{"@context":"https://schema.org","@type":"DataCatalog","name":"Maha Research Atlas — aggregated claim index","url":"https://research.mahastrategies.com/atlas/claims.json","isPartOf":"https://research.mahastrategies.com/atlas","role":"discovery-index","note":"A cross-atlas discovery index of claim records. Identifying fields and canonical URLs only — this is not the record of authority. Follow canonicalUrl for the full record, or the owning atlas's own JSON endpoint listed in atlases[].endpoints of /atlas/manifest.json.","boundary":"The atlases are source-linked research maps. They are not consensus statements, forecasts, investment advice, or operational guidance. Record status labels and limitations are part of the record and must not be dropped when reused.","license":"https://creativecommons.org/licenses/by/4.0/","count":46,"byAtlas":{"de-sitter-swampland":12,"quantum-computing":12,"synthetic-intelligence":22},"fieldsOmitted":["explanation","limitations","conceptIds","sourceIds"],"claims":[{"atlasId":"de-sitter-swampland","atlasUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland","id":"ds-001","status":"established","claim":"The observed late-time acceleration of the universe is standardly modeled with a small positive cosmological constant, and current data place the dark-energy equation of state near w = −1.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims/ds-001","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims.json"},{"atlasId":"de-sitter-swampland","atlasUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland","id":"ds-002","status":"active","claim":"Whether string/M-theory admits stable or metastable de Sitter vacua remains an open question, and there is no community consensus.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims/ds-002","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims.json"},{"atlasId":"de-sitter-swampland","atlasUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland","id":"ds-003","status":"established","claim":"KKLT is a specific, well-defined proposal for obtaining a de Sitter vacuum: flux and non-perturbative stabilization into a supersymmetric anti-de Sitter minimum, followed by an anti-D3-brane uplift.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims/ds-003","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims.json"},{"atlasId":"de-sitter-swampland","atlasUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland","id":"ds-004","status":"active","claim":"Whether the KKLT anti-brane uplift is under parametric ten-dimensional supergravity control is actively debated.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims/ds-004","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims.json"},{"atlasId":"de-sitter-swampland","atlasUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland","id":"ds-005","status":"established","claim":"The Large Volume Scenario stabilizes Kähler moduli at exponentially large volume, which nominally suppresses α′ and loop corrections.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims/ds-005","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims.json"},{"atlasId":"de-sitter-swampland","atlasUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland","id":"ds-006","status":"active","claim":"It has been argued that keeping LVS de Sitter constructions under control requires a negative D3-tadpole exceeding known Calabi-Yau bounds.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims/ds-006","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims.json"},{"atlasId":"de-sitter-swampland","atlasUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland","id":"ds-007","status":"conjecture","claim":"The de Sitter swampland conjecture proposes that scalar potentials in consistent quantum-gravity effective theories obey a gradient bound |∇V| ≥ c·V, with c of order one.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims/ds-007","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims.json"},{"atlasId":"de-sitter-swampland","atlasUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland","id":"ds-008","status":"active","claim":"An order-one value of c has been argued to be in tension with slow-roll inflation and with the observed value of the cosmological constant.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims/ds-008","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims.json"},{"atlasId":"de-sitter-swampland","atlasUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland","id":"ds-009","status":"conjecture","claim":"The Trans-Planckian Censorship Conjecture proposes that sub-Planckian fluctuations never cross the Hubble horizon and freeze, which bounds the lifetime of a de Sitter phase.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims/ds-009","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims.json"},{"atlasId":"de-sitter-swampland","atlasUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland","id":"ds-010","status":"active","claim":"If metastable de Sitter vacua are excluded, dark energy would have to be dynamical — a rolling scalar field rather than a constant.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims/ds-010","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims.json"},{"atlasId":"de-sitter-swampland","atlasUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland","id":"ds-011","status":"speculative","claim":"The Dark Dimension scenario ties the dark-energy scale to a Kaluza-Klein scale via the AdS distance conjecture, predicting a single extra dimension of roughly micron size.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims/ds-011","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims.json"},{"atlasId":"de-sitter-swampland","atlasUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland","id":"ds-012","status":"established","claim":"Torsion-balance experiments report no deviation from the gravitational inverse-square law down to separations of about 52 micrometres.