--- id: ASI-PATH-001 title: "Pathways to Superintelligence" alternative_names: - "Routes to Superintelligence" - "Paths to ASI" category: "Advanced Intelligence → Superintelligence" volume: VI chapter: 32 entry_type: "Technical Article" level: 4 status: "Draft" last_reviewed: 2026-10-15 version: "0.1.0" prerequisites: - "ASI-001" - "ASI-EXP-001" - "AGI-PATH-001" related_entries: - "ASI-DEF-001" - "ASI-AIRD-001" - "TECH-SCALE-001" - "SAFE-OVERSIGHT-001" - "ECON-INFRA-001" --- # Pathways to Superintelligence ## 1. Concise Definition **Pathways to superintelligence** are the proposed technical and social routes by which an intellect greatly exceeding human performance might come to exist. The best-known classification, from Bostrom (2014), distinguishes artificial intelligence, whole brain emulation, biological cognitive enhancement, brain-computer interfaces, and networks and organisations. All of these remain Hypothetical: no route has produced a system that meets a general definition of superintelligence (see [ASI-DEF-001](../vol06_asi/ASI-DEF-001_defining_and_measuring_superintelligence.md)). ## 2. Overview & Significance [AGI-PATH-001](../vol05_agi/AGI-PATH-001_pathways_to_agi.md) covers routes to human-level machine intelligence, and [ASI-EXP-001](../vol06_asi/ASI-EXP-001_intelligence_explosion.md) covers the hypothesis that capability could grow rapidly once certain thresholds are passed. This article is concerned with the intermediate question: given human-level systems, or without them, by what mechanisms might capability go substantially beyond the human range, and what would each mechanism require? The question matters because the routes differ in the following respects, which bear on safety and governance. - **Predictability.** A route that extends known scaling trends is easier to anticipate than one that requires a conceptual breakthrough. - **Speed and discontinuity.** Routes differ in whether progress is likely to be smooth, stepwise or abrupt. - **Controllability.** Whether the resulting system's goals and behaviour can be inspected and shaped differs between routes. For example, an emulation of a human brain would inherit human-like psychology in some respects, while an artificial system trained by optimisation may not. - **Actors and infrastructure.** Different routes require different resources, which determines who could pursue them (see [ECON-INFRA-001](../vol09_economy/ECON-INFRA-001_energy_data_centers_and_semiconductor_supply_chains.md)). Beyond classification, this article identifies what evidence bears on each route and which claims are matters of argument rather than data. ## 3. Historical Development Good (1966) raised the possibility that a machine surpassing human intellect could improve its own design. Bostrom (1998) discussed estimates of the brain's processing power, when hardware might match it, and how quickly superintelligence could follow human-level AI; the paper's abstract argues that superhuman AI is likely within the first third of the twenty-first century [VERIFICATION NEEDED: wording of the timing claim taken from a secondary summary of the abstract]. No system meeting the definition exists as of this review, and the date is mentioned only as an example of a timing claim in this literature. Sandberg and Bostrom (2008) produced a roadmap for whole brain emulation, the replication of a brain's function in a computer, following a workshop at the Future of Humanity Institute in Oxford in 2007. Bostrom (2014) then set out the classification used here and analysed the dynamics of a transition. In the machine-learning era, the empirical discussion turned to scaling laws (Kaplan et al., 2020; Hoffmann et al., 2022), to self-play systems that exceeded human performance without human game data (Silver et al., 2017), and to the question of whether AI systems could take part in AI research itself (see [ASI-AIRD-001](../vol06_asi/ASI-AIRD-001_automated_ai_research.md)). ## 4. Technical / Conceptual Explanation ### 4.1 Pathway A: Artificial intelligence This is the route most discussed in current research and policy. It has several mechanisms, which can combine. **A1. Scaling of learning systems.