{ "id": "S17", "slug": "automated-discovery-step", "title": "AI-directed research produces a reproducible discovery step", "version": "0.5.0", "status": "active", "orientation": "upside", "family": "D", "short_title": "Automated discovery step", "thematics": [ "ai-cyber-digital", "economic-security-supply-chains", "demographics-human-capital" ], "likelihood_type": "systemic", "likelihood_note": "The bands rate the Standard Severe case: a verified discovery step in three scientific fields. The first-pass superintelligence takeoff is carried as the Extreme setting and is not what these bands rate.", "scores": { "likelihood_2y": { "band": 3, "rationale": "Structure prediction and materials screening already run at scale, and one autonomous laboratory has reported synthesising predicted compounds. The scenario requires more: independent reproduction of the whole loop from hypothesis to useful result in three fields, with licensed systems inside a year. In mid-2026 no AI-designed drug had full approval and early automated-laboratory claims were disputed on novelty. A verified step within two years is possible, not likely." }, "likelihood_10y": { "band": 4, "rationale": "The length of tasks AI agents complete has risen steadily, laboratory automation is spreading and the three fields have outputs that can be tested. Ten years allows several model generations and time for replication studies. The band is 4 and not 5 because the definition is demanding: verification in three fields and a halving of development time. Earlier general-purpose technologies took two decades to show in productivity." }, "impact_systemic": { "level": 4, "rationale": "Rated on scale of change. Pharmaceuticals, chemicals, materials and biomanufacturing reprice within five years and the location of discovery moves towards nations with laboratories and compute. The effect on global output inside five years is limited by trials, approval and manufacturing, so level 5 is not reached at Standard Severe. Level 5 applies at the Extreme setting, which carries the superintelligence takeoff." }, "impact_national": { "level": 4, "rationale": "For a highly exposed nation the highest dimension reached is international position. A research base and export sectors built on long development cycles lose standing within a few years where laboratories, evaluators and approval capacity are missing. Security is engaged through dual-use access. Welfare improves where discoveries are captured. The change is large but can be reversed by investment, so level 4." }, "confidence": { "level": "low", "reason": "Single-field precedents exist, but no verified case of the whole loop. The three-field threshold and the 50 per cent figures are specification choices, not observed values." }, "priority": 12, "tier": "II", "source": "model-panel-v0.5" }, "scorecard": { "onset": "months", "duration": "3 years", "warning": "months", "scope": "global", "origin": "accidental (technological)", "external_support": "partial (licensed and open systems)", "policy_response": "current plans executed", "recovery_horizon": "structural" }, "summary": "Independent laboratories verify that AI-directed research systems can formulate hypotheses, run automated experiments and produce useful, reproducible results in three commercially important fields: therapeutic candidates, battery and catalyst materials and engineered enzymes. Licensed and open systems become available within a year. For qualifying programmes, the time from target to validated candidate halves and cost per validated candidate falls 50 per cent over three years. General superintelligence is not assumed at Standard Severe; the first-pass takeoff is carried as the Extreme setting. Nations must supply laboratories, compute, skilled evaluators and approval pathways to capture discoveries, while managing dual-use access and concentration of intellectual property.", "distribution": { "gains": [ "Nations with automated laboratories, compute, skilled evaluators and capable regulators", "Owners of verified research systems, proprietary data and laboratory automation", "Patients and industries in fields where validated candidates reach approval", "Research institutions that direct open systems at neglected local priorities" ], "loses": [ "Contract research and screening businesses that sell long development cycles", "Nations with access to the systems but no laboratories or evaluators to use them", "Holders of early-stage pipelines and patent portfolios overtaken by faster programmes", "Jurisdictions whose regulators and patent offices cannot handle the volume and lose first filings" ] }, "narrative": { "dateline": "November 2031", "text": "The verification report was published in February 2029. Eleven laboratories in seven countries, none funded by the developers, had been given the same systems and the same three tasks. In each field the system proposed hypotheses, designed experiments, ran them on automated equipment and revised its plan from the results. The laboratories reproduced a class of antibiotic candidates active against resistant infections, a solid electrolyte that kept its conductivity through 2,000 charge cycles and an enzyme that reduced the energy used in a common chemical step. The report's conclusion was narrow. The systems failed on problems outside the three fields. Within them they worked, and the results could be repeated.