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims/ds-012","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/de-sitter-swampland/claims.json"},{"atlasId":"quantum-computing","atlasUrl":"https://research.mahastrategies.com/atlas/quantum-computing","id":"qc-001","status":"established","claim":"A qubit is a quantum information carrier whose state can be prepared, transformed, and measured; it is not simply a classical bit with two labels.","reviewDate":"2026-07-28","canonicalUrl":"https://research.mahastrategies.com/atlas/quantum-computing/claims/qc-001","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/quantum-computing/claims.json"},{"atlasId":"quantum-computing","atlasUrl":"https://research.mahastrategies.com/atlas/quantum-computing","id":"qc-002","status":"established","claim":"Superconducting circuits, trapped ions, neutral atoms, photonics, and semiconductor spins are distinct hardware approaches with different engineering trade-offs.","reviewDate":"2026-07-28","canonicalUrl":"https://research.mahastrategies.com/atlas/quantum-computing/claims/qc-002","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/quantum-computing/claims.json"},{"atlasId":"quantum-computing","atlasUrl":"https://research.mahastrategies.com/atlas/quantum-computing","id":"qc-003","status":"established","claim":"Noise and imperfect operations limit the depth and reliability of computations performed directly on physical qubits.","reviewDate":"2026-07-28","canonicalUrl":"https://research.mahastrategies.com/atlas/quantum-computing/claims/qc-003","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/quantum-computing/claims.json"},{"atlasId":"quantum-computing","atlasUrl":"https://research.mahastrategies.com/atlas/quantum-computing","id":"qc-004","status":"established","claim":"Quantum error correction protects logical quantum information by distributing it across multiple physical qubits and repeatedly extracting error information.","reviewDate":"2026-07-28","canonicalUrl":"https://research.mahastrategies.com/atlas/quantum-computing/claims/qc-004","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/quantum-computing/claims.json"},{"atlasId":"quantum-computing","atlasUrl":"https://research.mahastrategies.com/atlas/quantum-computing","id":"qc-005","status":"established","claim":"Below-threshold surface-code memory performance has been experimentally reported, but it is not yet a demonstration of general-purpose fault-tolerant quantum computing.","reviewDate":"2026-07-28","canonicalUrl":"https://research.mahastrategies.com/atlas/quantum-computing/claims/qc-005","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/quantum-computing/claims.json"},{"atlasId":"quantum-computing","atlasUrl":"https://research.mahastrategies.com/atlas/quantum-computing","id":"qc-006","status":"established","claim":"Shor’s algorithm gives polynomial-time quantum algorithms for integer factorization and discrete logarithms on a suitable quantum computer.","reviewDate":"2026-07-28","canonicalUrl":"https://research.mahastrategies.com/atlas/quantum-computing/claims/qc-006","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/quantum-computing/claims.json"},{"atlasId":"quantum-computing","atlasUrl":"https://research.mahastrategies.com/atlas/quantum-computing","id":"qc-007","status":"established","claim":"Grover’s search algorithm provides a quadratic query-speedup for unstructured search in its oracle model.","reviewDate":"2026-07-28","canonicalUrl":"https://research.mahastrategies.com/atlas/quantum-computing/claims/qc-007","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/quantum-computing/claims.json"},{"atlasId":"quantum-computing","atlasUrl":"https://research.mahastrategies.com/atlas/quantum-computing","id":"qc-008","status":"established","claim":"Simulating quantum physical systems is a foundational motivation for quantum computers.","reviewDate":"2026-07-28","canonicalUrl":"https://research.mahastrategies.com/atlas/quantum-computing/claims/qc-008","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/quantum-computing/claims.json"},{"atlasId":"quantum-computing","atlasUrl":"https://research.mahastrategies.com/atlas/quantum-computing","id":"qc-009","status":"established","claim":"Random-circuit sampling experiments demonstrate control over specific computational tasks, not a general-purpose application advantage.","reviewDate":"2026-07-28","canonicalUrl":"https://research.mahastrategies.com/atlas/quantum-computing/claims/qc-009","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/quantum-computing/claims.json"},{"atlasId":"quantum-computing","atlasUrl":"https://research.mahastrategies.com/atlas/quantum-computing","id":"qc-010","status":"active-research","claim":"Whether a NISQ device delivers a durable, practically relevant advantage for a specific real-world workload remains an active research question.","reviewDate":"2026-07-28","canonicalUrl":"https://research.mahastrategies.com/atlas/quantum-computing/claims/qc-010","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/quantum-computing/claims.json"},{"atlasId":"quantum-computing","atlasUrl":"https://research.mahastrategies.com/atlas/quantum-computing","id":"qc-011","status":"active-research","claim":"The physical resources, decoding, control, and integration needed for useful large-scale fault-tolerant computation remain active engineering and research problems.","reviewDate":"2026-07-28","canonicalUrl":"https://research.mahastrategies.com/atlas/quantum-computing/claims/qc-011","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/quantum-computing/claims.json"},{"atlasId":"quantum-computing","atlasUrl":"https://research.mahastrategies.com/atlas/quantum-computing","id":"qc-012","status":"established","claim":"Claims about quantum applications require task-specific comparisons to credible classical baselines.","reviewDate":"2026-07-28","canonicalUrl":"https://research.mahastrategies.com/atlas/quantum-computing/claims/qc-012","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/quantum-computing/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-001","status":"established","claim":"The transformer architecture replaces recurrence and convolution with attention as the primary sequence-modelling mechanism.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-001","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-002","status":"established","claim":"Sufficiently large autoregressive language models perform tasks from instructions and examples supplied in context, without gradient updates.