mlsanityData to 14 Oct 2026

Practical Principles for AI Cost and Compute Accounting

arXiv 2502.15873 ↗First posted 21 Feb 2025cs.AI cs.CYcited by 3

Not on the map. Why?

Abstract (arXiv)

Policymakers increasingly use development cost and compute as proxies for AI capabilities and risks. Recent laws have introduced regulatory requirements for models or developers that are contingent on specific thresholds. However, technical ambiguities in how to perform this accounting create loopholes that can undermine regulatory effectiveness. We propose seven principles for designing AI cost and compute accounting standards that (1) reduce opportunities for strategic gaming, (2) avoid disincentivizing responsible risk mitigation, and (3) enable consistent implementation across companies and jurisdictions.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 19 of its 38 references that are in the map. First posted on arXiv 21 Feb 2025.

Builds on

19 of 38 references.

References · 19

  1. 2020Scaling Laws for Neural Language Modelscited by 6,586
  2. 2024GPT-4o System Cardcited by 4,588
  3. 2018Model Cards for Model Reportingcited by 1,080
  4. 2021Process for Adapting Language Models to Society (PALMS) with Values-Targeted Datasetscited by 181
  5. 2023Model evaluation for extreme riskscited by 177
  6. 2019Adaptively Sparse Transformerscited by 155
  7. 2024The rising costs of training frontier AI modelscited by 95
  8. 2023The MiniPile Challenge for Data-Efficient Language Modelscited by 75
  9. 2023The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AIcited by 73
  10. 2024Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Trainingcited by 69
  11. 2024Safetywashing: Do AI Safety Benchmarks Actually Measure Safety Progress?cited by 61
  12. 2024Survey on Knowledge Distillation for Large Language Models: Methods, Evaluation, and Applicationcited by 56
7 more
  1. 2024Open Problems in Technical AI Governancecited by 49
  2. 2024Computing Power and the Governance of Artificial Intelligencecited by 40
  3. 2024Automated Data Curation for Robust Language Model Fine-Tuningcited by 27
  4. 2024On the Limitations of Compute Thresholds as a Governance Strategycited by 18
  5. 2024Training Compute Thresholds: Features and Functions in AI Regulationcited by 14
  6. 2023Increased Compute Efficiency and the Diffusion of AI Capabilitiescited by 8
  7. 2024How Do AI Companies "Fine-Tune" Policy? Examining Regulatory Capture in AI Governancecited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 1.

3 papers.

  1. 2025The 2025 Foundation Model Transparency Indexcited by 11
  2. 2025NeurIPS should lead scientific consensus on AI policycited by 3
  3. 2026Beyond Training: A Feasibility Taxonomy for Inference-Time AI Governancecited by 0