\n\nLicences were on sale by the summer and two open versions followed within the year. Access proved to be the smaller constraint. A system could propose 400 candidates in a week; a national laboratory with manual equipment could test 12. By 2031 the programmes that qualified, those with automated laboratories, reliable compute and staff able to judge the output, had halved the time from target to validated candidate and cut the cost per candidate by about the same. Programmes without those things reported little change.\n\nThe effect on business came quickly. Contract research firms that sold screening by the hour lost a third of their orders in two years. Three large pharmaceutical companies wrote down their early-stage pipelines and bought laboratory capacity. Patent offices received filings in volumes they could not examine, and courts in four jurisdictions gave different answers on whether a discovery made by a system could be patented and by whom. Medicines regulators became the limiting step. The agencies that had prepared rolling review and accepted automated evidence cleared candidates into trials a year ahead of the others, and developers moved their first filings to them. Trials themselves took as long as before.\n\nGovernments handled the dual-use problem unevenly. The biological design capability that produced the antibiotics could be directed at harmful agents, and the open versions could not be recalled. Fourteen countries adopted synthesis screening and tiered access rules; others did not. Defence ministries funded materials programmes on the same systems and classified the results.\n\nBy November 2031 the gains were measurable and concentrated. Five jurisdictions held most of the new patents. Countries that had equipped laboratories, trained evaluators and reformed approval pathways before 2029 captured discoveries in their own priorities, including neglected diseases and local crops. Countries that had relied on licences alone bought the products later, at the owners' prices. Research staff moved towards the equipped sites. No system had exceeded human experts across all fields, and the developers reported no self-sustaining acceleration of their own research. The step was a change in the cost of discovery in three fields. What separated nations was the capacity they had built to use it." }, "anchors": [ { "shock_id": "alphafold-protein-structure-prediction-2020", "date": "November 2020 to October 2024", "event": "AlphaFold protein structure prediction", "what_happened": "AlphaFold 2 predicted protein structures at close to experimental accuracy in the CASP14 blind assessment in November 2020. A public database released predictions for more than 200 million proteins in July 2022. Two of its developers shared the 2024 Nobel Prize in Chemistry.", "calibrates": "An AI system solving one defined scientific problem, verified by an independent blind assessment and then opened to all. The scenario requires the same standard of proof for the whole research loop, in three fields." }, { "shock_id": "gnome-and-a-lab-materials-discovery-2023", "date": "November 2023", "event": "GNoME and the A-Lab", "what_happened": "Google DeepMind reported 2.2 million predicted crystal structures, about 380,000 of them judged stable. An autonomous laboratory at Lawrence Berkeley National Laboratory reported synthesising 41 of 58 target compounds in 17 days. Other researchers later disputed how many of the compounds were new.", "calibrates": "Prediction joined to automated experiment in materials, and the reason the specification requires independent verification before the step counts." }, { "shock_id": "erooms-law-2012", "date": "2012", "event": "Eroom's law", "what_happened": "Scannell and co-authors showed that the number of new drugs approved per billion US dollars of research spending, adjusted for inflation, had halved about every nine years since 1950.", "calibrates": "The baseline the scenario reverses. A 50 per cent fall in cost per validated candidate in three years is large against a 60-year trend in the opposite direction." }, { "shock_id": "covid-19-mrna-vaccines-2020", "date": "2020", "event": "COVID-19 mRNA vaccines", "what_happened": "Vaccines were authorised within about 11 months of the pathogen