AI Evaluations
EvaluationactiveSystematic testing and measurement of AI system capabilities, alignment, and safety properties.
Organizations
4
Grants
122
Total Funding
$111M
Risks Addressed
2
Cluster: Evaluation
Tags
evaluationssafety-researchtesting
Organizations4
| Organization | Role |
|---|---|
| Anthropic | active |
| Alignment Research Center | active |
| Google DeepMind | active |
| OpenAI | active |
GrantsTop 50 of 122
| Name | Recipient | Amount | Funder | Date |
|---|---|---|---|---|
| OpenMined Foundation — Secure Enclaves for LLM Evaluation | OpenMined | $11M | Coefficient Giving | 2025-06 |
| UC Davis — Malaria Gene Drive Feasibility Analysis (Greg Lanzaro) (2021) | University of California, Davis | $10M | Coefficient Giving | 2021-12 |
| RAND — AI Evaluation and Testing | RAND | $10M | Coefficient Giving | 2025-09 |
| University of Oxford — Malaria Vaccine Manufacturing | University of Oxford | $4.7M | Coefficient Giving | 2024-12 |
| UC Berkeley — Cyberoffense Benchmark | University of California, Berkeley | $3.4M | Coefficient Giving | 2025-06 |
| General Support of Alignment Research Center (Evals Team) | ARC Evaluations | $3.2M | Survival and Flourishing Fund | 2023-01 |
| Center for Global Development — General Support (2019) | Center for Global Development | $3M | Coefficient Giving | 2019-06 |
| FutureHouse — Benchmarks for Biology Research and Development | FutureHouse | $2.9M | Coefficient Giving | 2024-03 |
| Stanford University — LLM Cybersecurity Benchmark | Stanford University | $2.9M | Coefficient Giving | 2024-07 |
| Harvard University — Antimalarial Bednet Development and Evaluation | Harvard University | $2.9M | Coefficient Giving | 2024-05 |
| African Union Development Agency — General Support (2020) | African Union Development Agency | $2.5M | Coefficient Giving | 2020-04 |
| New Partnership for Africa’s Development Planning and Coordinating Agency — General Support | New Partnership for Africa’s Development | $2.4M | Coefficient Giving | 2017-04 |
| IDinsight — Embedded GiveWell Team (2017) | IDinsight | $2.3M | Coefficient Giving | 2017-05 |
| Apollo Research — General Support | Apollo Research | $2.2M | Coefficient Giving | 2024-05 |
| IDinsight — Endline Evaluation of New Incentives RCT | IDinsight | $2.1M | Coefficient Giving | 2019-08 |
| New York University — LLM Cybersecurity Benchmark | New York University | $2.1M | Coefficient Giving | 2024-07 |
| Malaria Consortium — Monitoring and Evaluation of LLIN Distribution Campaign | Malaria Consortium | $2.1M | Coefficient Giving | 2021-07 |
| Center for Study of Science, Technology and Policy — Centre for Air Pollution Studies Initiative | CSTEP | $2M | Coefficient Giving | 2023-08 |
| IDinsight — General Support | IDinsight | $2M | Coefficient Giving | 2016-06 |
| University of Maryland — LLM Cybersecurity Benchmark | University of Maryland | $1.7M | Coefficient Giving | 2024-09 |
| Center for Open Science — LLM Research Benchmark | Center for Open Science | $1.7M | Coefficient Giving | 2024-07 |
| SeedAI — General Support | SeedAI | $1.6M | Coefficient Giving | 2025-03 |
| UC Davis — Malaria Gene Drive Feasibility Analysis (Greg Lanzaro) | University of California, Davis | $1.5M | Coefficient Giving | 2020-02 |
| UC Berkeley — Research on Rapid COVID-19 Serology Testing (Lisa Barcellos and Eva Harris) | University of California, Berkeley | $1.3M | Coefficient Giving | 2020-04 |
| Texas Organizing Project — Criminal Justice Reform (2017) | Texas Organizing Project | $1.2M | Coefficient Giving | 2017-03 |
| Owain Evans Research Group — AI Evaluations Research | Effective Ventures Foundation USA | $1.2M | Coefficient Giving | 2023-05 |
| Malaria Consortium — Monitoring and Evaluation of Net Distribution in Anambra, Nigeria | Malaria Consortium | $1.1M | Coefficient Giving | 2021-12 |
| Algorithmic Research Group — Language Model Capabilities Benchmarking (2024) | Algorithmic Research Group | $1.1M | Coefficient Giving | 2024-08 |
| Princeton University — Software Engineering LLM Benchmark | Princeton University | $1M | Coefficient Giving | 2024-05 |
| gui2de — Zusha! Road Safety Campaign (February 2017) | Georgetown University | $900K | Coefficient Giving | 2017-02 |
| Stanford University — LLM-Generated Research Ideation Benchmark | Stanford University | $880K | Coefficient Giving | 2024-05 |
| Princeton University — AI R&D Benchmark | Princeton University | $863K | Coefficient Giving | 2024-09 |
