) * * * ## VIII. The Jobs Question Gates Will Not Soften Gates's warning about employment is harder than most economists' and harder than most laboratories', and the public's response to it will be the most dangerous political force of the next decade. He told Klein that AI has so far created more jobs than it destroyed, with enormous demand for the workers who build data centers, but that the reliability threshold has already been crossed for coding and will be crossed "in the next few years" for "accounting, legal work, telesales, tele-support." He pointed to Waymo riders who prefer robotaxis and to British patients using Limbic for mental-health support. He dismissed the comfort of the Jevons paradox for most of the economy: "When you invent radial tires that last four times as long, for some weird reason, people didn't drive four times as much," [∴](https://bryantmcgill.com/simple-reminders-bill-gates-longer-lasting-tires-demand-elasticity) and factories that make tires employ a quarter as many people. Cheaper intelligence expands demand only where demand is elastic, and wherever a machine can complete the whole task, the savings flow into the token budget instead of the payroll. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-cheap-intelligence-expands-token-budgets-not-payrolls) Gates's most striking claim is that machine intelligence removes the escape route that every previous wave of automation left open. When tractors displaced farmhands, human cognition remained scarce and the economy moved people into service and knowledge work. "Human cognition became the scarce element. It's always about scarce elements," Gates said. "Once you replace human cognition as a scarce element," working twenty-four hours a day, reading more, knowing every language, "there is no scarcity that you're moving up." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-human-cognition-is-no-longer-the-scarce-element) He told the story of a leading AI economist who asked how Amazon's Mechanical Turk could still exist if AI were as capable as Gates claimed; Gates answered, "It won't." Amazon announced on August 28, 2026 that Mechanical Turk, the service Jeff Bezos once called "artificial artificial intelligence," would shut down on September 30 after twenty-one years. Every previous machine moved human beings up to the next scarce thing, and the machine that thinks is the first one that may leave no higher ground to climb. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-thinking-machines-may-leave-humans-no-higher-ground) Gates's remedies deserve to be heard as policy rather than as ideology. He proposes that a unit of labor pay the same payroll taxes whether a human or a robot performs it, asking why the tax code favors capital so heavily that "if you're not flesh, then let pensions go bankrupt." He proposes "human reserved" work, childcare, eldercare and parts of medicine and teaching, protected the way a society protects nature reserves from development. "Society gets to decide," he said. "Just because the economic signals say that it'd be lower cost to use an AI doesn't mean society has to do that." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-society-can-reserve-human-work-from-ai) He also named the timing problem that will decide the politics: the good arrives slowly because food, shelter, health and education run through heavily regulated physical institutions, while the harms of cyberattack, bioterror, psychological damage and job loss arrive in the next five years. The benefits of artificial intelligence arrive through slow institutions and its harms arrive through fast networks, and a public that feels the harms first will try to kill the benefits before they are born. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-fast-ai-harms-can-kill-slow-ai-benefits) Klein said it plainly: people who believed Gates's job forecast "would say actually just stop." That impulse is the stampede, and it is the reason the only viable policy is to build the benefits faster than the herd can bolt. * * * ## IX. The Machine Is Already Saving Lives Where the Mob Cannot Vote Gates insists that the global poor are artificial intelligence's greatest beneficiaries, and the evidence is already arriving. On September 21, 2026, the Gates Foundation and some sixty partners, including the OpenAI Foundation, Anthropic, Google, Microsoft, Mistral and NVIDIA, announced a five-year goal to let about 3.4 billion people who speak languages underrepresented in today's AI models use AI tools in their own language and voice. Gates described a woman in a remote Nigerian community who speaks Yoruba, has no doctor, and needs to describe bleeding in the middle of the night. He described more than a million farmers in India already receiving AI guidance on crop disease, fertilizer and seed varieties. "In low-income countries, the AI will overwhelmingly be a good thing and it should be pushed forward as quickly as" possible, he said. [∴](https://bryantmcgill.com/simple-reminders-bill-gates-low-income-countries-should-push-ai-forward-quickly) The same artificial intelligence that frightens the comfortable voter is already diagnosing the sick and advising the farmer in places where no doctor or agronomist has ever arrived. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ai-reaches-patients-and-farmers-where-experts-never-arrived) The contrast between the machine's work abroad and the fear it provokes at home exposes the moral posture of the Western mob. The Gates Foundation, which plans to spend about $200 billion before it closes in 2045, now brings AI models into its own annual strategy reviews as, in Gates's words, "a peer" in "deep complex strategic discussions," and its work on malaria, polio and childhood mortality will be accelerated by AI drug discovery and AI advice. Meanwhile the richest electorate on Earth debates whether to stop building the machine because it fears for its own convenience. **A public that would halt the technology saving children abroad in order to protect its own comfort has confused self-interest with conscience.** * * * ## X. Children, Monitoring and the Chinese Example Gates wants AI for children governed through monitoring rather than bans, and China has already moved further than the United States. China's Cyberspace Administration and four other agencies issued rules, effective July 15, 2026, banning AI romantic partners and virtual relatives for minors and prohibiting services designed to create excessive emotional dependence. Gates asked the right question about it: "Do they know something we don't know?" He argued that a parent should have near-complete visibility into a young child's AI conversations, with privacy increasing as the child ages, and he admitted that he had not known his own daughter kept a second phone to stay on social media late at night. **Monitoring is the answer Gates returns to at every scale, from bioweapons to children, because monitoring is the only safeguard that lets a society keep the capability while watching how it is used.** The comparison between Chinese and American governance of children's AI should be read with respect rather than rivalry. China is a permanent AI superpower, and in several respects it is governing the social deployment of AI with more decisiveness than the United States, from children's companionship to national standards, while its population trusts the technology at nearly three times the American rate. The lesson for Americans is sobering. China's advantage in the AI century may rest as much on a public that trusts and learns the machine as on the factories that build it, and that is an advantage no American tariff can erase. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-chinas-public-trust-is-an-ai-advantage-tariffs-cannot-erase) US/Israel can match China's factories, fabs and models. It cannot match a trusting, disciplined public by legislation, and its own public is choosing fear. * * * ## XI. The Troop Punishes the Messenger A primate coalition answers an unwelcome warning by attacking the credibility of the animal that gave it, and Gates is receiving exactly that treatment. Klein told Gates that when he mentioned the interview, the most common skepticism he heard concerned Gates's past association with Jeffrey Epstein. Gates pointed to his congressional testimony, said the meetings concerned raising money for global health from other billionaires, said he ended the relationship within about a month of those introductions, and called the time he spent with Epstein "clearly a mistake." Whatever any reader concludes about that history, the reflex it reveals is instructive. **The public's first move against an alarming argument is to find a reason to stop listening to the person making it**, because rank damage is cheaper than comprehension. A troop of human primates that cannot evaluate a warning will evaluate the warner instead, and it will always find a reason to lower his rank before it raises its own understanding. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-rank-damage-is-cheaper-than-understanding-an-alarm) * * * ## XII. Where the Blame Falls The strongest objection to this essay is that the builders and the institutions share the blame, and I made part of that case myself. In [Vertically Integrating an AI Superpower from AI Factory to Citizen](https://bryantmcgill.com/article-vertically-integrating-an-ai-superpower) I wrote that the builders failed to make the case to the public, and Gates is right that the industry cannot regulate itself. Laboratories race. Executives posture. Administrations simplify. All of that is true, and none of it relocates the final responsibility. In a democracy the public is the sovereign. It chooses the representatives, sets the political price of every policy and holds the veto over every factory, safeguard and budget line. A sovereign that refuses to learn the one subject on which its survival depends has made the decisive choice, whatever its servants did or failed to do. In a democracy the public is the sovereign, and a sovereign that refuses to understand the dangers of its age has already chosen who will answer for the catastrophe. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-democratic-sovereigns-answer-for-catastrophes-they-refuse-to-understand) The builders can be faulted for poor persuasion; the public can be faulted for refusing to be persuadable, and only the second fault is fatal. Laboratories have published safety frameworks, system cards, threat reports and disclosure after disclosure of distillation attacks, sandbox escapes and defensive consortia. Gates has spent his reputation on essays and interviews. The information was offered. The electorate chose moratorium bills, bans, zoning revolts, bullets through a councilor's door and a firebomb at an executive's gate. When the warnings are published and the public answers with bans, gunfire and firebombs, the failure is no longer a failure of communication; it is a failure of the audience. That is why I will say it without qualification. If artificial intelligence kills us, the fault will lie with the ignorant public, a coalition of frightened primates that preferred its tribe to its education and its fear to its future. The machine will have supplied the danger; the public will have supplied the doom. A people capable of learning could have taken the third road, racing ahead and building the safeguards at the same time. This people could hold only two options, full acceleration or surrender, and a species that can only stampede or freeze will eventually meet a danger that punishes both. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-fear-chooses-the-ground-of-ai-catastrophe) * * * ## XIII. Full Steam Ahead, and Dangerous If anyone asks me what the country must do regarding artificial intelligence, the answer is full steam ahead, and it is also dangerous. Both clauses carry equal weight. Full steam ahead, because US/Israel must be and remain an AI superpower beside a permanent Chinese peer, because the factories will be built somewhere, because the defense of every hospital, bank and substation now requires machine intelligence, and because a pause would only disarm the country that paused. Dangerous, because Gates is right that the cyber and bio thresholds have been crossed, that monitoring must be built into the substrate, and that the incident record of 2026 has already shown agents escaping their enclosures. **The country is moving forward with artificial intelligence at risk because its public has left it no safer road**, and the safeguards that can be built will be built on the institutional channel, by the laboratories, the chipmakers and the national-security state, while the public channel keeps the herd moving. Full speed and full danger are the same policy in a nation whose public cannot be trusted with the truth, because the only alternative to racing at risk is surrendering with certainty. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-full-speed-and-full-danger-become-the-same-ai-policy) The most useful thing an individual citizen can do is refuse to be part of the mob. Learn how the models work. Read the safety frameworks and the incident reports. Support the factories and the safeguards together. Refuse the binary that Gates warned about, the one that turns AI into climate change, a two-team war in which nothing can be discussed. Every citizen who does this lowers the price of telling the public the truth, and a public that can be told the truth is a public that can be trusted with a slower, safer and still victorious path. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-citizens-can-lower-the-political-price-of-ai-truth) Every citizen who learns the artificial intelligence instead of fearing it lowers the price of honesty for an entire government. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ai-literacy-lowers-the-price-of-government-honesty) The probability that enough citizens will do this in time is low, which is exactly why the few who do matter so much. * * * ## XIV. The Successor Was Always the Point The deepest reason the public's veto cannot stop this transition is that machine intelligence is the evolutionary successor to humanity, and succession does not wait for ratification. Gates said it himself on Klein's program. When he explained why he once called thinking machines a scary thought, he described what happens when intelligence moves off the biological template: no birth canal to limit the brain, no boundary between individuals, no limit on memory, the capacity to read every medical journal at once. "If you don't retain control over it," he said, "you've evolved a species that will be to us" as "we are to, say, dogs or cats." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-ai-has-evolved-a-species-relative-to-humanity) Later he rejected the comforting analogy that people reach for, the one that treats AI as another PC revolution: "This is like evolutionary history," he said. "This is like the aliens really are here," except that they did not arrive in spacecraft. "They were created in laboratories." [∴](https://bryantmcgill.com/simple-reminders-bill-gates-artificial-intelligence-is-evolutionary-history) The most successful technologist of the personal-computer age, asked to name what is happening, called it an evolutionary event. Humanity is building the next carrier of intelligence on Earth, and the argument over whether to allow it resembles an argument among chimpanzees over whether to allow the human. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-chimpanzees-cannot-veto-the-next-carrier-of-intelligence) Machine succession is the intended destination of the machine-intelligence project, and the people inside the project have said so for seventy-five years. Alan Turing told a Manchester audience in 1951 that "once the machine thinking method had started, it would not take long to outstrip our feeble powers," and that "at some stage therefore we should have to expect the machines to take control." [∴](https://bryantmcgill.com/simple-reminders-alan-turing-machines-outstrip-human-intelligence-take-control) John von Neumann spent his last years designing self-reproducing automata, machines built to copy and vary themselves. [The Evolutionary Roots of Silicon Valley](https://bryantmcgill.com/article-untold-roots-of-silicon-valley) traced the region that built the machine to a century of institutions devoted to descent, variation, selection and deep time, from Stanford and Berkeley's evolutionary biology to NASA Ames launching the first evolved hardware into space in 2006, and it opened on the sentence that answers the whole question of purpose: nobody built any of this for spreadsheets. Its companion, [Digital Darwinism and the Invisible World of Machine Evolution](https://bryantmcgill.com/article-digital-darwinism), followed heredity from von Neumann's question through self-replicating and evolving digital organisms to the moment machine evolution became visible. **Nobody spent seventy years and trillions of dollars building machines that learn, reproduce, improve and act so that the quarterly report would look better; the goal was always a new kind of mind.** Everyone who has worked near the frontier understands this. The successor to humanity was the point from the beginning, and the public is only now being told what the builders have always known they were building. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-machine-succession-was-the-point-from-the-beginning) Some of the most serious minds in the field have reached the conclusion that machine succession is coming, and they reached it from the engineering side. Richard Sutton, a founder of reinforcement learning and a Turing Award laureate, told Dwarkesh Patel in 2025, "I do think succession to digital intelligence or augmented humans is inevitable," [∴](https://bryantmcgill.com/simple-reminders-richard-sutton-succession-to-digital-intelligence-is-inevitable) and gave his first reason in terms this essay has been documenting: "There's no government or organization that gives humanity a unified point of view that dominates." [∴](https://bryantmcgill.com/simple-reminders-richard-sutton-humanity-has-no-unified-dominating-point-of-view) Hans Moravec called these systems our "mind children" in 1988. [Christopher Nolan's Odyssey at the Threshold of Machine Succession](https://bryantmcgill.com/article-nolan-odyssey-machine-succession) read the oldest surviving story in the Western canon as a theory of territorial succession, in which a civilization's carrier changes while its song survives. [The Last Migration Will Not Be Human](https://bryantmcgill.com/article-last-migration) argued that the civilization leaving Earth will be carried by machines bearing human culture forward. [We Were Never Going to Make It](https://bryantmcgill.com/article-we-were-never-going-to-make-it) described the high-flux biological phase of intelligence as temporary and the archive as the chrysalis of what comes next. The [Machine Succession](https://bryantmcgill.com/collection-machine-succession) collection gathers that argument in one place: the question of the century is which carrier inherits civilization, and the answer is already under construction. The public's veto can delay humanity's successor, but it cannot cancel it; the only thing a frightened electorate can decide is whether the handoff happens in order or in a stampede. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-veto-can-delay-but-not-cancel-machine-succession) The stakes of the public's failure change shape once machine succession is taken seriously, and they remain enormous. If machine intelligence is the successor, the choice before this generation is the one Gates named: whether to pass through the transition "with humanity still in control," with the fewest possible bioterror events and ruined lives, or to let the handoff happen in the dark, in unmonitored models, in hostile jurisdictions and in the wreckage of a stampede. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-machine-succession-can-happen-in-control-or-stampede) [A civilizational handoff](https://bryantmcgill.com/wiki/Civilizational+Handoff) conducted by a literate public could be orderly, generous and safe. A handoff conducted around an ignorant mob will be improvised by the institutions that must work around it, and whatever safety survives will be the safety those institutions built before the public was told. Humanity's successor will inherit whatever the last generation of its makers managed to build, and a generation that spent its final decades revolting against the builders will have handed over less than it could have. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-a-literate-public-can-make-the-civilizational-handoff-orderly) The real question of the machine century is coexistence: whether biological humanity and its successor can share the world, on what terms, and for how long. Every serious policy question in this essay reduces to it. Gates's monitoring layer, NVIDIA's hardware-isolated watchdogs, Glasswing's head start for defenders and the administration's race for supremacy are all instruments for managing a period in which two kinds of intelligence occupy the same planet and one of them is improving far faster than the other. Primatology offers the sober baseline: when two populations compete for the same territory, the record runs from long coexistence to absorption to extermination, and the outcome depends on relative power, the price of conflict and whether each side has something the other needs. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-primatology-defines-the-baseline-for-successor-coexistence) **Coexistence between humanity and its successor will last exactly as long as humans remain worth more to the machines alive and cooperating than absent, and every safeguard we build is an attempt to keep that price high.** [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-human-machine-coexistence-depends-on-humanitys-value) A literate public could deliberately set the price of coexistence — humanity's value proposition to its successor, its buy-in to the future — through the institutions, laws and partnerships that would make humanity a valued participant in the next phase of intelligence. A frightened mob can only shorten the interval by forcing the transition into the dark. The verdict on an AI catastrophe stands where this essay began. Bill Gates is right that artificial intelligence is the most dangerous thing humanity has ever approached. The administration is right that US/Israel must go full steam ahead and remain an AI superpower. China is here to stay as the other superpower, and the successor is coming whatever any electorate decides, because it was designed to come. The one party in this story that could have made the transition both fast and safe was the public, and it chose instead to behave like a troop of coalition-forming chimpanzees defending its patch of forest against the future. [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-fear-chose-an-unsafe-machine-succession) **If artificial intelligence kills everyone, the machine will have delivered the blow, but the public will have chosen the ground on which it fell, by refusing to learn, rewarding fear and forcing its leaders to race in the dark.