Shashwat Singh
Center for Data Science, New York University
Critic Current evidence and current methods do not support the claims being made
Shashwat Singh is a machine learning and interpretability researcher working with Shauli Ravfogel and Tal Linzen on causal probing, representation erasure, and the empirical validity of self-knowledge claims in large language models.
As first author of Can LLMs Introspect? A Reality Check (arXiv:2605.26242, May 2026), Singh systematically tested whether language models that pass intervention-detection or hidden-state-prediction benchmarks are genuinely reading their own internal states. By constructing relabeled control tasks and training external baseline classifiers restricted to prompt features, Singh, Linzen, and Ravfogel proved that existing introspection benchmarks are heavily confounded by input-level anomalies and surface heuristics, establishing a rigorous evidentiary bar for mechanistic introspection studies.
Singh is classified as a Critic on this site for designing the control experiments that constrain claims of metacognitive self-access in frontier language models.
Known for. Can LLMs Introspect? A Reality Check, concept erasure and causal representation auditing in language models
Coverage on this site
- LLM Introspection After the November 2026 Quanta Report, Concept Injection Against the Reality Check November 2026
Related researchers
- Jack LindseyAnthropic, Transformer CircuitsIntrospection circuits in language models, mind viruses in multi-agent systems, Global Workspace representations
- Tal LinzenDepartment of Linguistics and Center for Data Science, New York UniversityCan LLMs Introspect? A Reality Check, mechanistic evaluation of syntactic and metacognitive claims in language models
- Shauli RavfogelCenter for Data Science, New York University and Bar-Ilan UniversityCan LLMs Introspect? A Reality Check, linear concept erasure, causal probing of neural representations
- Scott AaronsonUniversity of Texas at Austin, Department of Computer ScienceQuantum complexity theory, mathematical counterexamples to Integrated Information Theory (Phi in expander graphs and Reed-Solomon grids), speaker at UT Austin Consciousness and Intelligence 2026
- Michel BitbolCNRSPhenomenology and quantum physics, the critique of objectivist reduction, first-person knowledge as irreducible, collaboration with Claire Petitmengin
- Ned BlockDepartment of Philosophy, New York UniversityDistinction between Phenomenal and Access Consciousness (1995), Blockhead argument against behavioral tests, The Border Between Seeing and Thinking (2023)
Shashwat Singh is listed with the other critics and skeptics, who hold that current evidence and current methods do not support the claims being made. The positions themselves are set out in the index of consciousness theories. Every researcher covered on this site is indexed in the directory of consciousness researchers.