| Friedrich Schiller University Jena — Analytical Chemistry Benchmark | Friedrich Schiller University Jena | $829K | Coefficient Giving | 2024-10 |
| Evidence Action — Impact Evaluation of Iron and Folic Acid Supplementation (“Phase 2”) | Evidence Action | $800K | Coefficient Giving | 2019-03 |
| University of Illinois Foundation — LLM Hacking Benchmarks | University of Illinois Urbana-Champaign | $800K | Coefficient Giving | 2024-01 |
| Trustees of Boston University — LLM Research Benchmark | Boston University | $756K | Coefficient Giving | 2024-07 |
| University of California, Berkeley — Software Engineering Benchmark | University of California, Berkeley | $740K | Coefficient Giving | 2024-08 |
| Daniel Kang — Research on AI Benchmarks | Daniel Kang | $680K | Coefficient Giving | 2024-12 |
| Abdul Latif Jameel Poverty Action Lab — Innovation and Science Research | Abdul Latif Jameel Poverty Action Lab | $649K | Coefficient Giving | 2022-08 |
| University of Oxford — LLM Research Replication | University of Oxford | $622K | Coefficient Giving | 2024-09 |
| University of Illinois Urbana-Champaign — Zero-knowledge Proofs for Secure AI Audits | University of Illinois Urbana-Champaign | $615K | Coefficient Giving | 2025-02 |
| FutureSearch – Benchmark for Language Model Forecasting | FutureSearch | $607K | Coefficient Giving | 2024-03 |
| Yale University — LLM Persuasiveness Evaluation | Yale University | $596K | Coefficient Giving | 2024-06 |
| Grant to Model Evaluation & Threat Research (METR) | METR | $548K | Survival and Flourishing Fund | 2025 |
| Carnegie Mellon University — Benchmark for Web-Based Tasks | Carnegie Mellon University | $547K | Coefficient Giving | 2024-03 |
| WestExec — Report on Assurance in Machine Learning Systems | WestExec | $540K | Coefficient Giving | 2020-02 |
| Precision Development – Trial Scoping | Precision Development | $540K | Coefficient Giving | 2022-04 |
| Metaculus — Forecasting Tournaments | Metaculus | $532K | Coefficient Giving | 2024-05 |
| iGEM — Synthetic Biology Safety and Security (2016) | International Genetically Engineered Machine Foundation | $520K | Coefficient Giving | 2016-05 |
| Cambridge in America — Data Science Benchmark | Cambridge in America | $518K | Coefficient Giving | 2024-07 |
Funding by Funder
| Funder | Grants | Total Amount |
|---|---|---|
| Coefficient Giving | 85 | $105M |
| Survival and Flourishing Fund | 3 | $4M |
| Manifund | 11 | $1.1M |
| Long-Term Future Fund (LTFF) | 15 | $608K |
| Centre for Effective Altruism | 4 | $131K |
| Centre for Effective Altruism | 4 | $131K |
Sub-Areas12
| Name | Status | Orgs | Papers |
|---|---|---|---|
| Alignment EvaluationsEvaluations specifically designed to measure alignment properties: honesty, helpfulness, harmlessness, and value adherence. | active | 4 | 0 |
| Alignment FakingResearch on whether and how AI systems might pretend to be aligned during evaluation while pursuing different goals at deployment. | emerging | 4 | 1 |
| Backdoor DetectionDetecting adversarially implanted vulnerabilities in model weights. | active | 4 | 0 |
| Capability ElicitationMethods for discovering hidden or latent capabilities in AI systems. | active | 4 | 0 |
| Control EvaluationsStress-testing systems designed to constrain AI behavior; monitoring for collusion. | emerging | 4 | 0 |
| Dangerous Capability EvaluationsTesting AI systems specifically for dangerous capabilities like CBRN knowledge, cyber offense, autonomous replication, and persuasion. | active | 4 | 0 |
| Evaluation AwarenessStudying how AI systems might game evaluations by detecting when they are being tested. | active | 4 | 0 |
| Jailbreak ResearchResearch on prompt injection, jailbreaking attacks, and defenses for language model safety filters. | active | 4 | 0 |
| Red TeamingAdversarial testing of AI systems to discover failure modes, safety issues, and vulnerabilities, both manual and automated. | active | 7 | 0 |
| Reward Hacking of Human OversightEmpirically investigating how AI systems deceive or manipulate human evaluators. | emerging | 4 | 0 |
| Scheming DetectionResearch on detecting when AI systems are engaged in deceptive alignment or strategic manipulation of their training process. | emerging | 4 | 0 |
| Sleeper Agent DetectionResearch on detecting and mitigating backdoors, trojans, and time-delayed deceptive behavior in AI systems. | emerging | 4 | 1 |