being sequenced, with regulators reviewing data while trials proceeded.", "calibrates": "Approval pathways can be compressed where regulators have prepared. This is the capability that decides which nations capture discoveries." }, { "shock_id": "ai-designed-drug-candidates-2026", "date": "July 2026", "event": "AI-designed drug candidates", "what_happened": "No AI-designed drug had full approval; the most advanced candidate entered a Phase III trial.", "calibrates": "Counter-anchor on speed. Discovery has accelerated ahead of trials and approval, which the Standard Severe case does not assume are shortened." }, { "shock_id": "metr-time-horizon-measurements-2025", "date": "March 2025 to April 2026", "event": "METR time-horizon measurements", "what_happened": "The length of tasks AI agents complete doubled about every seven months, then faster; in April 2026 laboratories reported no dramatic speed-up of their own research.", "calibrates": "Leading indicator for the step, and counter-evidence for the Extreme setting, which requires AI research itself to be automated." } ], "parameters": { "common": { "severity": { "major": "major (one field verified; systems licensed, not open)", "severe": "severe (verified discovery step in three fields; general superintelligence not assumed)", "extreme": "extreme (general superintelligence takeoff: the first-pass record, in which AI systems automate AI research)" }, "duration": { "major": "3 years", "severe": "3 years", "extreme": "3 years" }, "onset": { "major": "years", "severe": "months", "extreme": "months" }, "warning": { "major": "months", "severe": "months", "extreme": "months" }, "scope": { "major": "global", "severe": "global", "extreme": "global" }, "origin": { "major": "accidental (technological)", "severe": "accidental (technological)", "extreme": "accidental (technological)" }, "external_support": { "major": "partial (licensed systems)", "severe": "partial (licensed and open systems)", "extreme": "none (access on the owners' terms)" }, "concurrency": { "major": "standalone", "severe": "standalone", "extreme": "plus one named scenario (S08)" }, "policy_response": { "major": "current plans executed", "severe": "current plans executed", "extreme": "current plans executed" }, "recovery_horizon": { "major": "years", "severe": "structural", "extreme": "structural (irreversible at Extreme)" } }, "specific": [ { "name": "Capability of the leading systems", "default": "automated discovery in defined fields", "range": "options: automated discovery in defined fields / automated AI research / general superintelligence", "unit": null, "note": "General superintelligence is not assumed at Standard Severe. Extreme: the first-pass takeoff, in which systems automate AI research and outperform human experts in every field." }, { "name": "Scientific fields with independently verified automated discovery", "default": "3", "range": "1 to 6", "unit": "fields", "note": "Default fields: therapeutic candidates, battery and catalyst materials and engineered enzymes. Major: 1. Extreme: every field of science and engineering." }, { "name": "Fall in development time for qualifying programmes", "default": "50 per cent", "range": "25 to 75", "unit": "per cent over three years", "note": "Measured from target to validated candidate. Clinical trial and approval time is not assumed to fall. Extreme: research speed-up of tenfold or more." }, { "name": "Fall in cost per validated candidate", "default": "50 per cent", "range": "25 to 80", "unit": "per cent over three years", "note": "Applies to programmes with automated laboratories, compute and skilled evaluators." }, { "name": "Time from verification to licensed and open systems", "default": "12 months", "range": "6 to 36", "unit": "months", "note": "Extreme: time from automated AI researcher to superintelligence, 12 months." }, { "name": "Access for nations that host no developer", "default": "licensed and open systems", "range": "options: licensed only / restricted by export control", "unit": null, "note": "Extreme: paid access on the owners' terms, with open versions withheld." }, { "name": "Binding constraint on capture", "default": "automated laboratories and skilled evaluators", "range": "options: compute and power / approval pathways / access controls", "unit": null, "note": "Extreme: compute and power." } ] }, "transmission": [ "Independent laboratories reproduce AI-directed hypothesis generation and automated experiment in three fields; the result is verified, not claimed.", "Developers license the systems and open versions follow within a year; access spreads faster than the laboratories needed to use it.", "Qualifying programmes halve development time and cost per validated candidate; candidate volume outruns evaluators, trial capacity and approval pathways.", "Value moves to owners of