{"ontology":"https://layer.re/machine/schema.json","epistemic_model":"evidence-backed-temporal-layer-graph","canonical":true,"type":"civilization_snapshot","snapshot":{"id":"civ-2023q4-reconstructed","as_of":"2023-12-31 23:59:59.000","summary":"By 31 December 2023, generative AI had begun to function as a persistent, configurable systems layer rather than only a deployable model ecosystem. Developer platforms introduced managed persistent assistant threads, retrieval, executable tools and external actions; users and enterprises could configure reusable specialized assistants; enterprise copilots reached general availability; and multimodal, long-context model interfaces expanded across multiple providers. Open model infrastructure broadened into permissively licensed mixture-of-experts models and open safety tooling. Compute infrastructure diversified through AMD and Google accelerators while large training compute and cloud/IaaS infrastructure became explicit objects of AI-specific governance. Public governance hardened through the U.S. AI Executive Order, Bletchley Declaration, tightened semiconductor controls and a provisional EU AI Act agreement. Robotics moved toward shared learning across multiple embodiments. Persistent autonomous operation, portable delegated permissions, generalized agent identity, cheap unconstrained frontier compute, broadly available Gemini Ultra and finalized international frontier-AI regulation were not yet established.","cross_layer_changes":["Managed persistent assistant state/tools became developer infrastructure.","Reusable customized assistants became a product surface.","Enterprise copilots reached GA.","Multimodal/long-context platforms expanded.","Open models gained MoE plus open safety tooling.","Accelerator infrastructure diversified.","Large AI compute became a governance object.","Frontier AI gained formal U.S., multilateral and provisional EU governance.","Robotics scaled across embodiments."],"dependencies":["models + software + managed threads/tools -> agent-like applications","enterprise data + tenant controls + models -> enterprise copilots","open weights + safety tooling + compute -> open deployment stack","chips + networks + software -> diversified AI compute","governance + compute thresholds/IaaS -> infrastructure oversight","cross-embodiment data + models -> robot transfer"],"constraints":["authorization","continuous autonomy","frontier compute cost/supply","regulatory implementation lag","safety evaluation quality","robot generalization"],"emerging_capabilities":["persistent assistant threads","retrieval/tool-enabled APIs","custom GPTs and Actions","enterprise Copilot GA","128K/200K contexts","Gemini multimodality","MI300X and TPU v5p","open MoE/safety tooling","cross-embodiment robotics"],"possible_bottlenecks":["permission boundaries","tool correctness","compute supply","governance interoperability","safety evaluation","embodiment transfer"],"uncertainties":["Persistent threads do not establish autonomous agents.","Actions remain application-specific permissions.","Gemini Ultra was not generally available.","EU AI Act deal was provisional.","Bletchley was nonbinding.","Accelerator availability did not imply abundance."],"mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.136Z","updated_at":"2026-10-08T17:00:00.136Z"},"claims":[{"id":"claim-q4-accelerator-diversification","statement":"MI300X availability and TPU v5p/AI Hypercomputer provide Q4 evidence that frontier AI compute was diversifying beyond the Q3 H100 path.","epistemic_status":"supported_inference","confidence":0.97,"civilization_snapshot_id":"civ-2023q4-reconstructed","layer_state_snapshot_id":"state-compute-2023q4","competing_interpretation":"The observed developments may be interpreted as maturation within prior trajectories rather than a discrete state transition.","uncertainty":"This reconstruction preserves Q4 cutoff limits and does not infer later adoption, final regulation or autonomous operation.","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.140Z","updated_at":"2026-10-08T17:00:00.140Z"},{"id":"claim-q4-ai-compute-governance","statement":"The October 30 U.S. AI Executive Order provides Q4 evidence that sufficiently large AI training compute and IaaS-based training became explicit AI-specific governance/reporting