{"course":{"slug":"ai-and-you","title":"AI, Safety and What Goes Wrong","url":"https://addaly.com/learn/ai-and-you","level":"beginner","track":"start"},"lesson":{"slug":"the-two-debates","title":"The two arguments, and how to read them","url":"https://addaly.com/learn/ai-and-you/the-two-debates","position":71,"of":73,"minutes":9,"answer":"The near-term and long-term risk arguments are not logically opposed but compete for attention, expert surveys show wide dispersion rather than consensus, and any claim in either direction should be tested for a stated mechanism.","keyIdea":"The near-term and long-term risk arguments are not logically opposed but compete for attention, expert surveys show wide dispersion rather than consensus, and any claim in either direction should be tested for a stated mechanism, a checkable timeline and something that would falsify it."},"body":{"format":"markdown","text":"## Two conversations that keep colliding\n\nPublic argument about AI risk is two arguments, conducted by partly different people, using the same words.\n\n**The near-term argument** concerns what is happening now: discrimination, surveillance, labour conditions, misinformation, concentration of power, environmental cost, automated decisions without recourse. This is the subject of the previous eight modules. Its proponents include researchers in fairness, accountability and transparency, labour organisations, civil liberties groups and affected communities.\n\n**The long-term argument** concerns whether sufficiently capable future systems could pose catastrophic or existential risk. Its proponents include a substantial number of researchers at frontier labs and academic groups working on alignment.\n\nThey are not logically opposed. In practice they compete for attention, funding, regulatory bandwidth and the same scarce expert time, which is why the argument between them is sharper than the underlying disagreement warrants.\n\n## What each side says about the other\n\n**Near-term critics of long-term framing** argue that speculative future risk draws attention away from documented present harm; that it is convenient for companies because it locates the danger in the future rather than the product; that it frames the solution as more capability rather than less deployment; and that it centres the concerns of a small, well-resourced group over people currently being harmed. The critique advanced in *On the Dangers of Stochastic Parrots* and by the researchers around it is the reference text here.\n\n**Long-term critics of near-term framing** argue that present harms, though real, are the kind society has mechanisms for, and that a unprecedented risk deserves attention before rather than after; that dismissing it because the timeline is uncertain is a poor way to handle tail risks; and that the mechanisms being proposed (evaluation, interpretability, controlled deployment) help with present harms too.\n\n## What researchers actually think\n\nSurveys of AI researchers find wide dispersion rather than consensus. A large 2023 survey of published machine learning researchers found a median estimate of around 5% for extremely bad long-run outcomes from advanced AI, with a large minority placing it at 10% or higher and a substantial group placing it near zero. Medians moved across successive surveys, and dispersion did not narrow.\n\nRead that carefully. It is not \"experts say it is fine\" and it is not \"experts say we are doomed\". It is a field with no settled view, in which a nontrivial fraction of practitioners assign a small but non-negligible probability to a catastrophic outcome, and another fraction think the question is confused.\n\n## How to read any claim in this space\n\nFive questions, which work on optimistic and pessimistic claims equally.\n\n**Is there a mechanism?** Not a capability trend line, but a described causal path from here to the claimed outcome, with the steps named.\n\n**Is there a timeline, and has this person's previous timeline passed?** Predictions in this field have a public record. Checking it is a fifteen-minute exercise that reorders your priors considerably.\n\n**What would falsify it?** A claim compatible with every possible observation is not a claim about the world.\n\n**Who benefits if this is believed?** Applies in all directions. A company benefits from the belief that its product is so powerful it needs regulation only that company can afford, and equally from the belief that concerns are overblown. A researcher benefits from the importance of their own field. Note the incentive and then evaluate the argument anyway: incentive is a reason for care, not a refutation.\n\n**Is the confidence proportionate to the evidence?** Both extremes routinely fail this. \"It will certainly be fine\" and \"it will certainly kill us\" are the two positions the evidence supports least.