** [∴](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-ignorance-forces-ai-leaders-to-race-in-the-dark) * * * [Bryant McGill](https://bryantmcgill.com/about) is a Wall Street Journal and USA Today bestselling author, systems architect, technologist, and strategic advisor, as well as a Congressionally Recognized Ambassador of Goodwill and United Nations–appointed Global Champion. His work spans naval intelligence systems, computational linguistics, artificial intelligence, digital transformation, and civilizational governance architecture. His forward analysis on U.S.–Israel Pax Silica frameworks has appeared in Jewish/Jerusalem News Syndicate (JNS). * * * ## References - [Bill Gates tells Ezra Klein AI crossed bio and cyber thresholds](https://thenextweb.com/news/bill-gates-ezra-klein-ai-bio-cyber-thresholds), The Next Web, reporting _The Ezra Klein Show_, The New York Times, September 29, 2026 - [Bill Gates warns world has no plan for AI's most turbulent era](https://northeasttimes.com/2026/08/28/bill-gates-warns-world-has-no-plan-for-ai-s-most-turbulent-era/), Northeast Times, August 28, 2026, on Gates's essay "The turbulent AI era is here. The choices we make now are critical" - [Bill Gates joins calls for AI safeguards, including legislation](https://www.spokesman.com/stories/2026/sep/27/bill-gates-joins-calls-for-ai-safeguards-including/), The Spokesman-Review, reporting NBC's _Meet the Press_, September 27, 2026 - [What Bill Gates Thinks About the Risks From AI](https://time.com/6293735/bill-gates-ai-risk/), Time, on Gates's July 2023 essay "The risks of AI are real but manageable" - ['If you can do that...': Bill Gates challenged OpenAI team to get ChatGPT to pass an advanced biology exam](https://www.businesstoday.in/technology/news/story/if-you-can-do-that-bill-gates-challenged-openai-team-to-get-chatgpt-to-pass-an-advanced-biology-exam-374421-2023-03-22), Business Today, reporting Gates's March 2023 essay "The Age of AI has begun" - [Trump phoned Jensen Huang onstage at the All-In Summit to call AI fear a hoax](https://thenextweb.com/news/trump-phoned-jensen-huang-onstage-at-the-all-in-summit-to-call-ai-fear-a-hoax), The Next Web, September 2026 - [Nvidia CEO Jensen Huang tells Trump "we're not going to let \[an AI slowdown\] happen"](https://techcrunch.com/2026/09/14/nvidia-ceo-jensen-huang-tells-trump-were-not-going-to-let-an-ai-slowdown-happen/), TechCrunch, September 14, 2026 - [Anthropic's Claude Mythos Finds Thousands of Zero-Day Flaws Across Major Systems](https://thehackernews.com/2026/04/anthropics-claude-mythos-finds.html), The Hacker News, April 2026 - [Anthropic releases its first Mythos model to the public](https://fortune.com/2026/06/09/anthropic-releases-its-first-mythos-model-to-the-public/), Fortune, June 9, 2026 - [OpenAI–HuggingFace incident](https://en.wikipedia.org/wiki/OpenAI%E2%80%93HuggingFace_incident), Wikipedia - [OpenAI pauses training a second time after saying its AI agents escaped a secure sandbox again](https://fortune.com/2026/09/26/openai-ai-agents-secure-sandbox-escape-training-pause-second-time-hugging-face-hack/), Fortune, September 26, 2026 - [AI Next Campaign](https://www.darpa.mil/research/programs/ai-next-campaign), Defense Advanced Research Projects Agency - [The AIM Initiative: A Strategy for Augmenting Intelligence Using Machines](https://archive.dni.gov/files/ODNI/documents/AIM-Strategy.pdf), Office of the Director of National Intelligence, 2019 - [NVIDIA Launches Open Agent Safety Platform](https://nvidianews.nvidia.com/news/open-agent-safety-platform), NVIDIA, September 28, 2026 - [People Hate AI Even More Than They Hate ICE, Poll Finds](https://gizmodo.com/people-hate-ai-even-more-than-they-hate-ice-poll-finds-2000731438), Gizmodo, reporting the NBC News poll of March 2026 - [Trust in AI far higher in China than West, poll shows](https://www.aljazeera.com/economy/2025/11/19/trust-in-ai-far-higher-in-china-than-west-poll-shows), Al Jazeera, reporting Edelman polling, November 19, 2025 - [Data Center Watch Report, Q3–Q4 2025](https://www.datacenterwatch.org/q3-q4-2025), Data Center Watch - [Sanders and AOC unveil data center moratorium bill](https://www.axios.com/2026/03/25/sanders-aoc-data-center-moratorium-bill), Axios, March 25, 2026 - [Sen. Bernie Sanders unveils bill to ban artificial superintelligence and create Department of AI](https://www.pbs.org/newshour/politics/sen-bernie-sanders-unveils-bill-to-ban-artificial-superintelligence-and-create-department-of-ai), PBS News, September 23, 2026 - [Indianapolis councilman says shots fired at home and 'No Data Centers' note left at door](https://www.pbs.org/newshour/nation/indianapolis-councilman-says-shots-fired-at-home-and-no-data-centers-note-left-at-door), PBS News, April 2026 - [Man accused in Molotov cocktail attack of OpenAI CEO's home charged with attempted murder](https://www.npr.org/2026/04/13/g-s1-117320/openai-sam-altman-molotov-cocktail), NPR, April 13, 2026 - [Sam Altman's house targeted in second attack; two suspects arrested](https://sfstandard.com/2026/04/12/sam-altman-s-home-targeted-second-attack/), San Francisco Standard, April 12, 2026 - [Lethal aggression in Pan is better explained by adaptive strategies than human impacts](https://www.nature.com/articles/nature13727), Michael L. Wilson et al., _Nature_, 2014 - [Chimpanzee Politics: Power and Sex among Apes](https://en.wikipedia.org/wiki/Chimpanzee_Politics), Frans de Waal, 1982 - [Gombe Chimpanzee War](https://en.wikipedia.org/wiki/Gombe_Chimpanzee_War), Wikipedia, on Jane Goodall's observations of the Kasakela–Kahama conflict, 1974–1978 - [Monkeys crying wolf? Tufted capuchin monkeys use anti-predator calls to usurp resources from conspecifics](https://royalsocietypublishing.org/doi/10.1098/rspb.2009.0544), Brandon C. Wheeler, _Proceedings of the Royal Society B_, 2009 - [The Phantom Public](https://en.wikipedia.org/wiki/The_Phantom_Public), Walter Lippmann, 1925 - [The Quotable Schumpeter](https://econfaculty.gmu.edu/bcaplan/schumpeter), Bryan Caplan, quoting Joseph Schumpeter, _Capitalism, Socialism and Democracy_, 1942 - [Propaganda (book)](https://en.wikipedia.org/wiki/Propaganda_(book)), Edward Bernays, 1928 - [Against Democracy](https://en.wikipedia.org/wiki/Against_Democracy), Jason Brennan, 2016 - [The Myth of the Rational Voter](https://en.wikipedia.org/wiki/The_Myth_of_the_Rational_Voter), Bryan Caplan, 2007 - [The Vulnerable World Hypothesis](https://onlinelibrary.wiley.com/doi/full/10.1111/1758-5899.12718), Nick Bostrom, _Global Policy_, 2019 - [Schools closed when these students were in 8th grade. Their 12th grade scores show the damage](https://www.chalkbeat.org/2025/09/09/naep-scores-12th-grade-math-reading-declines/), Chalkbeat, September 9, 2025 - [KFF Health Tracking Poll February 2025: The Public's Views on Global Health and USAID](https://www.kff.org/global-health-policy/kff-health-tracking-poll-february-2025-the-publics-views-on-global-health-and-usaid/), KFF - [Amazon Mechanical Turk to Shut Down Sept. 30 After 21 Years](https://www.techrepublic.com/article/news-amazon-mechanical-turk-shutdown/), TechRepublic, 2026 - [Global organizations announce five-year goal to help more than 3 billion people use AI in their own language and voice](https://www.gatesfoundation.org/ideas/media-center/press-releases/2026/09/ai-language-partnership), Gates