systems, proprietary data and automated laboratories; research businesses built on long development cycles are repriced.", "Patent offices and regulators face volume and new questions of inventorship and evidence; jurisdictions with capable agencies receive first filings and set standards.", "Dual-use results in biology and materials prompt access tiers, synthesis screening and defence research programmes; open versions limit enforcement.", "Research staff and investment move to equipped sites; nations with laboratories, compute and evaluators capture discoveries in their own priorities.", "Nations with licences but no capacity buy products later at the owners' prices, and their research position declines." ], "loading": [ { "dimension": "Hard", "domain": "defence-security", "load": "High", "channel": "dual-use biology and materials results; defence research programmes run on the same systems" }, { "dimension": "Hard", "domain": "strategic-infrastructure", "load": "High", "channel": "automated laboratories, compute, power and data networks decide who can run qualifying programmes" }, { "dimension": "Hard", "domain": "critical-technology", "load": "High", "channel": "AI research systems, advanced computing, laboratory robotics, biotechnology and materials capability" }, { "dimension": "Soft", "domain": "government-effectiveness", "load": "High", "channel": "approval pathways, patent examination, research funding and dual-use access rules" }, { "dimension": "Soft", "domain": "human-capital", "load": "High", "channel": "skilled evaluators and researchers; health gains where discoveries are captured" }, { "dimension": "Soft", "domain": "influence-cohesion", "load": "Medium", "channel": "standard setting, access agreements and public trust in machine-generated evidence" }, { "dimension": "Economic", "domain": "macro-financial", "load": "Medium", "channel": "repricing of research-intensive assets; public funding for laboratories" }, { "dimension": "Economic", "domain": "industry-trade-supply", "load": "High", "channel": "pharmaceutical, chemical and materials industries repriced; intellectual property concentrates" }, { "dimension": "Economic", "domain": "energy-resources", "load": "Medium", "channel": "electricity for compute and laboratories; new battery and catalyst materials" } ], "stakeholders": { "government": { "exposure": "Research position, approval capacity, dual-use control and the terms on which the nation obtains licensed systems", "actions": [ "Fund automated laboratories and train evaluators before systems are licensed", "Prepare regulators and patent offices for higher volumes and machine-generated evidence", "Set tiered access and synthesis screening rules for dual-use design capability" ], "watch": [ "Independent replication of AI-directed discoveries", "Licence terms and export controls on scientific systems", "Review times at medicines regulators and patent offices" ], "relevance": 5 }, "technology": { "exposure": "System developers and laboratory automation firms gain; compute and data providers must meet research demand and access controls", "actions": [ "Submit systems to independent verification before making commercial claims", "Build misuse screening and access tiers into licensed and open releases", "Secure compute, power and laboratory robotics supply for research customers" ], "watch": [ "Task-horizon and research-automation benchmarks", "Open-weight releases of scientific models", "Rules on model weights and biological design tools" ], "relevance": 5 }, "investors": { "exposure": "Repricing of research-intensive sectors: early-stage pipelines, contract research and patent portfolios lose value while owners of systems and laboratories gain", "actions": [ "Test holdings on halved development time and cost per candidate", "Map portfolio exposure to businesses that sell long development cycles", "Assess patent validity risk where inventorship rules are unsettled" ], "watch": [ "Candidates entering trials from automated programmes", "Write-downs of early-stage pipelines", "Concentration of new patents by owner and jurisdiction" ], "relevance": 4 }, "public": { "exposure": "Earlier access to new medicines and materials where they are approved and affordable; research employment; biosafety", "actions": [ "Ask how public research funding secures access to the resulting products", "Support training in evaluating and checking automated research", "Hold local institutions to biosafety and screening standards" ], "watch": [ "Time from approval to availability in the national health system", "Research and technical hiring", "Reported biosafety incidents and screening coverage" ], "relevance": 4 } }, "regions": [ { "region": "north-america", "exposure": "High", "rationale": "Home to most system developers and much automated laboratory capacity; gains, patents