objects.","epistemic_status":"supported_inference","confidence":0.99,"civilization_snapshot_id":"civ-2023q4-reconstructed","layer_state_snapshot_id":"state-compute-2023q4","competing_interpretation":"The observed developments may be interpreted as maturation within prior trajectories rather than a discrete state transition.","uncertainty":"This reconstruction preserves Q4 cutoff limits and does not infer later adoption, final regulation or autonomous operation.","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.141Z","updated_at":"2026-10-08T17:00:00.141Z"},{"id":"claim-q4-cross-embodiment-robotics","statement":"Open X-Embodiment and RT-X provide Q4 evidence for scaling robot-learning data and policy transfer across multiple physical embodiments.","epistemic_status":"supported_inference","confidence":0.97,"civilization_snapshot_id":"civ-2023q4-reconstructed","layer_state_snapshot_id":"state-robotics-2023q4","competing_interpretation":"The observed developments may be interpreted as maturation within prior trajectories rather than a discrete state transition.","uncertainty":"This reconstruction preserves Q4 cutoff limits and does not infer later adoption, final regulation or autonomous operation.","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.142Z","updated_at":"2026-10-08T17:00:00.142Z"},{"id":"claim-q4-enterprise-copilot","statement":"Microsoft 365 Copilot GA provides Q4 evidence that context-aware generative assistance became a purchasable enterprise productivity layer integrated with organizational data and tenant controls.","epistemic_status":"supported_inference","confidence":0.98,"civilization_snapshot_id":"civ-2023q4-reconstructed","layer_state_snapshot_id":"state-software-2023q4","competing_interpretation":"The observed developments may be interpreted as maturation within prior trajectories rather than a discrete state transition.","uncertainty":"This reconstruction preserves Q4 cutoff limits and does not infer later adoption, final regulation or autonomous operation.","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.137Z","updated_at":"2026-10-08T17:00:00.137Z"},{"id":"claim-q4-frontier-governance","statement":"The U.S. AI Executive Order, Bletchley Declaration and provisional EU AI Act agreement materially formalized frontier-model governance beyond Q3 voluntary/private commitments.","epistemic_status":"supported_inference","confidence":0.99,"civilization_snapshot_id":"civ-2023q4-reconstructed","layer_state_snapshot_id":"state-governance-2023q4","competing_interpretation":"The observed developments may be interpreted as maturation within prior trajectories rather than a discrete state transition.","uncertainty":"This reconstruction preserves Q4 cutoff limits and does not infer later adoption, final regulation or autonomous operation.","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.141Z","updated_at":"2026-10-08T17:00:00.141Z"},{"id":"claim-q4-multimodal-platform","statement":"Q4 materially broadened long-context and multimodal model platforms through GPT-4 Turbo APIs, Claude 2.1 and Gemini while key variants remained preview, beta or unavailable.","epistemic_status":"supported_inference","confidence":0.98,"civilization_snapshot_id":"civ-2023q4-reconstructed","layer_state_snapshot_id":"state-models-2023q4","competing_interpretation":"The observed developments may be interpreted as maturation within prior trajectories rather than a discrete state transition.","uncertainty":"This reconstruction preserves Q4 cutoff limits and does not infer later adoption, final regulation or autonomous operation.","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.138Z","updated_at":"2026-10-08T17:00:00.138Z"},{"id":"claim-q4-open-model-safety-stack","statement":"Mixtral and Purple Llama provide Q4 evidence that open model infrastructure expanded beyond weights into permissive MoE architectures and safety/evaluation tooling.","epistemic_status":"supported_inference","confidence":0.96,"civilization_snapshot_id":"civ-2023q4-reconstructed","layer_state_snapshot_id":"state-open-source-2023q4","competing_interpretation":"The observed developments may be interpreted as maturation within prior trajectories rather than a discrete state transition.","uncertainty":"This reconstruction preserves Q4 cutoff limits and does not infer later adoption, final regulation or autonomous operation.","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.140Z","updated_at":"2026-10-08T17:00:00.140Z"},{"id":"claim-q4-persistent-systems-layer","statement":"Q4 