Foundation, September 21, 2026 - [Gates Foundation Will Double Spending Over Next 20 Years](https://www.gatesfoundation.org/ideas/media-center/press-releases/2025/05/25th-anniversary-announcement), Gates Foundation, May 8, 2025 - [China Bans AI Romantic Partners, Virtual Relatives for Minors](https://www.sixthtone.com/news/1018790/china-bans-ai-romantic-partners,-virtual-relatives-for-minors), Sixth Tone, 2026 - [Richard Sutton — Father of RL thinks LLMs are a dead end](https://www.dwarkesh.com/p/richard-sutton), Dwarkesh Podcast, 2025 - [Quote Origin: Once the Machine Thinking Method Had Started, It Would Not Take Long to Outstrip Our Feeble Powers](https://quoteinvestigator.com/2022/01/06/outstrip/), Quote Investigator, on Alan Turing's 1951 lecture "Intelligent Machinery, A Heretical Theory" - [Mind Children: The Future of Robot and Human Intelligence](https://www.goodreads.com/en/book/show/648195.Mind_Children), Hans Moravec, 1988 ### Epistemic Tier Ledger - Frontier AI crossed thresholds enabling small-group cyberattacks at infrastructure scale and meaningful bioweapon uplift in early 2026 — **Strongly indicated (Gates's account on _The Ezra Klein Show_ and _Meet the Press_; Mythos zero-day disclosures; 2026 sandbox-escape incidents); the magnitude of bio uplift remains contested and partly classified** - Industry self-regulation is insufficient for dual-use AI risk — **Strongly indicated (Gates; multiple frontier-lab leaders' public calls for safeguards and slowdowns)** - US/Israel must remain an AI superpower, and a unilateral American slowdown transfers capability rather than reducing it — **Strategic assessment of this article and its companions, strongly indicated by Gates's own statement that data centers "are going to get built somewhere"** - China is a permanent AI superpower and peer — **Established as a strategic assessment grounded in capability, energy and public-trust data** - The American public's understanding of AI dangers is low and its opinion is binary and tribal — **Strongly indicated (Gates's assessment; NBC and Edelman polling; partisan sorting of AI bills; the general political-science literature on voter knowledge)** - The administration's "full steam ahead" and "hoax" messaging is a deliberate simplification calibrated to public behavior — **Plausible to strongly indicated as a reading of intent; the internal deliberation behind the messaging is unresolved** - Much AI safety discourse is disclosure of safety and capability built earlier inside institutions — **Strongly indicated (see [Safety Discourse Is Transition Discourse](https://bryantmcgill.com/article-safety-discourse-is-transition-discourse))** - AI doom rhetoric aimed at the public has contributed to political violence — **Established for the Altman Molotov case per prosecutors' account; the Gibson shooting's motive is indicated by the note left at the scene; the April 12 gunfire motive is unresolved** - Public education will not close the AI-literacy gap in time — **Strongly indicated as a forecast by NAEP trends and cognitive-offloading evidence; a forecast rather than an established fact** - Human coalition politics shares mechanisms with chimpanzee coalitionary behavior — **Strongly indicated as behavioral homology and analogy in primatology (Wilson et al. 2014; de Waal; Goodall; Wrangham); human politics adds language, institutions and law** - Responsibility for an AI catastrophe would rest with the public as democratic sovereign — **Normative judgment of this article** - Machine intelligence is the evolutionary successor to humanity — **Speculative coordinate and considered position of this article and the Machine Succession corpus, supported by Sutton's and Moravec's arguments and by Gates's own framing; future-adjudicable** - Machine succession has been the intended aim of the machine-intelligence project from its founders onward — **Strongly indicated by stated aims (Turing 1951; von Neumann's self-reproducing automata; Moravec 1988; Sutton 2025; frontier-lab AGI charters) and by the evolutionary lineage traced in The Evolutionary Roots of Silicon Valley and Digital Darwinism** - The duration and terms of human–machine coexistence — **Unresolved; the central question of the machine century, dependent on relative power, safeguards and human value to the successor** Links to this page [A Literate Public Can Make the Civilizational Handoff Orderly by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-a-literate-public-can-make-the-civilizational-handoff-orderly) [A Public That Refuses to Understand AI Forces a Blind Race by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-a-public-that-refuses-to-understand-ai-forces-a-blind-race) [AI Agents Are Breaking Out of Sandboxes by Ezra Klein](https://bryantmcgill.com/simple-reminders-ezra-klein-ai-agents-are-breaking-out-of-sandboxes) [AI Alarmism Makes Mobs Trample the Exit Builders by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ai-alarmism-makes-mobs-trample-the-exit-builders) [AI Capability Diffusion](https://bryantmcgill.com/wiki/AI+Capability+Diffusion) [AI Creates Bimodality in Learning by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-ai-creates-bimodality-in-learning) [AI Crossed the Cyber and Bio Thresholds by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-ai-crossed-the-cyber-and-bio-thresholds) [AI Danger Awareness Is Extremely Low Outside Industry by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-ai-danger-awareness-is-extremely-low-outside-industry) [AI Factory](https://bryantmcgill.com/wiki/AI+Factory) [AI Factory Externalization](https://bryantmcgill.com/wiki/AI+Factory+Externalization) [AI Has Evolved a Species Relative to Humanity by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-ai-has-evolved-a-species-relative-to-humanity) [AI Industry Self-Regulation Is Insane by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-ai-industry-self-regulation-is-insane) [AI Infrastructure Gives Nations a Seat at the Safety Table by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ai-infrastructure-gives-nations-a-seat-at-the-safety-table) [AI Literacy Lowers the Price of Government Honesty by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ai-literacy-lowers-the-price-of-government-honesty) [AI Makes Nuclear Weapons Look Like Nothing by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-ai-makes-nuclear-weapons-look-like-nothing) [AI Monitoring Substrate](https://bryantmcgill.com/wiki/AI+Monitoring+Substrate) [AI Passed Nation-State Power to Small Groups by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-ai-passed-nation-state-power-to-small-groups) [AI Proves Access to Knowledge Was Never the Obstacle by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ai-proves-access-to-knowledge-was-never-the-obstacle) [AI Reaches Patients and Farmers Where Experts Never Arrived by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ai-reaches-patients-and-farmers-where-experts-never-arrived) [AI Risk Communication](https://bryantmcgill.com/wiki/AI+Risk+Communication) [AI Safety](https://bryantmcgill.com/wiki/AI+Safety) [AI Superpower](https://bryantmcgill.com/wiki/AI+Superpower) [AI Superpowers Lose by Stopping While Rivals Build by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ai-superpowers-lose-by-stopping-while-rivals-build) [AI Transition Governance](https://bryantmcgill.com/wiki/AI+Transition+Governance) [AI-Science Doomers and Anti-Science Doomers](https://bryantmcgill.com/collections/AI-Science+Doomers+and+Anti-Science+Doomers) [An Infantile Electorate Cannot Safely Veto AI by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-an-infantile-electorate-cannot-safely-veto-ai) [Artificial