and the dual-use control problem concentrate here" }, { "region": "europe", "exposure": "High", "rationale": "Deep research base and large pharmaceutical and chemical industries; approval speed decides whether discoveries are captured or filed elsewhere" }, { "region": "china", "exposure": "High", "rationale": "Laboratory scale, state research funding and open-weight models support fast adoption; access to advanced chips sets the pace" }, { "region": "indo-pacific", "exposure": "High", "rationale": "Japan, Korea, Taiwan, Singapore and Australia hold materials, chip and biomanufacturing capacity; much of ASEAN and the Pacific depends on licences" }, { "region": "south-asia", "exposure": "Medium", "rationale": "Contract research and generics face repricing; a large trained workforce can capture gains if automated laboratories are built" }, { "region": "gulf-middle-east", "exposure": "Medium", "rationale": "Capital and power can buy compute and laboratories; skilled evaluators and a domestic research base are scarce outside Israel" }, { "region": "africa", "exposure": "Medium", "rationale": "Large potential gains in neglected diseases and crops; few laboratories or evaluators to direct the systems at local priorities" }, { "region": "latin-america", "exposure": "Medium", "rationale": "Strong public research institutes in a few states; regulatory capacity and funding limit capture elsewhere" }, { "region": "russia-eurasia", "exposure": "Medium", "rationale": "Skilled scientists but restricted access to chips and licensed systems; open versions help and defence applications take priority" } ], "indicators": [ { "name": "Independent replication of AI-directed discoveries", "source": null, "threshold": "Whole research loop reproduced by independent laboratories in three fields", "cadence": "as published" }, { "name": "Length of tasks AI agents complete autonomously", "source": "METR", "threshold": null, "cadence": "per model release" }, { "name": "AI-designed drug candidates in Phase II and Phase III trials", "source": "ClinicalTrials.gov, WHO International Clinical Trials Registry Platform", "threshold": "First full regulatory approval of an AI-designed medicine", "cadence": "quarterly" }, { "name": "Novel medicine approvals and review times", "source": "US Food and Drug Administration, European Medicines Agency", "threshold": null, "cadence": "annual" }, { "name": "Frontier compute and data-centre power under construction", "source": "Epoch AI, IEA", "threshold": null, "cadence": "annual" }, { "name": "Open-weight releases of scientific and biological design models", "source": "Stanford HAI AI Index", "threshold": null, "cadence": "annual" }, { "name": "Export controls on chips, model weights and biological design tools", "source": "US Bureau of Industry and Security", "threshold": "Controls extended to scientific research systems", "cadence": "as issued" }, { "name": "Research spending and researchers per million population", "source": "OECD Main Science and Technology Indicators, UNESCO Institute for Statistics", "threshold": null, "cadence": "annual" } ], "compounds": { "triggers": [ { "id": "S08", "note": "the same systems extend from research to professional workflows; certain at the Extreme setting" }, { "id": "S15", "note": "materials discovery for storage and firm clean generation" }, { "id": "S19", "note": "faster and cheaper candidate discovery for metabolic and preventive treatments" }, { "id": "S20", "note": "automated design of diagnostics and countermeasures shortens the response" }, { "id": "S10", "note": "only if dual-use access controls fail; a tail risk, not the central case" }, { "id": "S09", "note": "AI-directed research could shorten the path to fault-tolerant hardware or to better algorithms." }, { "id": "S16", "note": "AI-directed research shortens breeding and trait validation." } ], "triggered_by": [ { "id": "S08", "note": "the same step in agent reliability, applied to research" }, { "id": "S13", "note": "relaxed technology restrictions widen access to chips and licensed systems" } ], "amplified_by": [ { "id": "S01", "note": "bloc controls on chips and systems narrow access and widen the gap between nations" }, { "id": "S15", "note": "cheap firm power eases the compute and laboratory constraint" } ], "triggered_by_other": [], "amplifying_trends": [ "Automated AI research", "AI-accelerated science", "Compute scaling & AI infrastructure", "Open-weight AI diffusion", "Advanced materials", "Synthetic biology & biomanufacturing", "Research & development intensity", "Biosecurity & dual-use biology" ] }, "open_questions": [ "Whether the first-pass superintelligence takeoff, carried here as the Extreme setting, is the same event at higher severity or a different scenario that needs its own record and a paired loss-of-control case.", "Whether