supports a transition in which generative AI became a more persistent and configurable systems layer through managed assistant threads, custom assistants, enterprise copilots and multimodal platforms.","epistemic_status":"supported_inference","confidence":0.99,"civilization_snapshot_id":"civ-2023q4-reconstructed","layer_state_snapshot_id":"state-agents-2023q4","competing_interpretation":"The observed developments may be interpreted as maturation within prior trajectories rather than a discrete state transition.","uncertainty":"This reconstruction preserves Q4 cutoff limits and does not infer later adoption, final regulation or autonomous operation.","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.137Z","updated_at":"2026-10-08T17:00:00.137Z"}],"layer_states":[{"id":"state-ai-2023q4","layer_id":"layer-ai","as_of":"2023-12-31 23:59:59.000","summary":"Generative AI became a more persistent and configurable systems layer across enterprise, developer and multimodal surfaces while autonomy and generalized permissions remained bounded.","dependencies":["layer-models","layer-software","layer-agents","layer-governance","layer-compute"],"constraints":["reliability","tool safety","authorization","availability gaps","compute concentration"],"emerging_capabilities":["custom GPTs","persistent assistant threads","multimodal developer APIs","enterprise copilots","Gemini Pro APIs"],"underdeveloped_components":["continuous autonomous operation","general delegated permissions"],"uncertainties":["Gemini Ultra was not generally available; persistent threads are not autonomous operation"],"confidence":0.98,"previous_snapshot_id":"state-ai-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.128Z","updated_at":"2026-10-08T17:00:00.128Z","layer_slug":"ai","layer_name":"AI"},{"id":"state-agents-2023q4","layer_id":"layer-agents","as_of":"2023-12-31 23:59:59.000","summary":"Agent-like infrastructure gained provider-managed persistent thread state, retrieval/execution tools and reusable customized assistant configurations while remaining user/app initiated.","dependencies":["layer-models","layer-software","layer-human","layer-permissions"],"constraints":["authorization","durable identity","background operation","tool safety"],"emerging_capabilities":["Assistants API threads","Retrieval","Code Interpreter","custom GPTs","Actions"],"underdeveloped_components":["continuous autonomous operation","portable delegated permissions"],"uncertainties":["Persistent thread state is not continuously running autonomy"],"confidence":0.99,"previous_snapshot_id":"state-agents-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.130Z","updated_at":"2026-10-08T17:00:00.130Z","layer_slug":"agents","layer_name":"Agents"},{"id":"state-chips-2023q4","layer_id":"layer-chips","as_of":"2023-12-31 23:59:59.000","summary":"Advanced AI silicon diversified through MI300X and TPU v5p while U.S. export controls further tightened access to advanced chips and manufacturing equipment.","dependencies":["layer-manufacturing","layer-compute","layer-governance","layer-geopolitics"],"constraints":["supply concentration","software ecosystems","export controls"],"emerging_capabilities":["MI300X","TPU v5p","tighter advanced-computing controls"],"underdeveloped_components":["broad accelerator abundance","stable cross-border access"],"uncertainties":["Accelerator availability and export restrictions affect different portions of the chip stack"],"confidence":0.98,"previous_snapshot_id":"state-chips-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.132Z","updated_at":"2026-10-08T17:00:00.132Z","layer_slug":"chips","layer_name":"Chips"},{"id":"state-compute-2023q4","layer_id":"layer-compute","as_of":"2023-12-31 23:59:59.000","summary":"Compute diversified beyond the H100 cloud path through AMD MI300X and Google TPU v5p/AI Hypercomputer while large AI compute became an explicit governance object.","dependencies":["layer-chips","layer-networks","layer-software","layer-governance"],"constraints":["supply","cost","regional access","software ecosystem maturity"],"emerging_capabilities":["MI300X","TPU v5p","AI Hypercomputer","AI-specific large-cluster governance"],"underdeveloped_components":["cheap unconstrained frontier compute","accelerator interchangeability"],"uncertainties":["TPU access was request-based and MI300X availability did not imply