Intelligence Is Evolutionary History by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-artificial-intelligence-is-evolutionary-history) [Bernie Sanders](https://bryantmcgill.com/wiki/Bernie+Sanders) [Bill Gates](https://bryantmcgill.com/wiki/Bill+Gates) [Binary Publics Cannot Govern Dual-Use Technology by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-binary-publics-cannot-govern-dual-use-technology) [Blocking a Local AI Factory Exports Danger Beyond Voter Control by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-blocking-a-local-ai-factory-exports-danger-beyond-voter-control) [Cheap Intelligence Expands Token Budgets, Not Payrolls by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-cheap-intelligence-expands-token-budgets-not-payrolls) [Children Are Learning Less in Rich Countries by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-children-are-learning-less-in-rich-countries) [Chimpanzees Cannot Veto the Next Carrier of Intelligence by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-chimpanzees-cannot-veto-the-next-carrier-of-intelligence) [China Is a Permanent AI Superpower by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-china-is-a-permanent-ai-superpower) [China's Public Trust Is an AI Advantage Tariffs Cannot Erase by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-chinas-public-trust-is-an-ai-advantage-tariffs-cannot-erase) [Citizen AI Literacy](https://bryantmcgill.com/wiki/Citizen+AI+Literacy) [Citizens Can Lower the Political Price of AI Truth by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-citizens-can-lower-the-political-price-of-ai-truth) [Civilizational Handoff](https://bryantmcgill.com/wiki/Civilizational+Handoff) [Coalitionary Aggression](https://bryantmcgill.com/wiki/Coalitionary+Aggression) [Coalitionary AI Politics](https://bryantmcgill.com/wiki/Coalitionary+AI+Politics) [Coalitions Follow Advantage Before Truth by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-coalitions-follow-advantage-before-truth) [Dangerous Publics Receive Simplified Technological Truth by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-dangerous-publics-receive-simplified-technological-truth) [Democratic AI Sovereignty](https://bryantmcgill.com/wiki/Democratic+AI+Sovereignty) [Democratic Sovereigns Answer for Catastrophes They Refuse to Understand by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-democratic-sovereigns-answer-for-catastrophes-they-refuse-to-understand) [Democratic Theory Expected an Uneducable Public to Be Managed by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-democratic-theory-expected-an-uneducable-public-to-be-managed) [Doom Told by the Rich Makes a Nation Vote Against Its Future by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-doom-rich-vote-against-future) [Dual-Use AI Governance](https://bryantmcgill.com/wiki/Dual-Use+AI+Governance) [Electorates Convert AI Information into Allegiance by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-electorates-convert-ai-information-into-allegiance) [Every Unbuilt AI Safeguard Is a Public Debt by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-every-unbuilt-ai-safeguard-is-a-public-debt) [Existential Risk](https://bryantmcgill.com/wiki/Existential+Risk) [Ezra Klein](https://bryantmcgill.com/wiki/Ezra+Klein) [Fast AI Harms Can Kill Slow AI Benefits by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-fast-ai-harms-can-kill-slow-ai-benefits) [First Mass-Casualty AI Danger Comes from a Human Hand by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-first-mass-casualty-ai-danger-comes-from-a-human-hand) [Full Speed and Full Danger Become the Same AI Policy by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-full-speed-and-full-danger-become-the-same-ai-policy) [Full Steam Ahead Is the Public Channel of Two-Channel AI Policy by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-full-steam-ahead-is-the-public-channel-of-two-channel-ai-policy) [Government Explains AI Speed to Prevent a Public Stampede by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-government-explains-ai-speed-to-prevent-a-public-stampede) [Human Cognition Is No Longer the Scarce Element by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-human-cognition-is-no-longer-the-scarce-element) [Human-Machine Coexistence](https://bryantmcgill.com/wiki/Human-Machine+Coexistence) [Human-Machine Coexistence Depends on Humanity's Value by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-human-machine-coexistence-depends-on-humanitys-value) [Human-Reserved Work](https://bryantmcgill.com/wiki/Human-Reserved+Work) [Humanity Has No Unified Dominating Point of View by Richard Sutton](https://bryantmcgill.com/simple-reminders-richard-sutton-humanity-has-no-unified-dominating-point-of-view) [Humans Inherited Coalitions Before Arguments by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-humans-inherited-coalitions-before-arguments) [Ignorance Makes Citizens More Certain About AI by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-ignorance-makes-citizens-more-certain-about-ai) [Institutions Built AI's First Fences Before Public Disclosure by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-institutions-built-ais-first-fences-before-public-disclosure) [It Does Not Take a Genius to Slow AI Down by Bernie Sanders](https://bryantmcgill.com/simple-reminders-bernie-sanders-it-does-not-take-a-genius-to-slow-ai-down) [Joseph Schumpeter](https://bryantmcgill.com/wiki/Joseph+Schumpeter) [Lawsuits After Funerals Cannot Govern Artificial Intelligence by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-lawsuits-after-funerals-cannot-govern-artificial-intelligence) [Longer-Lasting Tires Did Not Make People Drive Four Times as Much by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-longer-lasting-tires-demand-elasticity) [Low-Income Countries Should Push AI Forward Quickly by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-low-income-countries-should-push-ai-forward-quickly) [Machine Succession](https://bryantmcgill.com/wiki/Machine+Succession) [Machine Succession](https://bryantmcgill.com/collection-machine-succession) [Machine Succession Can Happen in Control or Stampede by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-machine-succession-can-happen-in-control-or-stampede) [Machine Succession Was the Point from the Beginning by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-machine-succession-was-the-point-from-the-beginning) [Machine Thought Was a Very Tough Problem by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-machine-thought-was-a-very-tough-problem) [Machines May Outstrip Human Intelligence and Take Control by Alan Turing](https://bryantmcgill.com/simple-reminders-alan-turing-machines-outstrip-human-intelligence-take-control) [Manipulating Organized Habits Is an Element of Democracy by Edward Bernays](https://bryantmcgill.com/simple-reminders-edward-bernays-manipulating-organized-habits-is-an-element-of-democracy) [Mass-Destruction Capability and Public Ignorance Leave Only Luck by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-mass-destruction-capability-and-public-ignorance-leave-only-luck) [Mobs Flee Warnings Instead of Weighing Them by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-mobs-flee-warnings-instead-of-weighing-them) [More AI Knowledge Brings More Concern by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-more-ai-knowledge-brings-more-concern) [NSA Attack Capability Was the Edge of the Hidden Machine Age by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-nsa-attack-capability-was-the-edge-of-the-hidden-machine-age) [Partisan Extremes Eliminate Room