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-002","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-003","status":"established","claim":"Language-model loss follows empirical power-law relationships with training compute, parameter count, and dataset size over the studied ranges.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-003","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-004","status":"active","claim":"The compute-optimal allocation between parameters and training tokens has been revised in the literature, and deployed practice diverges from compute-optimal training.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-004","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-005","status":"established","claim":"Fine-tuning with human feedback substantially changes how closely a model follows instructions relative to its pre-trained base.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-005","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-006","status":"established","claim":"Prompting a model to produce intermediate reasoning steps changes its measured performance on multi-step problems.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-006","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-007","status":"active","claim":"Allocating additional computation at inference time can, in studied settings, improve results more than spending the equivalent compute on additional parameters.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-007","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-008","status":"established","claim":"Retrieval-augmented generation conditions outputs on documents retrieved at inference time rather than on parametric memory alone.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-008","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-009","status":"established","claim":"Language models can be trained or prompted to invoke external tools and incorporate returned results.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-009","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-010","status":"established","claim":"The design of the agent-computer interface materially changes measured coding-agent performance holding the underlying model fixed.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-010","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-011","status":"established","claim":"SWE-bench measures whether a system produces a patch that passes a repository’s existing test suite for a real issue.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-011","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-012","status":"established","claim":"MMLU and GPQA are multiple-choice benchmarks over defined subject distributions, with human baselines reported by their authors.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-012","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-013","status":"established","claim":"Benchmarks are revised by their owners in response to saturation and robustness problems, producing successor versions that are not directly comparable to the originals.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-013","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-014","status":"active","claim":"Contamination of evaluation data into training corpora is a measured problem that must be assessed per benchmark rather than assumed absent.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-014","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-015","status":"active","claim":"The use of general-purpose benchmarks as measures of general capability is contested in the literature on construct-validity grounds.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-015","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-016","status":"established","claim":"Multi-metric evaluation across scenarios has been proposed and implemented as an alternative to single headline scores.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-016","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-017","status":"active","claim":"The behaviour of a served model accessed through an API can change over time on fixed prompts.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-017","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-018","status":"established","claim":"METR defines a 50%-task-completion time horizon: the time humans typically take on tasks a model completes with 50% success. Its published measurement reports a horizon of around 50 minutes for one frontier model of that period, and a doubling of the horizon roughly every seven months since 2019, with a possible acceleration in 2024.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-018","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-019","status":"active","claim":"Increases in measured long-horizon performance are attributed in part to improved reliability and error recovery rather than to reasoning improvements alone.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-019","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-020","status":"established","claim":"Provider technical reports and model cards are self-reports by the organization that built the system, and may withhold architecture, data, and training details.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-020","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-021","status":"active","claim":"Aggregated expert forecasts of AI capability timelines have shifted substantially between survey rounds.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-021","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"},{"atlasId":"synthetic-intelligence","atlasUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence","id":"si-023","status":"established","claim":"A benchmark score is a measurement of a system on a task distribution under a scaffolding configuration at a date. It is not a measurement of general capability.","reviewDate":"2026-07-27","canonicalUrl":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims/si-023","atlasClaimsEndpoint":"https://research.mahastrategies.com/atlas/synthetic-intelligence/claims.json"}]}