** Kaplan et al. (2020) reported that the cross-entropy loss of language models falls as a power law with model size, dataset size and training compute, over the ranges they studied. Hoffmann et al. (2022) reported, for compute-optimal training, that model size and training tokens should be scaled in roughly equal proportion. These are empirical regularities for a loss metric on a given model family, and extrapolating them to superhuman capability is an inference, not a result. A stylised form is $$L(N, D) \approx E + \frac{A}{N^{\alpha}} + \frac{B}{D^{\beta}},$$ with $N$ parameters, $D$ tokens and fitted constants $E, A, B, \alpha, \beta$; the constants and functional form are from the fitted literature and differ across studies. Whether lower loss translates into capabilities beyond human experts across domains is Contested. **A2. Algorithmic progress.** Ho et al. (2024) analysed more than 200 language-model evaluations on two benchmarks from 2012 to 2023 and estimated that the compute needed to reach a fixed performance threshold halved approximately every 8 months, with a 95% confidence interval of roughly 5 to 14 months. They also concluded that increased compute contributed more to overall performance gains than algorithmic innovations did, and noted that noisy benchmark data limits the precision of the estimates. These figures concern those benchmarks and that period; they are not a forecast of future algorithmic efficiency. **A3. Self-play and search.** Silver et al. (2017) reported a Go system trained by reinforcement learning from self-play without human game data that outperformed earlier human-data-based systems. This demonstrates a mechanism by which performance beyond the best humans can be reached in a domain with a fixed, perfectly specified objective and cheap simulation. Whether the mechanism transfers to open-ended domains without such an objective is Contested and depends on the availability of reliable verification. **A4. Automated AI research.** If AI systems become able to do research that improves AI systems, feedback could accelerate progress. This is the mechanism behind the intelligence-explosion hypothesis; its evidence status is discussed in [ASI-AIRD-001](../vol06_asi/ASI-AIRD-001_automated_ai_research.md) and [ASI-EXP-001](../vol06_asi/ASI-EXP-001_intelligence_explosion.md). Erdil and Besiroglu (2023) review arguments for explosive economic growth from AI automation, identify drivers (scalable AI systems, a fast increase in the number of AI workers, and a sharp jump in output from automation) and assess counterarguments such as regulation and production bottlenecks. They conclude that explosive growth is plausible if AI can largely replace human labour, but that high confidence is not yet justified. Hutter (2012) analysed conditions under which an intelligence explosion would or would not occur, in a mathematical framework; the conclusions depend on assumptions about how intelligence translates into the ability to improve itself. **A5. Amplified supervision and oversight.** Christiano et al. (2018) proposed iterated amplification, a training scheme for tasks that humans cannot directly evaluate, in which a weak expert is amplified by help from copies of the learning system. Burns et al. (2023) test weak-to-strong generalisation as an analogy for supervising systems beyond human ability. These are not routes to superintelligence in themselves but research on whether a system trained under such schemes can safely be capable beyond its supervisors (see [SAFE-OVERSIGHT-001](../vol08_safety_governance/SAFE-OVERSIGHT-001_scalable_oversight_and_corrigibility.md)). ### 4.2 Pathway B: Whole brain emulation Sandberg and Bostrom (2008) describe emulation as obtaining a detailed scan of a brain and implementing its function in a computer, and discuss the required technologies: scanning, modelling the neural substrate and computation. The report was written to examine feasibility rather than to predict a date. Emulation would yield human-like minds, possibly run faster or copied; this corresponds to the speed or collective forms in Bostrom (2014). The approach is bounded by the state of neuroscience and scanning technology, and no human brain has been emulated. As of the sources reviewed, this pathway is Hypothetical, and its relation to AI progress is debated: some argue that AI will arrive first. The sources reviewed contain no evidence settling that question. ### 4.3 Pathway C: Biological cognitive enhancement