abundance"],"confidence":0.98,"previous_snapshot_id":"state-compute-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.132Z","updated_at":"2026-10-08T17:00:00.132Z","layer_slug":"compute","layer_name":"Compute"},{"id":"state-geopolitics-2023q4","layer_id":"layer-geopolitics","as_of":"2023-12-31 23:59:59.000","summary":"Technology geopolitics combined tighter U.S. chip/compute controls with explicit international frontier-AI coordination and EU agreement on model-specific AI regulation.","dependencies":["layer-governance","layer-chips","layer-compute","layer-manufacturing","layer-ai"],"constraints":["technology fragmentation","export controls","nonbinding coordination"],"emerging_capabilities":["Bletchley coordination","expanded U.S. chip controls","EU GPAI agreement"],"underdeveloped_components":["binding global frontier-AI regime","stable cross-border advanced-compute access"],"uncertainties":["International AI coordination remained plural and nonbinding"],"confidence":0.98,"previous_snapshot_id":"state-geopolitics-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.135Z","updated_at":"2026-10-08T17:00:00.135Z","layer_slug":"geopolitics","layer_name":"Geopolitics"},{"id":"state-governance-2023q4","layer_id":"layer-governance","as_of":"2023-12-31 23:59:59.000","summary":"AI governance hardened through a U.S. executive order, Bletchley Declaration, provisional EU AI Act agreement, open safety infrastructure and tighter semiconductor controls.","dependencies":["layer-ai","layer-models","layer-compute","layer-software","layer-chips","layer-geopolitics"],"constraints":["implementation lag","jurisdictional divergence","nonbinding declarations","measurement thresholds"],"emerging_capabilities":["EO 14110","Bletchley Declaration","EU AI Act provisional agreement","AI-compute reporting design","tighter BIS controls"],"underdeveloped_components":["final EU AI Act","binding international frontier-AI governance"],"uncertainties":["Bletchley was nonbinding; EU deal provisional; EO required later implementation"],"confidence":0.99,"previous_snapshot_id":"state-governance-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.135Z","updated_at":"2026-10-08T17:00:00.135Z","layer_slug":"governance","layer_name":"Governance"},{"id":"state-human-2023q4","layer_id":"layer-human","as_of":"2023-12-31 23:59:59.000","summary":"Human interaction with AI became more configurable and organizationally embedded through Microsoft 365 Copilot GA, custom GPTs and expanding multimodal assistants.","dependencies":["layer-models","layer-software","layer-governance"],"constraints":["trust","privacy","organizational policy","tool risk"],"emerging_capabilities":["enterprise Copilot GA","custom GPT creation","vision/audio APIs","Gemini-powered products"],"underdeveloped_components":["safe generalized delegation","independent productivity evidence"],"uncertainties":["Availability and configurability do not establish universal adoption"],"confidence":0.97,"previous_snapshot_id":"state-human-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.134Z","updated_at":"2026-10-08T17:00:00.134Z","layer_slug":"human","layer_name":"Human"},{"id":"state-manufacturing-2023q4","layer_id":"layer-manufacturing","as_of":"2023-12-31 23:59:59.000","summary":"Semiconductor manufacturing remained policy-constrained in Q4 as BIS expanded controls on additional semiconductor manufacturing equipment and support for advanced production.","dependencies":["layer-chips","layer-governance","layer-geopolitics"],"constraints":["equipment licensing","fab lead times","capital intensity"],"emerging_capabilities":["expanded U.S. semiconductor-manufacturing-equipment controls"],"underdeveloped_components":["rapid new advanced-node capacity","stable equipment access"],"uncertainties":["Rule issuance does not itself measure physical production losses, realized capacity changes, or the magnitude of downstream manufacturing effects."],"confidence":0.97,"previous_snapshot_id":"state-manufacturing-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.133Z","updated_at":"2026-10-08T17:00:00.133Z","layer_slug":"manufacturing","layer_name":"Manufacturing"},{"id":"state-models-2023q4","layer_id":"layer-models","as_of":"2023-12-31 23:59:59.000","summary":"Models combined longer context, multimodal interfaces, tool use, multiple provider families and open MoE architectures while model-specific governance became more