halved development time covers discovery up to a validated candidate only, as this record assumes, or also trials and approval, where clinical and regulatory capacity is the constraint.", "Who defines independent verification and by what test, given that early claims of automated discovery have been disputed on novelty.", "Whether a scenario that loads defence and security High through dual-use access belongs on the upside list without a paired misuse case." ], "commentary": null, "overview": { "source": "model-panel-v0.5", "lenses": { "political": "Governments compete on laboratories, evaluators and approval pathways. **Fourteen countries adopt tiered access rules** for dual-use design capability; others do not.", "economic": "Qualifying programmes **halve time and cost per validated candidate**. Contract research and early-stage pipelines built on long cycles are written down.", "social": "Patients gain where regulators and health systems can absorb new candidates. **Research staff move to equipped sites**, widening the gap between nations.", "technological": "Eleven independent laboratories reproduce the whole loop in **three fields, not all of science**. Automated laboratories, not access to systems, become the constraint.", "legal": "**Inventorship and evidence rules** diverge across jurisdictions. Patent offices and medicines regulators become the limiting step.", "environmental": "New electrolytes, catalysts and enzymes **reduce energy and material use** in industry, while compute and automated laboratories add electricity demand." }, "regions": { "north-america": "**Home to most system developers** and much automated laboratory capacity. Gains and patents concentrate here first, with the dual-use control problem.", "europe": "A deep research base and large pharmaceutical and chemical industries. **Approval speed** decides whether discoveries are captured or filed elsewhere.", "china": "Laboratory scale, state research funding and open-weight models support fast adoption. **Access to advanced chips** sets the pace.", "indo-pacific": "Japan, Korea, Taiwan, Singapore and Australia hold **materials, chip and biomanufacturing capacity**. Much of ASEAN and the Pacific depends on licences.", "south-asia": "**Contract research and generics** face repricing. A large trained workforce can capture gains if automated laboratories are built.", "gulf-middle-east": "Capital and power can **buy compute and laboratories**. The shortage is skilled evaluators and a domestic research base.", "africa": "Large potential gains in neglected diseases and crops, with **few laboratories or evaluators** to direct the systems at local priorities.", "latin-america": "Strong public research institutes in a few states. **Regulatory capacity and funding** limit capture; most nations buy products later.", "russia-eurasia": "Skilled scientists but **restricted access to chips and licensed systems**. Open versions help and defence applications take priority." }, "evolution": { "lead_indicators": "**Independent laboratories reproduce** AI-directed discoveries. The length of tasks AI agents complete keeps rising and the first AI-designed medicine gains full regulatory approval.", "triggers": "Eleven laboratories in seven countries, none funded by the developers, **reproduce the whole research loop in three fields**: therapeutic candidates, battery and catalyst materials and engineered enzymes.", "immediate": "Licences are on sale within months and **open versions follow within the year**. Access spreads faster than the automated laboratories needed to test what the systems propose.", "post_immediate": "Qualifying programmes **halve time and cost per validated candidate** over three years. Contract research firms lose a third of their orders and early-stage pipelines are written down.", "stabilisation": "**Regulators and patent offices become the limiting step.** Prepared agencies clear candidates into trials a year ahead and receive first filings. Trials take as long as before.", "long_term": "**Five jurisdictions hold most new patents** and research staff move to equipped sites. Nations with licences but no laboratories buy products later, at the owners' prices." } }, "history": [ { "edition": 2027, "rank": null, "L2y": 3, "L10y": 4, "impact_systemic": 4, "impact_national": 4, "confidence": "low" } ], "changelog": [ { "version": "0.5.0", "date": "2026-10-09", "change": "Rebuilt in the second pass from Superintelligence takeoff: the Standard Severe case is now a verified discovery step in three scientific fields, both impact levels fall from 5 to 4 and the takeoff is carried as the Extreme setting." } ], "citation": "GINC (2027). Scenario S17 AI-directed research produces a reproducible discovery step, Scenario Library v0.5. scenarios.ginc.org/scenarios/automated-discovery-step" }