Genetic selection, genetic modification, pharmacology and training could in principle raise human cognitive ability. Bostrom (2014) discusses this route and argues that biological enhancement is likely to be slow relative to machine routes because of generation times, though the discussion is argumentative and the claim is Contested. [VERIFICATION NEEDED: the specific chapter and wording of Bostrom's comparison of biological and machine routes were not re-checked]. Ethical and legal constraints are significant and outside the scope of this article. ### 4.4 Pathway D: Brain-computer interfaces Interfaces between brains and computers could extend human cognition. The sources reviewed for this article do not provide evidence that current interfaces enhance general cognition beyond the human range; this pathway is Hypothetical. ### 4.5 Pathway E: Networks and organisations Collective intelligence arises from organisations, markets and the internet. Bostrom (2014) includes the strengthening of such networks as a route to collective superintelligence. Human institutions already outperform individuals at many tasks, which raises the question, treated in [ASI-DEF-001](../vol06_asi/ASI-DEF-001_defining_and_measuring_superintelligence.md), of how the reference class is defined. ## 5. Types, Components & Classifications | Pathway | Core mechanism | Main bottleneck | Evidence status | | :--- | :--- | :--- | :--- | | A1 Scaling | Compute, data and parameters | Compute and data supply; whether loss reflects capability | Scaling regularities Well Supported for loss; extrapolation to superhuman capability Contested | | A2 Algorithmic progress | Efficiency improvements | Measurement noise; diminishing returns | Preliminary (Ho et al., 2024) | | A3 Self-play and search | Optimisation against a verifier or simulator | Availability of a verifiable objective | Well Supported in games; open-ended extension Contested | | A4 Automated AI research | Feedback from AI to AI progress | Capability at research tasks; physical constraints | Hypothetical | | B Whole brain emulation | Scanning and modelling a brain | Scanning and neuroscience | Hypothetical | | C Biological enhancement | Selection and modification | Slow generations; constraints | Hypothetical | | D Brain-computer interface | Hardware linking brain and machine | Bandwidth; safety | Hypothetical | | E Networks and organisations | Coordination of many minds | Coordination costs | Contested | Pathways are not exclusive. For instance, scaling, algorithmic progress and automated research may operate together, and Pathway A is generally treated as the principal route in current discussion because it is the one on which active research is concentrated. ## 6. Applications & Concrete Examples There are no applications of superintelligence; the following are partial demonstrations of mechanisms. - **Superhuman in a narrow domain by self-play.** Silver et al. (2017). A laboratory result in a fixed-rule game. - **Measured scaling behaviour.** Kaplan et al. (2020); Hoffmann et al. (2022). Empirical regularities for language-model loss. - **Measured algorithmic efficiency.** Ho et al. (2024), for language models on the benchmarks they studied. - **Oversight analogues.** Burns et al. (2023) report experiments with weak supervisors and strong models as an analogue; Christiano et al. (2018) is a proposal for amplified supervision. - **Emulation.** Sandberg and Bostrom (2008) is a roadmap and not an implementation. Readers should not infer from any of these that a particular pathway is close to completion. ## 7. Advantages, Limitations & Unresolved Questions **Value of the pathway framework.** It structures a speculative field, identifies the bottlenecks specific to each route and clarifies which empirical observations bear on which claims. **Limitations.** - Pathways were classified before the current generation of AI systems, and the boundaries between them have shifted. - Evidence for the machine-learning pathway concerns loss, benchmark performance and narrow domains, not general superhuman capability. - Hardware, energy and data constraints may bind before algorithmic limits do. These are covered in [ECON-INFRA-001](../vol09_economy/ECON-INFRA-001_energy_data_centers_and_semiconductor_supply_chains.md) and [ECON-COMP-001](../vol09_economy/ECON-COMP-001_compute_economics.md). - The relevant physical and theoretical limits to intelligence are not known. Skeptical arguments, for example about diminishing returns to self-improvement, are discussed in [ASI-EXP-001](../vol06_asi/ASI-EXP-001_intelligence_explosion.md). **Open questions.