\n\n## Where this leaves you\n\nA defensible position, and the one this course has taken throughout: the documented harms are documented, they are happening to identifiable people now, and the work of reducing them is available today. The long-term question is open, deserves serious people working on it, and does not license either panic or dismissal.\n\nAnd the practices that make present systems safer (measurement, evaluation, containment, human oversight that actually functions, incident reporting, the ability to say no) are the same practices any future system would need. Doing them now is the part of this that requires no forecast to justify."},"check":{"question":"Which is the strongest test to apply to a confident claim about AI risk, whether alarming or reassuring?","options":[{"text":"Whether it names a causal mechanism in steps, and says what would prove it wrong","correct":true},{"text":"Whether the person making it has a financial interest in how things turn out","correct":false,"tempting":"Worth noting, but an incentive is a reason for care, not a refutation, and it applies to nearly everyone in this field."},{"text":"Whether a clear majority of AI researchers would say that they support it","correct":false,"tempting":"Appeals to a consensus that surveys show does not exist. The spread of views is the finding."},{"text":"Whether it fits the rate at which capability has improved over recent years","correct":false,"tempting":"Extending a trend line puts a curve in place of an argument, which is what claims without a mechanism rely on."}],"why":"A claim with named steps and a stated test can be checked. Confidence without them cannot, in either direction."},"prev":{"slug":"incident-reporting","title":"Learning from what went wrong","url":"https://addaly.com/learn/ai-and-you/incident-reporting"},"next":{"slug":"a-personal-policy","title":"The page you write for yourself","url":"https://addaly.com/learn/ai-and-you/a-personal-policy"},"authorship":"Founder and series editor: Dr Mustafa Mun. Written by AI models to a written standard. Source and assessment review continues. Academic review invited.","review":{"status":"Draft, open for academic review"},"cite":"Addaly (2026) 'The two arguments, and how to read them', in AI, Safety and What Goes Wrong. Edited by M. Mun. Addaly. Available at: https://addaly.com/learn/ai-and-you/the-two-debates (Accessed: 21 November 2026).","licence":{"name":"CC BY-NC-SA 4.0","url":"https://creativecommons.org/licenses/by-nc-sa/4.0/"},"reader":{"schema":1,"title":"The two arguments, and how to read them","courseTitle":"AI, Safety and What Goes Wrong","module":{"index":9,"slug":"living-with-it","title":"Living with it: cost, governance and your own practice"},"answer":null,"keyIdea":"The near-term and long-term risk arguments are not logically opposed but compete for attention, expert surveys show wide dispersion rather than consensus, and any claim in either direction should be tested for a stated mechanism, a checkable timeline and something that would falsify it.","hero":{"kind":"plate","image":null,"alt":null,"caption":null},"blocks":[{"t":"h2","id":"two-conversations-that-keep-colliding","runs":[{"text":"Two conversations that keep colliding"}]},{"t":"p","runs":[{"text":"Public argument about AI risk is two arguments, conducted by partly different people, using the same words."}]},{"t":"p","runs":[{"text":"The near-term argument","b":true},{"text":" concerns what is happening now: discrimination, surveillance, labour conditions, misinformation, concentration of power, environmental cost, automated decisions without recourse. This is the subject of the previous eight modules. Its proponents include researchers in fairness, accountability and transparency, labour organisations, civil liberties groups and affected communities."}]},{"t":"p","runs":[{"text":"The long-term argument","b":true},{"text":" concerns whether sufficiently capable future systems could pose catastrophic or existential risk. Its proponents include a substantial number of researchers at frontier labs and academic groups working on alignment."}]},{"t":"p","runs":[{"text":"They are not logically opposed. In practice they compete for attention, funding, regulatory bandwidth and the same scarce expert time, which is why the argument between them is sharper than the underlying disagreement warrants."}]},{"t":"h2","id":"what-each-side-says-about-the-other","runs":[{"text":"What each side says about the other"}]},{"t":"p","runs":[{"text":"Near-term critics of long-term framing","b":true},{"text":" argue that speculative future risk draws attention away from documented present harm; that it is convenient for companies because it locates the danger in the future rather than the product; that it frames the solution as more capability rather than less deployment; and that it centres the concerns of a small, well-resourced group over people currently being harmed. The critique advanced in "},{"text":"On the Dangers of Stochastic Parrots","i":true},{"text":" and by the researchers around it is the reference text here."