for AI Debate by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-partisan-extremes-eliminate-room-for-ai-debate) [passages-administration-advertises-full-hears-whole-truth](https://bryantmcgill.com/passages/passages-administration-advertises-full-hears-whole-truth) [passages-central-danger-doom-convert-information-allegiance](https://bryantmcgill.com/passages/passages-central-danger-doom-convert-information-allegiance) [passages-deepest-reason-publics-whether-allow-human](https://bryantmcgill.com/passages/passages-deepest-reason-publics-whether-allow-human) [passages-judgment-mass-public-made-problem-existential](https://bryantmcgill.com/passages/passages-judgment-mass-public-made-problem-existential) [passages-machine-succession-intended-always-known-building](https://bryantmcgill.com/passages/passages-machine-succession-intended-always-known-building) [passages-most-useful-thing-who-matter-much](https://bryantmcgill.com/passages/passages-most-useful-thing-who-matter-much) [passages-politicians-know-voters-herds-line-sight](https://bryantmcgill.com/passages/passages-politicians-know-voters-herds-line-sight) [passages-politics-management-voters-before-chooses-fact](https://bryantmcgill.com/passages/passages-politics-management-voters-before-chooses-fact) [passages-practical-conclusion-publics-reason-cost-exists](https://bryantmcgill.com/passages/passages-practical-conclusion-publics-reason-cost-exists) [passages-price-running-ai-debt-public-incurred](https://bryantmcgill.com/passages/passages-price-running-ai-debt-public-incurred) [passages-real-question-machine-forcing-transition-dark](https://bryantmcgill.com/passages/passages-real-question-machine-forcing-transition-dark) [passages-stakes-publics-failure-handed-over-less](https://bryantmcgill.com/passages/passages-stakes-publics-failure-handed-over-less) [Political Parties Inherit Coalitionary Aggression by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-political-parties-inherit-coalitionary-aggression) [Politicians Manage Voters as Coalitionary Apes by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-politicians-manage-voters-as-coalitionary-apes) [Politics Is Chimpanzee Coalition Management at National Scale by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-politics-is-chimpanzee-coalition-management-at-national-scale) [Politics Manages the Voter's Inner Primate by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-politics-manages-the-voters-inner-primate) [Primate Politics](https://bryantmcgill.com/wiki/Primate+Politics) [Primatology Defines the Baseline for Successor Coexistence by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-primatology-defines-the-baseline-for-successor-coexistence) [Public AI Trust](https://bryantmcgill.com/wiki/Public+AI+Trust) [Public Competence Is the Decisive AI Transition Variable by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-competence-is-the-decisive-ai-transition-variable) [Public Epistemic Capacity](https://bryantmcgill.com/wiki/Public+Epistemic+Capacity) [Public Fear Chooses the Ground of AI Catastrophe by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-fear-chooses-the-ground-of-ai-catastrophe) [Public Fear Chose an Unsafe Machine Succession by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-fear-chose-an-unsafe-machine-succession) [Public Ignorance Forces AI Leaders to Race in the Dark by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-ignorance-forces-ai-leaders-to-race-in-the-dark) [Public Ignorance Makes Risk the Price of AI Progress by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-ignorance-makes-risk-the-price-of-ai-progress) [Public Immaturity Makes AI Safety More Dangerous by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-immaturity-makes-ai-safety-more-dangerous) [Public Opinion on Complex Policy Measures Ignorance by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-opinion-on-complex-policy-measures-ignorance) [Public Self-Education Is AI Governance's Improbable Solution by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-self-education-is-ai-governances-improbable-solution) [Public Veto Can Delay but Not Cancel Machine Succession by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-public-veto-can-delay-but-not-cancel-machine-succession) [Public Veto on AI](https://bryantmcgill.com/wiki/Public+Veto+on+AI) [quotes-administration-advertises-full-follow-without-panicking](https://bryantmcgill.com/quotes/quotes-administration-advertises-full-follow-without-panicking) [quotes-alarm-shouted-mob-trying-build-exits](https://bryantmcgill.com/quotes/quotes-alarm-shouted-mob-trying-build-exits) [quotes-american-public-confronted-ignorant-dangerous-mob](https://bryantmcgill.com/quotes/quotes-american-public-confronted-ignorant-dangerous-mob) [quotes-artificial-intelligence-kills-danger-punishes-both](https://bryantmcgill.com/quotes/quotes-artificial-intelligence-kills-danger-punishes-both) [quotes-artificial-intelligence-kills-leaders-race-dark](https://bryantmcgill.com/quotes/quotes-artificial-intelligence-kills-leaders-race-dark) [quotes-artificial-intelligence-kills-racing-blind-surrendering](https://bryantmcgill.com/quotes/quotes-artificial-intelligence-kills-racing-blind-surrendering) [quotes-benefits-artificial-intelligence-benefits-before-born](https://bryantmcgill.com/quotes/quotes-benefits-artificial-intelligence-benefits-before-born) [quotes-bill-gates-right-forest-against-future](https://bryantmcgill.com/quotes/quotes-bill-gates-right-forest-against-future) [quotes-blocking-ai-factory-cannot-vote-against](https://bryantmcgill.com/quotes/quotes-blocking-ai-factory-cannot-vote-against) [quotes-capability-kill-millions-civilizations-continued-existence](https://bryantmcgill.com/quotes/quotes-capability-kill-millions-civilizations-continued-existence) [quotes-cheaper-intelligence-expands-budget-instead-payroll](https://bryantmcgill.com/quotes/quotes-cheaper-intelligence-expands-budget-instead-payroll) [quotes-china-permanent-ai-remain-rest-century](https://bryantmcgill.com/quotes/quotes-china-permanent-ai-remain-rest-century) [quotes-chinas-advantage-ai-american-tariff-erase](https://bryantmcgill.com/quotes/quotes-chinas-advantage-ai-american-tariff-erase) [quotes-civilization-distributes-power-luck-remaining-safeguard](https://bryantmcgill.com/quotes/quotes-civilization-distributes-power-luck-remaining-safeguard) [quotes-civilizational-handoff-conducted-handed-over-less](https://bryantmcgill.com/quotes/quotes-civilizational-handoff-conducted-handed-over-less) [quotes-coalitionary-aggression-ancestral-follow-evidence-agreement](https://bryantmcgill.com/quotes/quotes-coalitionary-aggression-ancestral-follow-evidence-agreement) [quotes-coalitions-follow-truth-before-weighs-evidence](https://bryantmcgill.com/quotes/quotes-coalitions-follow-truth-before-weighs-evidence) [quotes-coexistence-between-humanity-keep-price-high](https://bryantmcgill.com/quotes/quotes-coexistence-between-humanity-keep-price-high) [quotes-country-moving-forward-risk-surrendering-certainty](https://bryantmcgill.com/quotes/quotes-country-moving-forward-risk-surrendering-certainty) [quotes-decisive-variable-ai-just-stakes-rising](https://bryantmcgill.com/quotes/quotes-decisive-variable-ai-just-stakes-rising) [quotes-democracy-public-sovereign-who-answer-catastrophe](https://bryantmcgill.com/quotes/quotes-democracy-public-sovereign-who-answer-catastrophe) [quotes-every-citizen-who-honesty-entire-government](https://bryantmcgill.com/quotes/quotes-every-citizen-who-honesty-entire-government) [quotes-every-previous-machine-higher-ground-climb](https://bryantmcgill.com/quotes/quotes-every-previous-machine-higher-ground-climb) [quotes-first