** 1. Does scaling plus algorithmic progress suffice to reach general superhuman capability, or are new ideas required? Contested. 2. Do verification-limited domains (research, strategy, social reasoning) permit the kind of self-play improvement seen in games? 3. Would automated AI research accelerate progress, or be limited by compute and experiments? See [ASI-AIRD-001](../vol06_asi/ASI-AIRD-001_automated_ai_research.md). 4. How would emulation or enhancement pathways interact with AI progress? ## 8. Current State of Knowledge As of 2026-10-15, no pathway has produced a system meeting a general definition of superintelligence, and the sources reviewed for this article do not claim otherwise. The best-documented mechanisms are in the machine-learning pathway: loss scaling regularities (Kaplan et al., 2020; Hoffmann et al., 2022), measured algorithmic efficiency gains (Ho et al., 2024), and superhuman narrow performance via self-play (Silver et al., 2017). Whether these extend to general superhuman capability is Contested. Whole brain emulation, biological enhancement and brain-computer interfaces are Hypothetical, in the sense that no demonstration relevant to superintelligence is documented in the sources reviewed. For AGI timeline surveys see [AGI-TIME-001](../vol05_agi/AGI-TIME-001_agi_forecasting.md); for quantitative current inputs (compute, model sizes, costs), consult the Living Reference (`MODEL-001`, `FACT-001`) because such figures change quickly. ## 9. Future Trajectory & Strategic Outlook This section is descriptive. It registers no forecast; dated forecasts, if any, belong in the Forecast Register (`FCST-REG-001`). - **Near term (0 to 3 years).** Continued research on scaling, algorithmic efficiency, agentic systems and automated research, together with oversight research on supervising systems whose outputs are difficult to verify. - **Medium term (3 to 10 years).** Evidence on whether verification-limited domains yield to the methods that succeeded in games and on whether automated research accelerates AI progress would discriminate between the AI mechanisms. The sources reviewed do not indicate how these questions will resolve. - **Long term (10+ years).** Whole brain emulation and enhancement pathways cannot be assessed from the sources reviewed beyond their roadmaps. Scenarios about the aftermath of superintelligence are covered in [PHIL-CIV-001](../vol11_philosophy/PHIL-CIV-001_civilization_and_advanced_intelligence.md). ## 10. Related Entries & Knowledge Graph - **Foundational Concepts**: - [[ASI-001] Artificial Superintelligence](../vol06_asi/ASI-001_artificial_superintelligence.md): Overview of ASI. - [[ASI-DEF-001] Defining and Measuring Superintelligence](../vol06_asi/ASI-DEF-001_defining_and_measuring_superintelligence.md): What counts as superintelligence. - [[AGI-PATH-001] Pathways to AGI](../vol05_agi/AGI-PATH-001_pathways_to_agi.md): The preceding stage of the argument. - **Related Methods & Technologies**: - [[TECH-SCALE-001] Scaling Laws](../vol03_technical/TECH-SCALE-001_scaling_laws.md): Scaling regularities in detail. - [[ASI-AIRD-001] Automated AI Research](../vol06_asi/ASI-AIRD-001_automated_ai_research.md): Mechanism A4. - [[ECON-INFRA-001] Energy, Data Centers and Semiconductor Supply Chains](../vol09_economy/ECON-INFRA-001_energy_data_centers_and_semiconductor_supply_chains.md): Physical constraints. - [[SAFE-OVERSIGHT-001] Scalable Oversight and Corrigibility](../vol08_safety_governance/SAFE-OVERSIGHT-001_scalable_oversight_and_corrigibility.md): Supervising systems beyond human ability. - **Advanced Implications & Future Trajectories**: - [[ASI-EXP-001] The Intelligence Explosion Hypothesis](../vol06_asi/ASI-EXP-001_intelligence_explosion.md): Takeoff dynamics. - [[SAFE-XRISK-001] AI Existential Risk & Catastrophic Scenarios](../vol08_safety_governance/SAFE-XRISK-001_existential_risk.md): Risk implications. - [[PHIL-CIV-001] Civilization and Advanced Intelligence](../vol11_philosophy/PHIL-CIV-001_civilization_and_advanced_intelligence.md): Long-run scenarios. ## 11. References & Foundational Literature 1. Bostrom, N. (1998). *How Long Before Superintelligence?