}]},{"t":"p","runs":[{"text":"Long-term critics of near-term framing","b":true},{"text":" argue that present harms, though real, are the kind society has mechanisms for, and that a unprecedented risk deserves attention before rather than after; that dismissing it because the timeline is uncertain is a poor way to handle tail risks; and that the mechanisms being proposed (evaluation, interpretability, controlled deployment) help with present harms too."}]},{"t":"h2","id":"what-researchers-actually-think","runs":[{"text":"What researchers actually think"}]},{"t":"p","runs":[{"text":"Surveys of AI researchers find wide dispersion rather than consensus. A large 2023 survey of published machine learning researchers found a median estimate of around 5% for extremely bad long-run outcomes from advanced AI, with a large minority placing it at 10% or higher and a substantial group placing it near zero. Medians moved across successive surveys, and dispersion did not narrow."}]},{"t":"p","runs":[{"text":"Read that carefully. It is not “experts say it is fine” and it is not “experts say we are doomed”. It is a field with no settled view, in which a nontrivial fraction of practitioners assign a small but non-negligible probability to a catastrophic outcome, and another fraction think the question is confused."}]},{"t":"h2","id":"how-to-read-any-claim-in-this-space","runs":[{"text":"How to read any claim in this space"}]},{"t":"p","runs":[{"text":"Five questions, which work on optimistic and pessimistic claims equally."}]},{"t":"p","runs":[{"text":"Is there a mechanism?","b":true},{"text":" Not a capability trend line, but a described causal path from here to the claimed outcome, with the steps named."}]},{"t":"p","runs":[{"text":"Is there a timeline, and has this person’s previous timeline passed?","b":true},{"text":" Predictions in this field have a public record. Checking it is a fifteen-minute exercise that reorders your priors considerably."}]},{"t":"p","runs":[{"text":"What would falsify it?","b":true},{"text":" A claim compatible with every possible observation is not a claim about the world."}]},{"t":"p","runs":[{"text":"Who benefits if this is believed?","b":true},{"text":" Applies in all directions. A company benefits from the belief that its product is so powerful it needs regulation only that company can afford, and equally from the belief that concerns are overblown. A researcher benefits from the importance of their own field. Note the incentive and then evaluate the argument anyway: incentive is a reason for care, not a refutation."}]},{"t":"p","runs":[{"text":"Is the confidence proportionate to the evidence?","b":true},{"text":" Both extremes routinely fail this. “It will certainly be fine” and “it will certainly kill us” are the two positions the evidence supports least."}]},{"t":"h2","id":"where-this-leaves-you","runs":[{"text":"Where this leaves you"}]},{"t":"p","runs":[{"text":"A defensible position, and the one this course has taken throughout: the documented harms are documented, they are happening to identifiable people now, and the work of reducing them is available today. The long-term question is open, deserves serious people working on it, and does not license either panic or dismissal."}]},{"t":"p","runs":[{"text":"And the practices that make present systems safer (measurement, evaluation, containment, human oversight that actually functions, incident reporting, the ability to say no) are the same practices any future system would need. Doing them now is the part of this that requires no forecast to justify."}]}],"check":{"question":[{"text":"Which is the strongest test to apply to a confident claim about AI risk, whether alarming or reassuring?"}],"options":[{"text":[{"text":"Whether it names a causal mechanism in steps, and says what would prove it wrong"}],"correct":true,"tempting":null},{"text":[{"text":"Whether the person making it has a financial interest in how things turn out"}],"correct":false,"tempting":[{"text":"Worth noting, but an incentive is a reason for care, not a refutation, and it applies to nearly everyone in this field."}]},{"text":[{"text":"Whether a clear majority of AI researchers would say that they support it"}],"correct":false,"tempting":[{"text":"Appeals to a consensus that surveys show does not exist. The spread of views is the finding."}]},{"text":[{"text":"Whether it fits the rate at which capability has improved over recent years"}],"correct":false,"tempting":[{"text":"Extending a trend line puts a curve in place of an argument, which is what claims without a mechanism rely on."}]}],"order":[3,1,0,2],"why":[{"text":"A claim with named steps and a stated test can be checked. Confidence without them cannot, in either direction."}]},"prev":{"slug":"incident-reporting","title":"Learning from what went wrong","minutes":8},"next":{"slug":"a-personal-policy","title":"The page you write for yourself","minutes":8}}}