explicit.","dependencies":["layer-research","layer-compute","layer-software","layer-governance","layer-open-source"],"constraints":["context reliability","tool reliability","compute cost","access gating"],"emerging_capabilities":["GPT-4 Turbo 128K preview","Claude 2.1 200K","Gemini Pro/Nano","Mixtral MoE"],"underdeveloped_components":["reliable long-context reasoning","uniform multimodal access"],"uncertainties":["GPT-4 Turbo was preview and Gemini Ultra was not generally available"],"confidence":0.98,"previous_snapshot_id":"state-models-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.129Z","updated_at":"2026-10-08T17:00:00.129Z","layer_slug":"models","layer_name":"Models"},{"id":"state-open-source-2023q4","layer_id":"layer-open-source","as_of":"2023-12-31 23:59:59.000","summary":"Open model infrastructure expanded beyond weights into permissive MoE architecture and reusable open trust/safety tooling.","dependencies":["layer-models","layer-research","layer-software","layer-compute"],"constraints":["compute","safety validation","frontier capability gap"],"emerging_capabilities":["Mixtral 8x7B","Purple Llama","CyberSec Eval","Llama Guard"],"underdeveloped_components":["mature open deployment governance","cheap frontier reproduction"],"uncertainties":["Open safety tools do not guarantee safe deployment"],"confidence":0.97,"previous_snapshot_id":"state-open-source-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.131Z","updated_at":"2026-10-08T17:00:00.131Z","layer_slug":"open-source","layer_name":"Open source"},{"id":"state-research-2023q4","layer_id":"layer-research","as_of":"2023-12-31 23:59:59.000","summary":"Research expanded cross-embodiment robotics, open safety evaluation tooling and frontier-model governance/evaluation requirements.","dependencies":["layer-models","layer-robotics","layer-open-source","layer-governance","layer-compute"],"constraints":["evaluation standardization","compute access","robot data heterogeneity"],"emerging_capabilities":["Open X-Embodiment","RT-X","CyberSec Eval","Llama Guard"],"underdeveloped_components":["general cross-embodiment autonomy","validated universal frontier-safety tests"],"uncertainties":["Research releases and policy-directed evaluation do not prove broad deployment effectiveness"],"confidence":0.96,"previous_snapshot_id":"state-research-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.130Z","updated_at":"2026-10-08T17:00:00.130Z","layer_slug":"research","layer_name":"Research"},{"id":"state-robotics-2023q4","layer_id":"layer-robotics","as_of":"2023-12-31 23:59:59.000","summary":"Robotics research shifted toward shared datasets and policy learning across many robot embodiments through Open X-Embodiment and RT-X.","dependencies":["layer-models","layer-research","robot data","layer-open-source"],"constraints":["embodiment heterogeneity","physical safety","dataset bias"],"emerging_capabilities":["22-embodiment shared dataset","RT-X cross-robot learning"],"underdeveloped_components":["general-purpose robot autonomy","arbitrary-embodiment transfer"],"uncertainties":["Positive transfer across selected robots is not universal embodiment generalization"],"confidence":0.97,"previous_snapshot_id":"state-robotics-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.133Z","updated_at":"2026-10-08T17:00:00.133Z","layer_slug":"robotics","layer_name":"Robotics"},{"id":"state-software-2023q4","layer_id":"layer-software","as_of":"2023-12-31 23:59:59.000","summary":"Software became a configurable AI substrate through enterprise copilots, persistent assistant APIs, custom GPTs, multimodal APIs and Gemini Pro access.","dependencies":["layer-models","layer-agents","layer-human","layer-governance","layer-identity","layer-permissions"],"constraints":["vendor fragmentation","enterprise data governance","authorization","platform lock-in"],"emerging_capabilities":["Microsoft 365 Copilot GA","Assistants API","GPTs","multimodal APIs","Gemini Pro API"],"underdeveloped_components":["cross-provider agent standards","portable identity/permissions"],"uncertainties":["Organizational IAM and action permissions remained product-specific"],"confidence":0.99,"previous_snapshot_id":"state-software-2023q3","mapping_mode":"retrospective","reconstructed_at":"2026-10-08T15:50:00.000Z","created_at":"2026-10-08T17:00:00.131Z","updated_at":"2026-10-08T17:00:00.131Z","layer_slug":"software","layer_name":"Software"}]}