* International Journal of Futures Studies, 2. https://nickbostrom.com/superintelligence. [Early multi-route discussion of superintelligence and hardware estimates.] 2. Bostrom, N. (2014). *Superintelligence: Paths, Dangers, Strategies*. Oxford University Press. ISBN: 978-0199678112. [Classification of pathways and forms of superintelligence.] 3. Burns, C., Izmailov, P., Kirchner, J. H., Baker, B., Gao, L., Aschenbrenner, L., Chen, Y., Ecoffet, A., Joglekar, M., Leike, J., Sutskever, I., & Wu, J. (2023). *Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision*. arXiv preprint. arXiv:2312.09390. [Experimental analogy for supervising systems beyond human ability.] 4. Christiano, P., Shlegeris, B., & Amodei, D. (2018). *Supervising strong learners by amplifying weak experts*. arXiv preprint. arXiv:1810.08575. [Iterated amplification proposal for tasks humans cannot directly evaluate.] 5. Erdil, E., & Besiroglu, T. (2023). *Explosive growth from AI automation: A review of the arguments*. arXiv preprint. arXiv:2309.11690. [Reviews drivers of and objections to explosive growth from AI automation.] 6. Good, I. J. (1966). *Speculations Concerning the First Ultraintelligent Machine*. Advances in Computers, 6, 31–88. https://doi.org/10.1016/S0065-2458(08)60418-0. [Origin of the self-improvement argument.] 7. Ho, A., Besiroglu, T., Erdil, E., Owen, D., Rahman, R., Guo, Z. C., Atkinson, D., Thompson, N., & Sevilla, J. (2024). *Algorithmic progress in language models*. arXiv preprint. arXiv:2403.05812. [Compute-halving time of about 8 months (95% CI roughly 5 to 14 months) on two benchmarks, 2012 to 2023.] 8. Hoffmann, J., Borgeaud, S., Mensch, A., Buchatskaya, E., Cai, T., Rutherford, E., et al. (2022). *Training Compute-Optimal Large Language Models*. arXiv preprint. arXiv:2203.15556 (22 authors). [Compute-optimal scaling of model size and tokens.] 9. Hutter, M. (2012). *Can Intelligence Explode?* Journal of Consciousness Studies, 19(1–2), 143–166. arXiv:1202.6177. [Analysis of conditions for an intelligence explosion.] 10. Kaplan, J., McCandlish, S., Henighan, T., Brown, T. B., Chess, B., Child, R., Gray, S., Radford, A., Wu, J., & Amodei, D. (2020). *Scaling Laws for Neural Language Models*. arXiv preprint. arXiv:2001.08361. [Power-law dependence of loss on scale.] 11. Sandberg, A., & Bostrom, N. (2008). *Whole Brain Emulation: A Roadmap* (Technical Report #2008-3). Future of Humanity Institute, University of Oxford. [Feasibility roadmap for whole brain emulation.] 12. Silver, D., Schrittwieser, J., Simonyan, K., Antonoglou, I., Huang, A., Guez, A., et al. (2017). *Mastering the game of Go without human knowledge*. Nature, 550(7676), 354–359. https://doi.org/10.1038/nature24270. [Self-play Go system without human game data.] ## 12. Evidential Classification Matrix | Section / Claim | Evidence Category | Supporting Source | Confidence / Status | | :--- | :--- | :--- | :--- | | Language-model loss follows power laws in scale over studied ranges | Well Supported | Kaplan et al. (2020); Hoffmann et al. (2022) | High for loss; low for capability extrapolation | | Scaling will lead to general superhuman capability | Contested | Kaplan et al. (2020); Bostrom (2014) | Active debate | | Compute needed for fixed performance halved about every 8 months (95% CI 5 to 14), 2012 to 2023 | Preliminary (preprint, benchmark-specific) | Ho et al. (2024) | Moderate | | Self-play can exceed human performance in fixed-rule games | Well Supported | Silver et al. (2017) | High | | Self-play extends to open-ended domains | Contested | Silver et al. (2017) | Low | | Automated AI research could accelerate AI progress | Hypothetical | Good (1966); Erdil & Besiroglu (2023) | Low to moderate | | Whole brain emulation is feasible | Hypothetical | Sandberg & Bostrom (2008) | Low | | Biological enhancement or brain-computer interfaces reach superintelligence | Hypothetical | Bostrom (2014) | Low | | Explosive growth from AI automation | Contested | Erdil & Besiroglu (2023) | Plausible if labour is largely replaced; high confidence not justified, per authors | ## 13. Revision History - **v0.1.0** (2026-10-15): Initial draft. Citations checked against arXiv/Crossref/publisher pages where marked; not independently fact-checked.