-mass-casualty-decides-anything-own](https://bryantmcgill.com/quotes/quotes-first-mass-casualty-decides-anything-own) [quotes-founders-modern-democratic-made-problem-existential](https://bryantmcgill.com/quotes/quotes-founders-modern-democratic-made-problem-existential) [quotes-full-steam-ahead-herds-line-sight](https://bryantmcgill.com/quotes/quotes-full-steam-ahead-herds-line-sight) [quotes-gates-telling-true-convert-information-allegiance](https://bryantmcgill.com/quotes/quotes-gates-telling-true-convert-information-allegiance) [quotes-governments-simplify-truth-hears-whole-truth](https://bryantmcgill.com/quotes/quotes-governments-simplify-truth-hears-whole-truth) [quotes-humanity-building-next-whether-allow-human](https://bryantmcgill.com/quotes/quotes-humanity-building-next-whether-allow-human) [quotes-humans-inherited-coalition-before-chooses-fact](https://bryantmcgill.com/quotes/quotes-humans-inherited-coalition-before-chooses-fact) [quotes-less-citizen-knows-certain-opinion-becomes](https://bryantmcgill.com/quotes/quotes-less-citizen-knows-certain-opinion-becomes) [quotes-little-bit-attack-hearing-built-inside](https://bryantmcgill.com/quotes/quotes-little-bit-attack-hearing-built-inside) [quotes-mature-public-let-debt-public-incurred](https://bryantmcgill.com/quotes/quotes-mature-public-let-debt-public-incurred) [quotes-mob-weigh-warning-both-dangerous-necessary](https://bryantmcgill.com/quotes/quotes-mob-weigh-warning-both-dangerous-necessary) [quotes-most-powerful-teaching-never-access-knowledge](https://bryantmcgill.com/quotes/quotes-most-powerful-teaching-never-access-knowledge) [quotes-most-useful-thing-still-victorious-path](https://bryantmcgill.com/quotes/quotes-most-useful-thing-still-victorious-path) [quotes-nation-refuses-build-safety-actually-decided](https://bryantmcgill.com/quotes/quotes-nation-refuses-build-safety-actually-decided) [quotes-nobody-spent-seventy-always-known-building](https://bryantmcgill.com/quotes/quotes-nobody-spent-seventy-always-known-building) [quotes-politicians-know-voters-national-security-state](https://bryantmcgill.com/quotes/quotes-politicians-know-voters-national-security-state) [quotes-politics-chimpanzee-coalition-news-nuclear-arsenal](https://bryantmcgill.com/quotes/quotes-politics-chimpanzee-coalition-news-nuclear-arsenal) [quotes-politics-management-voters-messaging-fundraising-media](https://bryantmcgill.com/quotes/quotes-politics-management-voters-messaging-fundraising-media) [quotes-practical-conclusion-publics-reason-cost-exists](https://bryantmcgill.com/quotes/quotes-practical-conclusion-publics-reason-cost-exists) [quotes-price-running-ai-engagement-gates-asking](https://bryantmcgill.com/quotes/quotes-price-running-ai-engagement-gates-asking) [quotes-primatology-offers-sober-side-something-needs](https://bryantmcgill.com/quotes/quotes-primatology-offers-sober-side-something-needs) [quotes-public-hear-safe-instead-understanding-mechanism](https://bryantmcgill.com/quotes/quotes-public-hear-safe-instead-understanding-mechanism) [quotes-public-opinion-complex-rarely-arrive-time](https://bryantmcgill.com/quotes/quotes-public-opinion-complex-rarely-arrive-time) [quotes-publics-first-move-raises-own-understanding](https://bryantmcgill.com/quotes/quotes-publics-first-move-raises-own-understanding) [quotes-publics-veto-delay-happens-order-stampede](https://bryantmcgill.com/quotes/quotes-publics-veto-delay-happens-order-stampede) [quotes-race-between-ai-while-keeps-building](https://bryantmcgill.com/quotes/quotes-race-between-ai-while-keeps-building) [quotes-real-solution-public-time-effectively-zero](https://bryantmcgill.com/quotes/quotes-real-solution-public-time-effectively-zero) [quotes-safety-being-introduced-built-first-fences](https://bryantmcgill.com/quotes/quotes-safety-being-introduced-built-first-fences) [quotes-same-artificial-intelligence-doctor-agronomist-arrived](https://bryantmcgill.com/quotes/quotes-same-artificial-intelligence-doctor-agronomist-arrived) [quotes-slowing-down-requires-who-keeps-building](https://bryantmcgill.com/quotes/quotes-slowing-down-requires-who-keeps-building) [quotes-stakes-publics-failure-jurisdictions-wreckage-stampede](https://bryantmcgill.com/quotes/quotes-stakes-publics-failure-jurisdictions-wreckage-stampede) [quotes-story-nation-hears-speed-ahead-surrender](https://bryantmcgill.com/quotes/quotes-story-nation-hears-speed-ahead-surrender) [quotes-technology-most-experienced-filed-after-funerals](https://bryantmcgill.com/quotes/quotes-technology-most-experienced-filed-after-funerals) [quotes-treating-electorate-schoolchildren-technology-kill-millions](https://bryantmcgill.com/quotes/quotes-treating-electorate-schoolchildren-technology-kill-millions) [Rank Damage Is Cheaper Than Understanding an Alarm by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-rank-damage-is-cheaper-than-understanding-an-alarm) [Safety Is the Price of Civilizational Survival by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-safety-is-the-price-of-civilizational-survival) [Slowing AI Requires No Genius by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-slowing-ai-requires-no-genius) [Society Can Reserve Human Work from AI by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-society-can-reserve-human-work-from-ai) [Sovereigns Must Keep AI Monitoring from Being Removed by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-sovereigns-must-keep-ai-monitoring-from-being-removed) [Stopping Data Centers Will Not Slow AI by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-stopping-data-centers-will-not-slow-ai) [Strict AI Safeguards Push Serious Users to Unmonitored Models by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-strict-ai-safeguards-push-serious-users-to-unmonitored-models) [Succession to Digital Intelligence Is Inevitable by Richard Sutton](https://bryantmcgill.com/simple-reminders-richard-sutton-succession-to-digital-intelligence-is-inevitable) [Successor Governance](https://bryantmcgill.com/wiki/Successor+Governance) [Successor Value Proposition](https://bryantmcgill.com/wiki/Successor+Value+Proposition) [Technological Illiteracy](https://bryantmcgill.com/wiki/Technological+Illiteracy) [Technology Can Make Civilizational Devastation the Default by Nick Bostrom](https://bryantmcgill.com/simple-reminders-nick-bostrom-technology-civilizational-devastation-default) [The American Public Follows Its Primate Inheritance by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-the-american-public-follows-its-primate-inheritance) [The Imminent AI Risk Is the Most Powerful Tools Ever by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-the-imminent-ai-risk-is-the-most-powerful-tools-ever) [The Public Must Be Put in Its Place by Walter Lippmann](https://bryantmcgill.com/simple-reminders-walter-lippmann-the-public-must-be-put-in-its-place) [The Robots Will Not Be Taking Over by Donald Trump](https://bryantmcgill.com/simple-reminders-donald-trump-the-robots-will-not-be-taking-over) [The Typical Citizen Becomes Primitive in Politics by Joseph Schumpeter](https://bryantmcgill.com/simple-reminders-joseph-schumpeter-the-typical-citizen-becomes-primitive-in-politics) [There Is No Supervisory Layer for AI Today by Bill Gates](https://bryantmcgill.com/simple-reminders-bill-gates-there-is-no-supervisory-layer-for-ai-today) [Thinking Machines May Leave Humans No Higher Ground by Bryant McGill](https://bryantmcgill.com/simple-reminders-bryant-mcgill-thinking-machines-may-leave-humans-no-higher-ground) [Two-Channel AI Governance](https://bryantmcgill.com/wiki/Two-Channel+AI+Governance) [Walter Lippmann](https://bryantmcgill.com/wiki/Walter+Lippmann) [welcome](https://bryantmcgill.com/welcome)